NUNC COGNOSCO EX PARTE

THOMAS J. BATA LIBRARY

TRENT UNIVERSITY
 Digitized by the Internet Archive
in 2019 with funding from
Kahle/Austin Foundation

https://archive.org/details/sexpersonalityOOOOeyse
SEX AND PERSONALITY
Sex and Personality

H. J. Eysenck Ph.D., D.Sc.

University of Texas Press

Austin
Uvz- Vas

International Standard Book Number 0-292-77529-6

Library of Congress Catalog Card Number 76-15479

© H. J. Eysenck 1976

All rights reserved

Printed in England
 Contents

Introduction 1
1 Individual differences in sexual behaviour 5

2 Sexual attitudes and behaviour in student populations 30

3 Sexual attitudes and behaviour in adult populations 78

4 Social attitudes and sexual behaviour 150

5 Sexual attitudes and behaviour in psychiatric patients 177

6 Genetic factors in sexual behaviour 192

7 Sexual attitudes and social consequences 220

References and Author index 243

General index 253

The greatness of God is infinite, for while with one die man
impresses many coins and they are all alike, the King of Kings,
the Holy One, with one die he impresses the same image on all
men, yet not one of them is like his companions.
Talmud
 Introduction

Psychology holds a central place in our efforts to make sense of

the world, and of our place in it; unless we can understand our¬
selves, and have some firm knowledge of our ways of perceiving,
acting and behaving, we are likely to misinterpret the evidence of
our senses, and derive entirely false conclusions about our state.
It would however be quite wrong to imagine that efforts to place
psychology on a scientific footing are likely to receive any whole¬
hearted welcome from most people; quite the opposite is true. Just
as humanity resented the efforts of Copernicus and Galileo to dis¬
place the earth from its place in the centre of the universe, or the
work of Darwin which displaced man from his unique position in
the animal kingdom, so nowadays most people resent the psycho¬
logist’s efforts to replace unreasoned belief by factual demonstration.
We are determined to believe certain things about ourselves, and
under extreme provocation resort to the famous words of the war¬
time cartoon: ‘Don’t confuse me with facts; my mind is made up!’
Even suggest that homo sapiens may not be as sapiens as all that,
and all hell breaks loose; we may under duress agree that people
who do not share our own political, social, sexual and other views
may indeed be misguided, irrational and neurotic; but we would
most certainly defend to the death the proposition that our own
views are based entirely on sound reasoning, factual research and
objective appraisal.
In the sexual field, in particular, there has for centuries (and
probably since the beginning of recognisable societies) existed
a battle between puritans and libertines, between restrictive tradi¬
tion and uninhibited permissiveness. Both sides defend their position
vociferously, often appealing to such scientific research as may seem
to support their point of view and heaping scorn on anything that
might seem to go counter to their beliefs; neither side seems willing
to realise that perhaps their quarrel may be based on a misunder-

1
 SEX AND PERSONALITY

standing. When I say that I like strawberries and dislike oysters

and you say that you like oysters and dislike strawberries, we are
not really taking issue with each other; we are both stating facts
that have a personal reference, but no universal validity. The error
comes in only when we generalise our own preferences into pre¬
scriptive statements, saying that everyone should like strawberries
more than oysters (or vice versa). This is too silly for anyone to do
in this particular case, but on other issues that is precisely what
people do do. Libertines suggest that their way of expressing their
sexuality is the only proper way, and that those who disagree with
them are neurotic, inhibited and sick. Puritans in their turn believe
that their way of expressing their sexuality is the only proper way,
and that disagreement indicates possession by the devil, lack of
morality and probably mental disorder as well. The implication
in such debates always is that there is one and only one proper way
of conducting oneself, and that this of course is the way one happens
to have chosen oneself. Such beliefs, and such debates, generate more
heat than light, and they are likely to lead to intolerance, suppression
and persecution of those not sharing one’s own views.
This book grew out of the author’s feeling that perhaps there
was no single, orthodox way to salvation; that perhaps different
people are born with different desires, values and attitudes; and
that perhaps the desire for a universal Gleichschaltung was a piece
of authoritarian impudence which had no scientific warrant what¬
ever. Human diversity has always fascinated me, and the belief,
inherent in Darwinian evolutionary theories, that such diversity is
essential for survival (by allowing pressure of changing conditions
to favour now one, now another, form of adjustment, depending
on circumstances) seems to me an essential biological under¬
pinning of democratic government. If you find value in diversity
you are perhaps less likely to wish to tie everyone on to a Pro¬
crustean bed, and if you realise the tremendous pressure exerted
by genetic factors in pushing us one way or another in our efforts
at adjustment, you are likely to be more modest in your claims for
your particular plans of universal salvation, and more desirous
of seeing diversity of conduct preserved. It seemed worthwhile,
as a first step in the direction of finding out more about the myriad
ways in which human beings in our society express their sexual
desires, to carry out some empirical investigations, guided by the
wish to find out what people were actually thinking and doing.

2
 INTRODUCTION

rather than by a desire to tell them what they ought to be thinking

and doing.
Scientific studies are seldom of a purely inductive kind, simply
gathering facts along Baconian lines; usually there is some more
or less explicit theory guiding the collection of data. In this case
the theory was quite explicit; it seemed likely to me that people
characterised by different personalities would also differ in predict¬
able ways in their sexual conduct, and that such predictions could
be tested and would in turn cast some interesting light on the
theories of personality that inspired the research. Consequently
much of the work reported in this book looks at the interaction
between personality on the one hand and sexual attitudes and
behaviours on the other. The subjects are very varied, comprising
normal adult men and women, students and adolescents, criminals,
inmates of Broadmoor Hospital for the criminally insane and so
forth. On the whole, results were mostly along the lines anticipated,
and this was true regardless of the particular samples studied.
An area of particular interest was that of the differences between
males and females. Are the sexual attitudes of men and women in
our day and age really as different as tradition has it? Are the
relationships between attitudes and personality similar for the two
sexes, or are they quite different? These and other similar questions
seemed of some interest, and consequently particular attention was
paid to the detailed analysis of sex comparisons.
A final item of interest was the question of genetic determination
of sexual attitudes and behaviours. It has become fashionable to
deny the importance of biological factors in human behaviour,
and to attribute all differences that are observed to role-playing
generated by environmental pressures. There is too much evidence
to the contrary to accept such whole-hogging sociological views
in their entirety; there does seem to exist the alternative possibility
that some of these roles are enforced by society on men and women,
say, because throughout history genetic and other biological factors
have exerted pressure in that direction. Without wishing to preempt
the discussion, it seemed reasonable to carry out some direct studies
of these questions, using our Twin Register; the results would
seem to support a view insisting on the complex interaction of
genetic and environmental factors in the determination of human
conduct, particularly in the sexual sphere.
What do the results of our inquiries have to teach us regarding

3
 SEX AND PERSONALITY

the general ordering of our affairs in the sexual field? I have

tried to indicate the general thrust of the argument in the final
chapter, but of course the reader is free to interpret the import of
the results in any way he or she chooses. Scientific facts are relevant
to ethical, social and legal decisions, but they do not determine
them; the facts here brought forward are relevant to such issues
as censorship, prostitution, marriage, perverse practices and so forth,
but decisions regarding these affairs derive from a wider religious,
ideological or political context, and this context will determine to
some extent the regard paid to the factual material. My own ex¬
pertise ends with the presentation of the empirical studies; all I
have tried to do is to suggest some practical consequences that
seem to me to follow from these. The reader is at liberty to dis¬
regard my comments completely; they do not have the same status
as experimentally ascertained facts.
I am indebted to many people who helped in the collection
of data, too numerous to mention. I also encountered much hostility
when approaching vice-chancellors and other eminent academic
persons in order to obtain their permission to invite students under
their care to take part in the investigation; one of them offered
to pray for me in order to make me desist from what he was con¬
vinced was not only an impious and odious effort, but was also
likely to debauch and corrupt the students involved. Students them¬
selves, as well as the older and more mature men and women
who constituted my samples, took a rather more realistic and
sensible view of the proceedings; they were interested, helpful
and entirely serious in their endeavours to give a realistic picture of
their attitudes and practices. Undoubtedly they were not a random
sample of the population, but they represented all the diverse ways
of looking at sex, indulging or not indulging in sex, and of thinking
about sex that one is likely to find in this country at this time.
For their wholehearted collaboration in spending countless hours
on lengthy and perhaps rather boring questionnaires I wish here
to express my heartiest thanks; without their help this book
assuredly could not have been written!
H. J. Eysenck Ph.D., D.Sc.
Maudsley Hospital
15 January 1976

4
 1

Individual differences in sexual

behaviour

The literature on human sexual behaviour is large; even larger

is the literature on animal sex. Although there was much interest in
the topic prior to Kinsey, it is with his large-scale work that the
proper investigation of human sexual behaviour is usually con¬
sidered to start.(l) Nearly all of this literature has followed Kinsey
in being concerned with population parameters; interest centred on
such questions as the average number of times men and women
had intercourse per week; the percentage of men who had had
homosexual experiences, or of women who had Lesbian experiences;
the average number of premarital or extra-marital experiences of
men and women; or the average age at which men and women
had their first intercourse. These are interesting questions, but
they are not of very great scientific importance. Nor is it likely
that they can be answered with any reasonable degree of accuracy.
Kinsey and his followers have proceded in a somewhat Baconian
fashion, gathering facts and hoping that later on theories would
be developed on the basis of these collections of facts. Unfor¬
tunately they disregarded T. H. Huxley’s injunction; ‘Those who
refuse to go beyond fact seldom get as far as fact.’
My belief that these ‘facts’ are not really very factual is based
on the following considerations. In the first place, population para¬
meters cannot be estimated properly except on the basis of reason¬
ably random samples of the population. Now Kinsey was quite
well aware of the fact that his samples were by no means random;
he tried to compensate for the deficiencies in his samples, but un¬
fortunately this is not possible; his estimates may be near the true
values, but it is impossible to know whether this is so, or whether

5
 SEX AND PERSONALITY

they depart to a considerable extent from these true values. It

would be wrong to blame him for not obtaining random, or even
quota or stratified, samples; in the nature of things, this would have
been impossible when such a very personal thing as ‘sexual ex¬
perience’ was being investigated. There is an inevitably high refusal
rate for studies of this kind, and no proper replacement can
be made for these ‘refusees’.
In the second place, Kinsey’s interviewers asked for information
that is almost impossible to recover, and that in the form asked for
probably never existed in the mind of the interviewee. Consider
questions regarding frequency of intercourse. Very few people have
kept the sort of accurate record of their sexual behaviour that alone
could produce a reasonably accurate answer. Guessing how many
times one has had intercourse during the past month, or year,
or decade is likely to give very erroneous answers; frequencies for
many people are so variable (particularly for non-married people),
and so subject to illnesses, ‘periods’ and other disturbing influences,
that guesses cannot in the nature of things be at all accurate.
Relations between answers given spontaneously to questions of this
kind, and records kept afterwards to check on their accuracy, in our
experience have not shown a very high degree of agreement.
This is true when the question is purely factual; even greater com¬
plications ensue when the answers are value-laden. This is our
third point. There is much ‘social desirability’ in some answers, as
compared to others; psychologists have done much work in recent
years to demonstrate the importance of such social desirability in
shaping responses in questionnaires and interviews. Admittedly
Kinsey and his successors attempted to get over this difficulty, but
there is no evidence that they were successful in this, or just how
successful they were. These three objections are the main ones
that have been brought forward; they argue strongly against taking
the ‘facts’ unearthed too seriously.
But even if all the figures given in the literature could be taken
as accurate and reliable, it could still be argued that they have
little scientific value. Facts in science are important and relevant
to the extent that they support or refute a theory; as Darwin pointed
out, ‘how odd it is that anyone should not see that all observation
must be for or against some view if it is to be of any service! ’
These facts (even assuming them to be such) are quite atheoretical;
they neither support nor refute any meaningful scientific theory.

6
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

and must consequently be judged to be of little systematic im¬

portance. This position could of course be altered if the Kinsey
studies were to be repeated every ten or twenty years, so that
temporal trends could be established (assuming all the difficulties
listed above to have been overcome); these trends might be used as
evidence for certain psychological or sociological hypotheses re¬
garding changes in sexual mores. But by themselves they are not
of great scientific interest or importance.
There is one further objection to the Kinsey-type investigation.
Means and averages are meaningful only if there are no great indi¬
vidual differences among the units so averaged. To know that
the average number of times people have intercourse is twice a
week is meaningful if the numbers averaged vary from 2 to 4, or
even from 1 to 5. But in actual fact, when the details are plotted it
becomes apparent that many people have intercourse only once a
month, or even less frequently, while others have it several times
a day. The variability is so extreme that averages tell us little. In
line with his intention to provide us with population parameter
estimates, Kinsey gives means and variances for most of his tables
(or at least these can be estimated from his figures), but all they do
is illustrate the extreme degree of variability that is so characteristic
of human sexual responses.
Figure 1.1 shows the kind of variability referred to; it gives
the distribution of people in the 16-24, 25-34 and 35-54 and
55+ age groups on a scale showing the number of times inter¬
course took place on the average. The figures are taken from a
national sampling survey, and averaged for males and females;
within the limitations of accuracy of estimates, limitations of
sampling refusals and the other objections noted above, the results
should be fairly representative of British men and women in the
1970s. The decline of frequency of intercourse with age is noticeable
but hardly surprising; what is interesting in this figure is the great
spread, for each age group, of the frequencies. Even in the 55 +
group, some have intercourse four to six times a week, others
never! This variability makes generalisations pretty meaningless.
Kinsey makes little effort to account for individual differences;
he only tries to do so indirectly by reference to sociological
variables (like social class) and biological variables (like age). Thus
he presents means and variances for different age groups, or different
social classes. There is little in the way of psychological theorising.

7
Figure 1.1 Frequency of intercourse for men and women of four different
age groups

and practically no reference to personality variables, in spite of the

fact that such variables are likely to play an important part within
age and social class groups. This neglect of individual differences,
and this failure to attempt to account for them along psychological
lines, is one of the outstanding failures of recent work on human
sexual responses.
It might be thought that this stricture does not apply to psycho¬
analysis; would it not be true to say that psychoanalysts too have
criticised Kinsey on similar lines, and have proffered extensive and
intricate theories to account for individual differences in this field? I
will not here repeat the criticisms I have elsewhere made of psycho¬
analytic theories, or the failure of these theories to stand up to
experimental testing (Eysenck and Wilson 1974); let it suffice to say
that ‘explanations’ given by psychoanalysts of human sexual be¬
haviour, and individual differences, have no general validity or
application. They are usually ad hoc applications of arbitrary
principles, from which no general deductions can be made, and
which therefore cannot be tested scientifically. Psychoanalysis'does
not enable us to answer the urgent question raised by the existence
of very marked individual differences in human sexual behaviour.
It is the purpose of this book to make a beginning in the business

8
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

of taking a serious look at this vexed question. We will do so, not

along Baconian lines, i.e. by arbitrarily relating bits of personality
to bits of sexual behaviour, but along the lines of the hypothetico-
deductive method, i.e. by trying to make certain predictions from
general personality theory (Eysenck 1967) on to selected aspects of
sexual behaviour. It is the purpose of this first chapter to outline
the major parts of this personality theory, and to suggest how the
theory can be used to make predictions in the field of sexual
behaviour. It is not expected that these predictions will be verified
in their entirety, or even that where they are verified the predicted
effects will be very strong. Obviously sexual behaviour is so
complex, and determined by so many factors, both systematic and
accidental, that no single factor could account for a very large
portion of the variance. Nevertheless, it seemed worthwhile to see
to what extent predictions from personality theory could be verified,
and to obtain a rough idea of the pattern of interrelations that exists
in our society. In addition, we were concerned with the possibility
of trying to account for sexual behaviour along genetic lines, deter¬
mining the amount of variance contributed to different types of
sexual behaviour by genetic and environmental factors, or by their
interaction. This line of thought could obviously be extended to
include studies of the degree to which relations between personality
and sexual behaviour themselves were determined by genetic and
environmental influences. Results of such studies, it was considered,
would not answer in any profound sense the questions we were
asking; they would form a beginning on which other investigators
might build.
One further thought was in our minds. The recurring battles over
pornography and censorship, over chastity and premarital inter¬
course, over marriage and ‘living in sin’, have usually been waged
on the basis that one answer is ‘right’ and another ‘wrong’, in some
all-embracing, generally valid sense. It seemed worthwhile to investi¬
gate people’s actual attitudes to these and many other problems in
the sexual field, to demonstrate the diversity of answers given, and
the complex structure revealed by the analysis of the answers. If the
acceptance or rejection of different ‘life styles’ is regulated to some
extent by personality variables, themselves largely determined by
genetic factors, then it might have to be concluded that the very
notion of ‘right’ and ‘wrong’ might have to be given up in relation
to these questions, and that what was right for a neurotic introvert

9
 SEX AND PERSONALITY

might be wrong for a stable extravert. These ideas will be referred

to again in the final chapter; here we only intend to mention the
wider context in which our investigation is set.
When we turn to the personality side, there is of course a bewilder¬
ing array of different theories, techniques of measurement and con¬
ceptualisations; even the meaning of the term ‘personality’ is not
agreed upon, and the very notion of personality as a persevering and
consistent set of behaviour patterns is sometimes doubted (e.g.
Mischel 1968). This is not the place to discuss such issues; the
theory adopted here to make predictions and to relate personality
to sexual behaviour is that which I have presented in some detail
in The Biological Basis of Personality (Eysenck 1967), and the
genesis of which I have described, together with supporting evidence
for its meaningfulness, in The Structure of Human Personality
(Eysenck 1970a). It is open to the reader to argue that other theories
might have been adapted for the purpose, and I would be the last
to deny that such a possibility cannot be ruled out. Nevertheless, the
deed is done, and argument about wisdom of the choice fruitless;
the only thing to be determined is the degree of success that this
effort has produced. Accordingly a brief outline of the theory will
now be presented, together with some of the deductions made from
it. It would be pointless to document here what is said about the
theory; readers must be referred to the works cited above for
detailed information and documentation.
Our personality theory maintains that personality can be des¬
cribed best in terms of a large set of traits (sociability, impulsive¬
ness, activity, moodiness, etc.), and that these tr&its cohere in certain
clusters; these clusters are the empirical basis for higher-order con¬
cepts which may be labelled ‘types’ or, as I would prefer to call
them, dimensions of personality. (The term ‘type’ carries connota¬
tions of bimodal distribution or qualitatively different groupings
which go counter to empirical fact.) Three main such clusters have
been isolated in the very large number of empirical studies carried
out during the past seventy years; these have received many differ¬
ent names, but will here be referred to as N (neuroticism), E (ex¬
traversion), and P (psychoticism). The methods used for isolating
such dimensions, and for identifying the resulting factors, have
been discussed in detail in Personality Structure and Measurement
(Eysenck and Eysenck 1969). These major ‘types’ or dimensions
are conceived as being securely anchored in innate physiological

10
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

structures, differences in which are responsible,- in interaction with

environmental determinants, for producing the phenotypic behaviour
patterns that we observe, rate and measure; a discussion of the
evidence for this genetic hypothesis will be given in a later chapter.
Extraversion is defined in behavioural terms by the various traits
that are correlated together to define this factor, traits such as
sociability, impulsiveness, activity, carefreeness, liveliness, jocularity
and so forth. The physiological basis underlying these and many
other traits and behaviour patterns is thought to be low cortical
arousal, probably mediated by the reticular formation. The extra-
vert consequently requires greater external stimulation in order to
arrive at an optimum level of arousal than the normal (ambivert)
person, while the introvert requires less external stimulation than the
normal (ambivert) person. This general feature of human behaviour
(preferring intermediate degrees of arousal) is shown in diagram¬
matic form in figure 1.2; the thick curved fine indicates the general
relation obtaining between level of stimulation and hedonic tone,
with intermediate levels of stimulation giving rise to positive hedonic
tone, extreme levels of stimulation (pain) or sensory deprivation
giving rise to negative hedonic tone. Introverts and extraverts
are displaced from this mean in opposite directions, with extraverts
having optimum levels of stimulation above (O.L.E), introverts
having optimum levels of stimulation below (O.L.j) the optimum
level of the average person (O.L.p). It follows that at a given
point A which is indifferent to the average person, the level of
stimulation would be acceptable to the introvert and mildly pleasant,
but unacceptable to the extravert and mildly unpleasant, as lacking
in sufficient stimulation. Conversely, at point B, where also the
level of stimulation would be giving rise to indifference to the
average person, the extravert would be experiencing mildly positive
hedonic tone, the introvert mildly negative hedonic tone. The
evidence for these generalisations is discussed in Eysenck (1967),
and is quite strong.
The prediction from this general formulation would be, in the
sexual field, that extraverts would seek for stronger stimulation
than ambiverts, ambiverts than introverts. Sexual stimulation being
among the strongest sensory experiences commonly encountered,
one would conclude from this that introverts would approach it
rather more circumspectly, and would shy away from the more
stimulating forms of sexual behaviour. Conversely, extraverts would

11
 POPULATION
INTROVERT AVERAGE EXTRAVERT

LEVEL OF STIMULATION
Medium
LOW HIGH
[Sensory Deprivation] [Pain]

Figure 1.2 Relation between sensory stimulation and positive or negative

hedonic tone for extroverts, ambiverts and introverts respectively

approach precisely these more stimulating forms of sexual behaviour

eagerly and passionately. This prediction is supported by another
deduction from the general theory. I have theorised that introverts,
because of their higher level of cortical arousal, should form condi¬
tioned responses more quickly than extraverts; I have also theorised
that socialised behaviour is produced through a process of Pav-
lovian conditioning, and that consequently extraverts, being poor
conditioners, would be likely to show less evidence of socialisation
than introverts. The evidence on the whole supports both deductions
(Eysenck 1967, Eysenck 1970b). Now much sexual behaviour (per¬
versions, premarital and extra-marital intercourse) is considered to
be anti-social, and is consequently more likely to be indulged in
by extraverts than by introverts. These two causal chains, through
sensation-seeking and through lack of socialisation, should lead
to extravert in the same direction, namely greater indulgence in all
sorts of sexual behaviour involving persons of the opposite sex;
and accordingly the following predictions have been made (Eysenck
1971a):

1 Extraverts will have intercourse earlier than introverts

2 Extraverts will have intercourse more frequently than intro¬
verts

12
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

3 Extraverts will have intercourse with more different partners

4 Extraverts will have intercourse in more different positions
than introverts
5 Extraverts will indulge in more varied sexual behaviour out¬
side intercourse *
6 Extraverts will indulge in longer pre-coital love play than
introverts

There is one further deduction from general theory that is relevant

to these predictions. Extraverts, given their lower level of arousal,
would be expected to show greater habituation in experimental
situations; i.e., they would react as strongly as introverts to novel
stimuli, but would quickly get used to these stimuli and respond less
and less, while introverts would keep their original reaction much
longer. The evidence supports such a view (Eysenck 1967, Quirion
1970), and we would suggest that the changeableness of the extravert
in his sexual behaviour posited above (changes in partner, changes
in positions adopted, etc.) might be related to habituation. Clearly
habituation to a given stimulus leads to loss of arousal, boredom and
disinterest; this in turn would be expected to motivate search for
new stimuli. Thus extraverts are genetically predisposed to seek
strong stimuli, to get bored by repetition of identical or similar
stimuli and to be less likely to be inhibited from searching for such
strong and novel stimuli by social taboos. It is this combination
of behavioural tendencies on which our predictions are based.
There is some objective evidence that habituation is indeed more
pronounced in extraverts in the sexual field. In an unpublished
study, E. Nelson measured the physiological responses of male sub¬
jects to explicit sexual films, using the penis plethysmograph. (This
is a device for measuring the changing volume of the penis from
flaccid to erect.) Nine four-minute films were constructed from a
large amount of commercial material, each film concentrating on
one aspect of sexual activity, e.g. petting, intercourse male above
female, cunnilingus, fellatio, soixante-neuf, orgies, intercourse a
tergo, etc. Subjects were carefully chosen on the basis of personality
inventories (E.P.I.) to represent the four groups of stable extraverts,
stable introverts, unstable extraverts and unstable introverts; there
were ten subjects in each group. Three films were shown to each
person on three separate occasions, at least four days elapsing from
occasion to occasion; there was a rest of four minutes between
films on each occasion. Records were analysed for changes occurring

13
 SEX AND PERSONALITY

during each film, from film to film on any one occasion, and from
one occasion to another.
This analysis showed that extraverts showed significantly more
habituation (i.e. a lowering of penis volume) during the showing
of each film, from film to film, and from occasion to occasion. This
tendency was clearer for the stable extraverts and introverts, but
it was also manifest for the unstable extraverts and introverts. Intro¬
verts showed no change in responsiveness from beginning to end,
while extraverts showed a drop of 86 per cent of maximum
erection to 66 per cent of maximum erection. Thus clearly habitu¬
ation occurs more readily in extraverts, not only to simple sensory
stimulation (as had been used in previous laboratory investigations)
but also to visual sexual stimulation.
Prior to the experiments here published, there was only one
study that attempted to look at these and similar predictions in
an empirical manner (apart from the work of Schofield 1968 and
Bynner 1969, which, although not making use of personality
questionnaires, arrived at similar conclusions about the relationship
between extraversion and explicit sexual conduct). The study in
question was conducted by Giese and Schmidt (1968), who used
a rather short scale for the measurement of E and administered
questionnaires regarding their sexual conduct to over six thousand
German students, both male and female; most of these were un¬
married. Some of their relevant results are shown below (table 1.1a);
only unmarried students are included in the tabulation, and groups
are subdivided according to their extraversion scores into introverts
(EJ, ambiverts (Ea) and extraverts (E3). There were more men than
women in this sample. It will be seen that extraverts masturbate
less, pet to orgasm more, have coitus more frequently, have coitus
earlier, adopt more different positions in coitus, indulge in longer
precoital love-play, and practise fellatio and cunnilingus more
frequently. It should be added that on some of these items differ¬
ences are much greater for men than for women; this is expected on
the grounds that in our society it is men who set the pace in sexual
relationships, so that their personality is expressed more clearly
in the procedure adopted. These results are in good agreement with
our predictions.
Two studies reported by Zuckerman and others (1972) and
Zuckerman (personal communication) did not use the E scale, but
are relevant because the sensation-seeking scales used in these

14
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

Table 1.1a Sexual activities of introverts (E,), ambiverts (E„) and

extroverts (Eg) (figures taken from Giese and Schmidt 1968)

Males Females
(percentages shown in italics)

Ei e2 e3 E> E„ e3
1 Masturbation at present 86 80 72 47 43 39
2 Petting: at 17 16 28 40 15 19 24
Petting: at 19 31 48 56 30 44 47
Petting: at present age 57 72 78 62 71 76
3 Coitus: at 17 5 13 21 4 4 8
Coitus: at 19 15 31 45 12 20 29
Coitus: at present age 47 70 77 42 57 71
4 Median frequency of coitus per month 30 3-7 5-5 3-1 4-5 7-5
(sexually active students only)
5 Number of coitus partners in last 1 75 64 46 72 77 60
12 months; unmarried students 2 — 3 18 25 30 25 17 23
only 4+ 7 12 25 4 6 17
6 Long precoital sex play 21 25 28 21 16 18
7 Cunnilingus 52 62 64 58 69 69
8 Fellatio 53 60 69 53 59 61
9 More than three different coital positions 10 16 26 12 18 13
10 Experience of orgasm nearly always — — — 17 32 29

studies correlate positively with E (and also with P), and thus
make prediction possible (Zuckerman 1974). There are four such
scales, concerned respectively with thrill and adventure-seeking,
experience-seeking, disinhibition and boredom-susceptibility; these
are summed to give a total score. Zuckerman constructed a sexual
experience scale, listing types of sexual behaviour that the individual
questioned either had or had not experienced (similar to our own
scales given in chapters 2 and 3), going from kissing and breast¬
fondling to fellatio and cunnilingus. He also asked questions about
the number of sexual partners each individual had had. The pre¬
diction would of course be that all the scales would correlate posi¬
tively with the number of heterosexual activities endorsed, and the
number of heterosexual partners. The scale involving heterosexual
activities was administered to two separate populations, that regard¬
ing heterosexual partners only to one. The results are shown in
table 1.1b; it will be seen that all the correlations are positive, both
for males and for females. These results fit in well with those of
Giese and Schmidt, and bear out prediction.
Neuroticism is conceived of as strong, labile emotionality, pre¬
disposing a person to develop neurotic symptoms in case of exces-

15
 SEX AND PERSONALITY

Table 1.1b Correlations between sensation-seeking scales and sexual

behaviour
Heterosexual Heterosexual No. heterosexual
activities 1 activities 2 partners

M F M F M F

General 0-51 015 0-39 029 0-40 0-27

Thrill and adventure- 0-44 016 0-42 0-35 0-47 0-20
seeking
Experience-seeking 0-37 0-32 0-45 0-37 0-35 0-28
Disinhibition 0-33 0-43 0-39 0-33 0-42 0-29
Boredom susceptibility 0-36 0-29 0-23 0-20 0-25 0-20

sive stress. Traits correlating to define this ‘type’ are moodiness,

sleeplessness, nervousness, inferiority feelings, irritability, etc. The
physiological basis of this dimension of personality is believed to
be the autonomic system, and in particular the visceral brain, which
coordinates sympathetic and parasympathetic reactivity. High N
scorers, in view of their strong, labile and lasting emotions of
fear and anxiety to even mildly stressful situations (in which category
we must include social contact generally, and perhaps sexual con¬
tact particularly), would perhaps be less likely to indulge in sexual
contacts, to worry more about sex, to be disgusted by certain aspects
of sex and to have fewer contacts with sexual partners; this would
be particularly true of unmarried subjects, because of their well-
known difficulties in social relations. In Miller’s terms, sexual rela¬
tions, to the neurotic, should present a conflict-laden area in which
the conflict takes the approach-avoidance form; sexual desire makes
the high N subject approach, while his fears and anxieties make him
avoid. This ambivalence is of course well documented in the clinical
literature, but our prediction would be that it could be found in
normal (i.e. non-psychiatric) populations also. This prediction of
conflict in the sexual field would be particularly acute in subjects
combining high N and high E\ high N would inflate the avoidance
part of the equation, high E the approach part. Our work suggests
that this is in fact so (Eysenck 1971c).
The only study of sexuality as related to neuroticism, prior to
our own, is the work of Giese and Schmidt, already mentioned.
Unfortunately they used a very short scale of N, which could not
have been very reliable; it is perhaps for this reason that they found
few significant correlation. High N scorers (male) were found to

16
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

masturbate more frequently, have greater desire for coitus, and claim
to have spontaneous erections more frequently than low N scorers.
Females with high N scores had less frequent orgasm and stronger
menstrual pains. As far as they go, these findings may lend some
slight support to the deductions from theory outlined above, but
clearly they do not help a great deal in understanding the relations
involved.
Some work has been done on the relationship between orgasm
capacity and psychiatric disorder, particularly neurotic disturbances,
but this is contradictory on the whole. Terman (1951) found that,
of a large sample of women, those lacking in orgasm capacity said,
in reply to inventory questions, that they often felt miserable, ex¬
perienced periods of loneliness, often felt grouchy, couldn’t be opti¬
mistic when others around them were depressed; these are all N
items, but so are others which did not discriminate. These items
had also shown discrimation in a previous study (Terman 1938);
others were significant only in the second, such as being burdened
by a sense of remorse, often being just miserable, worrying too long
over humiliating experiences, lacking in self-confidence and being
affected by other people’s praise and blame. There appears in these
two studies a relationship between incapacity and N, but not a very
strong one.
Thomason (1951) found similar differences between sexually well
adjusted and poorly adjusted women, but of course used other
criteria additional to orgasm capacity. Shope (1966) found a negative
relationship in a very small sample of unmarried women; this inver¬
sion may have been due in part to the unmarried state of the girls
in question. Fisher (1973), who gives a review of these studies, found
some significant correlations between orgasmic incapacity and the
Murray Anxiety Scale in one of his samples, but failed to do so in
another; he also used some rather unusual and probably invalid
techniques for the measurement of anxiety in other samples, without
success. However, these samples were all quite small. Terman’s con¬
clusions are probably better based than those of the other authors
mentioned, although even he used a rather poor measure of N.
Several authors have reported restricted orgasmic capacity in
neurotic psychiatric patients (Landis and others 1940, Winokur and
others 1958, Purtell and others 1951, and Coppen 1965). Cooper
(1969) failed to find significant elevations in the N scores of women
with orgasmic difficulties, but this of course is not a psychiatric

17
 SEX AND PERSONALITY

diagnosis. Fisher (1973) concludes a review of these and other

studies (mostly only marginally relevant) by saying that ‘the exist¬
ing data are not persuasive that general psychological maladjust¬
ment, as such, plays a major role in sexual responsiveness’. This
leaves the probability that high neuroticism plays some part in
orgasmic incapacity, even though perhaps not a major one.
Whereas E and N are dimensions of personality that have been
studied for many years (the first set of studies using this particular
model of personality was presented by Eysenck 1947), and for
which there is a good deal of theoretical and experimental back¬
ground, the third dimension of personality here considered, psycho-
ticism, has a much less opulent background. In conception it is not
unlike neuroticism; just as N purports to be a personality variable
stretching through the normal, non-psychiatric population, and
measurable through questionnaires that do not involve actual
psychiatric symptoms, so also psychoticism purports to be a person¬
ality variable stretching through the normal, non-psychiatric popula¬
tion, and the attempt has been made to measure it by means of
questions that do not involve psychiatric symptoms but rather per¬
sonality traits (Eysenck and Eysenck 1976). The assumptions under¬
lying these scales and dimensions may be stated as follows.

1 Neurosis and psychosis are entirely different and separate dis¬

orders
2 Neurosis and psychosis are not qualitatively different from
normal behaviour patterns, but are extremes along dimensions that
run through the normal population
3 Diagnosed neurotics and psychotics may be used as criteria
for measuring scales of N and P, in the sense that they ought
to have high scores on scales for the measurement of N and P
respectively

These points, and the evidence for them, are discussed elsewhere
(Eysenck 1970a, Eysenck and Eysenck 1969, Eysenck and Eysenck
1975, 1976). For a consideration of psychoticism, it is this last refer¬
ence that is of particular importance; it contains a survey of the
relevant literature.
It is of particular interest in relation to P that genetically there
appears to be a close link between psychosis* and psychopathy
(Eysenck 1975a); this would suggest that criminals as well as
psychotic patients would show high P scores. This has in fact been

18
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

found to be true (Eysenck and Eysenck 1970a, 1971a, 1971b) and

reinforced our belief that the P scale measures something important,
meaningful and genetically relevant.
It may be premature to attempt to describe the personality traits
characteristic of the high P scorer, but in the absence of full
knowledge on this point it may still be useful to quote the des¬
cription given in the Manual of the E.P.Q. (Eysenck Personality
Questionnaire) (Eysenck and Eysenck 1975) constructed for the
measurement of the P, E and N variables.

[A high scorer may be described] as being solitary, not caring

for people; he is often troublesome, not fitting in anywhere. He may
be cruel and inhumane, lacking in feeling and empathy, and alto¬
gether insensitive. He is hostile to others, even his own kith and
kin, and aggressive, even to loved ones. He has a liking for odd
and unusual things, and a disregard for danger; he likes to make
fools of other people, and to upset them. This ‘is a description
of adult high P scorers; as far as children are concerned, we obtain
a fairly congruent picture of an odd, isolated, troublesome child;
glacial and lacking in human feelings for his fellow-beings and for
animals; aggressive and hostile, even to near-and-dear ones. Such
children try to make up for lack of feeling by indulging in sensation¬
seeking ‘arousal jags’ without thinking of the dangers involved.
Socialization is a concept which is relatively alien to both children
and adults; empathy, feelings of guilt, sensitivity to other people
are notions which are strange and unfamiliar to them. This descrip¬
tion of course, refers in its entirety only to extreme samples;
persons perhaps scoring relatively high, but nearer the middle
range of scores, would of course be far more frequent than extremes,
and would only show these behaviour patterns to a much less highly
developed degree. Psychiatric terms which would seem to assimilate
this kind of behaviour pattern are ‘schizoid’ and ‘psychopathic’;
‘behaviour disorders’ is another term which springs to mind. Our
concept of ‘psychoticism’ overlaps with all three of these diagnostic
terms.

Predictions from this factor on to sexual behaviour are more

difficult to make in view of the limited amount of knowledge we
possess of its nature. The terms that spring to mind are ‘impersonal
sex’ and ‘aggressive sex’, with their implications of lack of love
and affection, the stress on physical relations purely and simply,
and the failure to ‘relate’ in any sense to the sex object. It also

19
 SEX AND PERSONALITY

seems likely that high P scorers might behave in generally anti¬

social ways, and indulge in socially disapproved acts. These pre¬
dictions are less explicit, and rather vague, as compared with those
made in connection with E\ the reason of course is simply that
we know a good deal about E, but much less about P.
One important feature of P scores is that men tend to have much
higher scores than women, and the general aggressiveness, lack
of empathy and impersonal nature of the high P scorer is clearly
allied to popular stereotypes of maleness.(2) It is also interesting
that criminals score much more highly than non-criminals; there
is a distinct tendency for criminals to be male. (The comparison
between criminals and non-criminals is of course within the given
sex; male criminals have higher P scores than male non-criminals,
female criminals than female non-criminals.) If there is indeed a
close and possibly causal link between maleness and high P scoring
(perhaps through the mediation of androgens), then we would expect
that the sexual attitudes and behaviours of high as opposed to low
P scorers would be similar to those of males as opposed to females.
This prediction is of course based on analogy, rather than on firm
deduction, but may repay investigation.
In addition to these three major personality scales, there is a fourth
scale which has been used in some but not all the studies to be
described; this is the L (Lie) or dissimulation scale.
This scale, which was first incorporated in our series of question¬
naires in the E.P.I. (Eysenck and Eysenck 1964), attempts to measure
a tendency on the part of some subjects to ‘fake good’; this tendency
is particularly marked when the questionnaire is administered under
conditions where such a tendency would seem appropriate (e.g. as
part of an employment interview). A series of factorial and experi¬
mental studies has been carried out to investigate the nature of
this scale in some detail (Eysenck and Eysenck 1970b, Michaelis
and Eysenck 1971, Eysenck, Nias and Eysenck 1971). It is clear
that the scale possesses a considerable degree of factorial unity,
with individual items having high loadings on this factor and on no
other; however, there are certain difficulties in regarding scores as
nothing but indicators of dissimulation. The main difficulty seems
to be that, in addition to measuring dissimulation, the L scale also
measures some stable personality factor which may possibly denote
some degree of orthodoxy or lawful rule-obeying behaviour (con¬
formity to conventionality). The evidence is strong on both points.

20
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

The L scale does measure dissimulation; Michaelis and Eysenck

(1971) have shown that it is possible to manipulate L scale scores
by varying the experimental conditions from high to low motivation
to dissimulate. However, if dissimulation were the only factor
affecting the variance of this score, then the reliability of the score
should be a function of the size of score; when scores are low, thus
indicating that subjects are not dissimulating, then the scale should
have low reliability. Empirically, this has not been found to be so;
there is no lowering of reliability of the L scale under conditions
of little dissimulation, and no increase in reliability under condi¬
tions of high dissimulation. Hence the scale must measure some
stable personality function; unfortunately little is known about
the precise nature of this function. (See Eysenck and Eysenck 1976.)
Michaelis and Eysenck have shown that the dissimulation-
motivating conditions can be distinguished fairly adequately from
the non-motivating conditions by noting the correlation between
N and L. When conditions are such as to provide high degrees of
motivation for dissimulation, the correlation between N and L is
relatively high (approaching or even exceeding —0-5). When condi¬
tions are such as to provide little motivation for dissimulation, the
correlation between N and L becomes quite small, or vanishes
altogether. Thus, under conditions of little motivation to dissimulate,
the L scale score may be used as a measure of whatever personality
function is being measured by the scale; this would seem to constitute
an important experimental problem. Under these conditions there
is little point in trying to correct for dissimulation, and hence the
scale should not be used for that purpose. Under conditions of
high motivation to dissimulate, i.e. when the correlation between
N and L is relatively high, the L scale may with advantage be used
to screen out dissimulators, e.g. the highest 5 per cent of L scorers.
No definite cut-off point is being suggested beyond which subjects
should be eliminated, as this depends on the general level of scoring
of the population, as well as its age; the L scale score decreases with
age in children, and increases with age in adults.
In our investigation the L scale therefore fulfils a dual role. As a
scale for the measurement of dissimulation it may be useful to dis¬
cover whether in fact there has been any strong tendency in our
population to dissimulate. As a personality scale for the measure¬
ment of some ill-defined dimension of conservative orthodoxy
or social obedience (conformity; conventionality) we may be able

21
 SEX AND PERSONALITY

to link this with sexual behaviour patterns, predicting that high L

scorers would prefer socially approved attitudes and behaviours.
We have now introduced the scales to be used to measure per¬
sonality, and the theories to be tested; there remains one further
point to be discussed before we describe the scales used for the
measurement of sexual attitudes and behaviours, and go into details
of the experimental procedures. We have criticised Kinsey and his
followers in the opening paragraphs of this chapter for failure to use
random samples; as will be seen presently, in our work too there
is a marked departure from such random sampling. Does this mean
that the criticisms that apply to Kinsey also apply to our own
work? Such a conclusion would seem to be based on a confusion
of the requirements of experimental designs subserving different
purposes, and as this confusion is widespread it may be useful to
discuss it in some detail. The discussion may be oriented by
reference to a recent paper by Cochrane and Duffy (1974), speci¬
fically written to impress readers with the need for random samples,
and criticising psychologists severely for their failure to use such
samples.
Having reviewed samples of papers published in the British
Journal of Psychology and the British Journal of Social and Clinical
Psychology, and having found the sampling in most researches un¬
acceptable, they write:

These findings obviously severely limit the generalizability of most

of the results of our research endeavours. The most common
procedure is for an author to assemble an aggregate of volunteers
(to say ‘sample’ would be misleading) from a university. These
may number less than 100 from a total student population of up
to 20,000. It is impossible to generalize to the total student popula¬
tion of one college and, of course, nothing whatsoever can be said
about the existence of any relationship found in the general popu¬
lation. Only 1-2 per cent of all studies reported in these two journals
over the past four years based their findings on a true sample of the
general adult population, [p. 120]

Even this, one might add, is an exaggeration; there never has been
a ‘true’ sample of the general population, including mental defectives,
schizophrenics, geriatric patients, alcoholics, colour-blind males, etc.,
in their correct proportions; the notion of a ‘true’ sample is a myth.
Is it necessary or desirable?
The point here made is a very simple one. Cochrane and Duffy

22
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

are basing their case on a quite erroneous conception of scientific

method, using an inductive model that has been out of date for many,
many years. In terms of this Baconian model, we discover facts from
a sample, and then generalise these facts to a particular universe.
But this is not the way modern science proceeds, and philosophers
of science are quite explicit about this (Popper 1959, 1972). The
purely inductive approach is philosophically invalid (even with a
‘true’ sample!) and scientifically meaningless. What happens instead
is that we proceed by theory and falsification (‘Conjectures and
Refutations’, in terms of Popper’s famous book). We state a specific
theory which has testable consequences; in so far as the consequences
are in fact discovered when an empirical test is made, then in so
far does our theory survive. Usually the theory will also predict
the type of population to which it applies; the theory that resistance
to the passage of an electric current grows as the conductor is
heated clearly applies only to conductors (although it may also
be found to apply to semi-conductors, etc.). Similarly in psychology;
our theories are either general, i.e. are supposed to apply to all
human beings (or to all mammals, or to all living organisms), or
else they apply to extraverts, as opposed to introverts (Eysenck
1967), or to working-class children, as opposed to middle-class
children (Jensen 1973). If a theory does not specify sub-groups to
which it does not apply, then the assumption is that it applies
universally. This assumption is of course subject to disproof, but
it must govern the design of the experiments conducted to test the
theory.
Putting the matter symbolically, we may say that the total popula¬
tion can be considered to be made up of (overlapping) sub-groups.
Letting P stand for the whole population, we can write:

P=m, /; e, r, y, o\ w, b; t, c; etc.

(in this expression m and / stand for male and female, e and i for
extravert and introvert, y and o for young and old, w and m for
working- and middle-class, w and b for white and black, and t
and c for town and country dwellers). The hypothesis being tested
states that m=f, e—i, y—o, w=m, w=b, t=c, etc.; consequently
these distinctions are not relevant to our choice of subject. It may
be of course that our hypothesis does include one of these terms;
I have argued strongly that extraversion and introversion are
relevant to many psychological experiments, and that we can formu-

23
 SEX AND PERSONALITY

late hypotheses that predict the way in which personality and

experimental paradigm interact (Eysenck 1967). These predictions
have often been verified, and in so far as such hypotheses exist, it
becomes important to include reference to them in the selection and
measurement of the experimental population. We may of course also
be able to make predictions from some well considered theory about
differences to be expected between young and old, male and female,
white and black; if such hypotheses exist, then we must pay atten¬
tion to them in our design. What is quite unscientific, however, is
to demand random samples regardless of existing hypotheses; in this
way no relevant hypotheses are being tested, and we are simply pro¬
ceeding along a path of uncodified induction. That way lies chaos.
Let us consider a typical hypothesis in experimental psychology.
To take one such at random, let us state that reaction time increases
linearly with increase in the number of bits of information in the
stimulus. This is a universal hypothesis; it does not specify that
the prediction applies only to children, or to women, or to students,
or to any particular group that can be induced to undergo the test¬
ing procedure and that understands and is willing to follow the
instructions. It is not clear to me why, in testing it, we should
go to the trouble of assembling a random sample of the population.
A group of subjects assembled in the manner condemned by
Cochrane and Duffy would do very well to test this prediction.
(In the same way can predictions made by astronomers about
planetary orbits be tested in our planetary system, although this
is a very poor sample of all possible planetary systems - on the
basis of Cochrane and Duffy’s argument astronomy could have
very little reason for generalising findings from our system!) There
may be personality and other differences between volunteers and
non-volunteers, as Cochrane and Duffy argue, but these would be
irrelevant to the prediction made by the theory, and consequently
the theory can be tested equally well with volunteers as with
conscripts.
Let us now turn to another type of research, involving the
correlational method. It has been predicted that when social attitude
items are intercorrelated and factor-analysed, then two major factors
would emerge, which have been called conservatism-radicalism and
toughmindedness-tendermindedness (Eysenck 1954). Such an hypo¬
thesis can be tested on any population within the Western world,
being as universal (with that restriction) as the reaction time pre-

24
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

diction. We have in fact tested it with as random a sample as the

public opinion survey firms could manage to give us, and with
many other more restricted samples, including student samples
assembled in the manner criticised by Cochrane and Duffy. In
all cases, the pattern of correlations behaved lawfully, and gave
very similar solutions. We even tried to construct severely truncated
samples, by using only Conservative, or only Labour, voters; even
then the prediction was borne out. Solutions from different popula¬
tions were mathematically transformable into identity by using
Thompson’s restriction of range formula, showing that only by
choosing a population having zero range for one or other of the
two variables could we fail to obtain the expected outcome. What
would a random sample tell us that a non-random sample would
not tell us just as well?
The only exception to this general rule would be a fine of
research that is of the least interest in science, namely the deter¬
mination of population parameters (as in Kinsey’s work). If the
manufacturers of brassieres wish to know the average size of female
breasts at a given age in this country, as well as the shape of the
distribution, then of course they would be forced to use as random
a sample as they could persuade to have their breasts measured.
If we wish to know the average latency of reaction times to a given
stimulus for the whole population, then of course we would have to
test a ‘random’ sample. It is not clear to me why anyone should
ever wish to do such a thing, but given the problem, then clearly
this would be the solution. Needless to say, the resulting figure would
have no scientific interest whatever.
Even problems of this kind can often be approached without
the need of a ‘random sample’. Let us take Milgram’s (1974) work
on obedience to authority as an example. When he demonstrated
that a high proportion of Yale students were ready and willing
to apparently torture and possibly kill a volunteer subject, it was
argued that this might have been due to their status as students,
and the special relation between students and instructor-experi¬
menters, who gave the orders. Later work on non-student volunteers
(who did not know what they were volunteering for, and who were
paid for their trouble), and replications in many other countries,
with many other types of sample, showed convincingly that this
was not the correct explanation. Milgram’s work generates a general
hypothesis regarding obedience; this can be tested equally well with

25
 SEX AND PERSONALITY

different types of samples. Differences between samples would be of

considerable interest, and would require new hypotheses (or possibly
they could be accounted for in terms of his own hypothesis, which
could be used to generate predictions regarding different samples).
But nowhere can it be said that a random sample would have added
noticeably to the general interest or convincingness of his work. The
precise mean value for the population, which could be obtained
only from such a random sample, would be of precisely no interest
at all.
Sometimes it is necessary to tailor an experiment to the specific
nature of the sample, or even to individuals. If we wish to study
the effect of stress on performance, we must make certain that our
experimental procedure does indeed produce stress. The moving
shape of a predatory bird produces stress in doves, but not in dino¬
saurs. Rapid presentation of serial rote-learning stimuli produces
stress in high N subjects, but not in low N subjects. Having to listen
to lectures on the inheritance of intelligence by A. R. Jensen without
disruption and fisticuffs would, we may perhaps conjecture, produce
stress in sociology students. Any experiment on stress would need
to contain specific hypotheses of this kind; the use of a random
sample of subjects would not guarantee equalisation of stress, and
might in fact make it much more difficult to achieve this.
Cochrane and Duffy are worried about the applicability of statis¬
tical methods to non-random samples. The answer again lies in
the hypothesis tested. If this hypothesis does not discriminate be¬
tween different members of the total population, then all samples
are random unless they are specifically selected by reference to
the variables to be tested. If Milgram had selected only people
known to be extremely obedient to authority, then obviously his
results would be suspect. But short of this the hypothesis in most
experiments mounted by psychologists contains the (usually
silent) minor premise that for the purpose of this particular experi¬
ment all persons at risk of selection are equal - possibly the only
occasion where extreme egalitarian demand is fully met! If this
minor premise is justified, then clearly any selection of subjects is
a random one, and the usual statistical methods are applicable. If
the premise is not justified, then it is open to the critic to disprove
it. In doing so explicitly the critic would add greatly to our know¬
ledge of the natural phenomena in question; he would do so far more
than if he simply argued for the desirability of random samples.

26
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

Cochrane and Duffy end by stating their distress at:

... the way in which the elementary rules of sampling are ignored,
along with the known facts about biases being introduced by using
college student volunteers as subjects. It is perhaps particularly un¬
fortunate that virtually no one has the inclination, or the available
resources, to test his hypothesis on the general population. Until we
can do this routinely it appears that most of our research efforts
into human behaviour will be essentially trivial, [p. 121]

The burden of the argument is that it is their criticisms that are

essentially trivial, and that psychologists would be very ill-advised
should they feel obliged to test random samples of the population
whenever they wished to confirm or disprove specific theories.
To use such samples routinely would imply a return to a long-
abandoned scientific philosophy which has been given up quite
universally by scientists and by philosophers of science. All the
arguments in the Cochrane and Duffy article ultimately derive from
this error.
It should not be thought that I have any wish to discourage
the use of random samples in psychological research, or an extension
of the usual research design from using sophomores to using other
types of group. That would be to misunderstand my argument com¬
pletely. There are many considerations that should be borne in
mind in selecting a sample for one’s research. To take but one in¬
stance, it would not be possible in the United States to use college
students for an experiment of the Milgram type, simply because
a large number of potential subjects would have heard all about
the deception involved in the experiment, and would not be fooled.
This would make a non-academic sample mandatory, and even
there considerable caution would have to be observed. What
we are opposing is a mindless rejection of non-random samples,
either in planning one’s own experiments or in evaluating other
people’s. The selection of a proper sample should derive from one’s
theoretical position, and not be imposed by irrelevant Baconian
principles. In certain circumstances a random sample may be appro¬
priate, in others not. These decisions cannot be anticipated by a fal¬
lacious appeal to imaginary ‘scientific methods’; scientific research
cannot be automated and made foolproof by laying down general
rules of this kind.
We would conclude from this discussion that for our purposes ran¬
dom samples would not be needed; the hypotheses tested should

27
 SEX AND PERSONALITY

be capable of being falsified in samples of students, variegated adults

of all sorts or indeed any sample of volunteers selected without
reference to their personality or their sexual behaviour. (If our
population consisted entirely of extreme extraverts, or sexual per¬
verts, then our predictions could not properly be tested, and we
would have to conclude that the choice of subjects was unsuitable.)
It is desirable that our sample should not deviate too much from
the normal distribution of scores on the P, E, N and L variables; as
we shall see this deviation is in fact very small. Similarly, there is
evidence that in sexual experience too at least some of our samples
do not deviate from normalcy. This is all that is required for our
purposes (McCloskey 1967).
The other problems considered in relation to Kinsey’s work, i.e.
conscious or unconscious falsification, do of course arise in relation
to our own work also. They do not arise to the same extent because
we have preserved strict anonymity, in contrast to the interviewing
methods used by Kinsey. We shall present some arguments later on
to show that subjects took the whole inquiry very seriously, and
made every effort to answer truthfully and accurately; this does not
guarantee truthful and accurate answers, but it does suggest a high
degree of cooperation.
There is one further argument that derives from the internal
structure of the results. It seems highly unlikely that the complex
pattern of intercorrelations found, meaningful and replicable as it
has turned out to be, could have been produced unless the great
majority of the subjects had answered truthfully and conscientiously.
But of course there is no final answer to the reader who chooses to
doubt the accuracy of the answers to our questionnaires, and to the
degree that this doubt may invalidate our conclusions there is no
definitive reply. The reader may like to reconsider this point after
reading the relevant comments in the following chapters.

1. For a survey of this work, and an extension to a French sample

see Simon (1972).
2. There are very many studies of sex differences in attitudes and
behaviour; the major results have been well summarised bv Kaean
(1964, p. 143): S

In sum, females are supposed to inhibit aggression and open displays

of sexual urges, to be passive with men, to be nurturant to others, to
cultivate attractiveness, and to maintain an effective, socially poised,
and friendly posture with others. Males are urged to be aggressive

28
 INDIVIDUAL DIFFERENCES IN SEXUAL BEHAVIOUR

in face of attack, independent in problem situations, sexually aggres¬

sive, in control of regressive urges, and suppressive of strong
emotions, especially anxiety.

Fisher (1973), who also reviews some of the literature on ‘being a

woman’, curiously enough attributes observed differences to cultural
factors; his whole book does not contain any mention of genetic
factors!
 2

Sexual attitudes and behaviour in

student populations

The first studies to be carried out in the attempt to determine the

influence of personality on sexual behaviour and attitudes made use
of an inventory specially constructed for the purpose, and covering
a large number of separate issues relating to sex. In constructing
this inventory we made use of existing questionnaires, particularly
one constructed by Thorne (1966), and of discussions in the relevant
psychological and psychiatric literature. Most of the major areas of
sex are covered in this inventory, which is given below; also given
in this table are the percentage ‘Yes’ answers of our male and female
respondents. The second part of the inventory, dealing with actual
behaviours, is given later in this chapter; here we are concerned
more particularly with attitudes. Also administered to our subjects
was another inventory, constructed to measure personality dimen¬
sions P (psychoticism), E (extraversion) and N (neuroticism). The
inventory used was an earlier form of the E.P.Q. (Eysenck Person¬
ality Questionnaire) recently published (Eysenck and Eysenck
1975) . Information about its construction and changes from earlier
forms is given in detail elsewhere (Eysenck and Eysenck
1976) .
The inventories described above were administered to 423 male
and 379 female university students, aged from eighteen to twenty-
two, all unmarried, under conditions of complete anonymity. The
students came from several different universities, most were ap¬
proached in groups and asked to volunteer for the purpose of con¬
tributing to a scientific study of sexual behaviour. There were few
refusals; those who accepted were given copies of the questionnaire
and asked not to give their names. Completed forms were either

30
 STUDENT POPULATIONS

put into a large box, where they were obviously so mixed up with
other questionnaires as to be untraceable to the person who had
filled it, or else sent by post to the Institute in specially provided
envelopes. Discussion with respondents indicated no worry about
possible identification, and considerable interest in the project; in
fact, many students who had not been approached wrote in or con¬
tacted the assistants who were distributing the questionnaires,
indicating their regrets at having been left out, and their desire to
participate. The rather lengthy questionnaires take a good deal of
time to fill in; it seems unlikely that ‘jokers’ and others with intent
to deceive would have gone to the trouble of spending so much
time in order to fool the investigator. A few questionnaires had to
be eliminated because questions had been left out by the subject,
but these were far fewer in proportion than is usual in our ex¬
perience. Many subjects added long comments to some of the
questions, or wrote lengthy additions at the end, often on separately
provided sheets of paper. There was no evidence of any ‘response
set’, i.e. wholesale endorsement of Yes or No answers, or regular
patterns of answering; these were carefully looked for by eye, and
also by computer. The general impression (which must of course be
subjective in the circumstances) was that the population sampled
was unusually responsible and careful in the way they answered
questions, and sincerely desired to make quite clear their actual
attitudes and behaviours. Internal evidence to support this view will
be given later on.
Ninety-six variables were taken from the inventory and inter-
correlated separately by product-moment correlation for males and
females; the items included were questions 1-94, age, and question
98 dichotomised as to whether respondent had or had not had inter¬
course. The resulting matrices were factor-analysed by principal
components procedures; over thirty latent roots exceeded unity, and
further analysis was restricted to the first fifteen factors. It will be
seen that even this may be too many, and in fact only a dozen or so
factors are meaningful. Factors so extracted were rotated by Promax
(Hendrickson and White 1964) into oblique structure; these primary
factors will be discussed first. Later sections will deal with higher-
order factors. Indices of factor comparison (Eysenck and Eysenck
1969) were run between the fifteen male and the fifteen female factors
extracted, and optimum matches were found reasonably easily for
thirteen factors. One female factor had two roughly equally probable

31
 w 2S
vo vo
I 1 n rt- it)
n
VO
m r4

iS £
G 09
o a. u
fc- 0)
c S » O G
L-
<*■> Tj* O *a> ra ^ o m on
rt .2 >, vo ^ m Tf m
O
o ® *G G a
g S &
° d » ” ■g 60 jh 60 O G
oo 2 o2§
*- ™ o
G
*>
9.s a>
CO
o G
<U
a o 8
5t3 G
60
G
G '-G 60
»G co
- — >, co ;-y
G
X §| >*
<u
<D G ’S
^ c3
G G m G *3
2 O ^ su s X) **2
o so X
0)
«>
«
t|
8
O
a
£< . * ’X >» g o *-<
.G. <D <1> a
^ fS w CO *-> G X D G CCJ
TJ**• 5 O
<u o gX
- •s © o D _ >
(U <D
aS
&.SP S
G Mp _
•M
&o;S O ^ 2 -S G <U G co
« * 8?U „> tfl •8 3 & G
*H a 2
U
8
O co “ u *
i- a> <U 60 X
I S « Ui tD a> q.
G
G
s &l-s o 'S
O
C4H M
<0 G3
ti
>
c3 «
G . .

S J3 V! <U ■♦-* «j ^ a> X

s g "o SaS u.’S3 so.S
“Os
t3 a o M C
.G G
cd »-• CM
. <D
.. 60
sa ^ as
G
w G .'G .s »
o -S o .60 co
>% M e| <D 3 ^
o 0) G
I§
C u £
<0 c *C G
<D
G ^
a
o arr
^43 P
O
60 £ J-4
IO
S3
H X <D
“ O G ^ —'
oo
>
O X O
So a m 4U .5
.co G o
!-H
’m
M
G
X sT c e .2
o E -a
§
2
.S
a-0
^ <U
C3 43
-*—> K
a « t-i >
W o O G *3 15 w > o ** a co 3
°ii .2 "3 flj
<1) <D co G O
Q >? ?P
oj
s G
s CO
o o-. G G
0) 5 G x x o
JD » - 3
-S x ft S >
ctf *—<
p s &l “ 0 S •° v- C G M £ « ui *J 3 « a o ~ 8> ^
H 3.S w O cj ±i G
- HH (X, S 00 OlHH coco 3C«2 as
2 »• o g>
| »• x - *■> <30
s; g
ti Vt vo ON O
CN CS fS
Csl
< S o> u
W
2 X *3 ^ W)
Ph S *j .9 §> c
o t;
41 C
2 5 ^‘S
Ui 41
•a a « a m ON O tJ- cn r- oo
3 <3 3 2t-
3 T3 O ^ O
»o 00 rf fN ^ ??3 5?
eSQa^
00
m
Os O vo vo r- VN
^t (N »n vo <s

G >» O
O .
>»r
X
<D
CO
a x>
a
41 T3 X'
o
G
_3 3
60 JS
■M o O
13 3 G
G & > x & O
o o
co
o 6
o M T3
CO
G S * co o G
o <l> o ■M * G
X
o
a
rj
G
vu
>»G3
co
t +3
a G G .* *~~! r—X
O <
n O Q) G
o
co
G . ■*—»
s-°
_ 2 C3 G -O O V
X Q}
O M
W5 03
o o
t) c3
> M o
,2-ft > T3 rt
~H o) 3
S *
u
« aa SO
G
co x g
<U ^
CO
ip M
CO T3
-4- «
^ >
0,’C
a
G <p ^ •ti g* e a G T3
O 5c'fl >.
o _ X a
Jg x a c 3 ■8'? 8 «
(U cj o o o <6
■> 3 :-s s .a a a
<M <U co 2
c ^3 1i.|-
•« xs v 41 ^ _b 30 X
M
o . ^ 60-0
S ^
41 C § 3 — 6 ’
S
^Orii
® 60 W3 • — >
rs g « G3 X os J3
.*3 O 9-* ° s1 G
a* .2 U 60< 8 > fthH 2Si
G
G h
2f x
G _

r4 X t> 00 ON O ^

32
 ON N r- *r> m oo vo co h «r> N o co NO
ON \o n 10 cm <m vo cm -h CM oo

Tj- Tf Tt o NO Tt oo CM VN VO tJ* vo cm NO
ON oo NO N VO -H m cm CM vo CM

>> X x <D

It’s hard to talk with people of the

I didn’t learn the facts of life until

X x h X >> CM
3
o a a <d $ o a> ^ 405> N O X

■a
73 05 o 05 cd
cd 3
X >» M
cd >.5 00
G e +-» o <D O
M • —< D 05 2 g<
M <u o G Ch

■S
Si *
o ^
Cd
«
3
jo SO 3 2 3
cd
00
<>
o
1>
3 D
00
3
cd
.75
O
x t;
c«
*72
2 §
G
£
>. g
£ 8
X
0)
05
3 .Sow
o
•5 '& S <D 00
11 C.S
_i oj
03 q
G,
o
0) TJ
D .C 4>
£ *c > <d
> X ^ «5 «3 G 4>
43
4> a
8
X
<D ®P
s
£
£ 13 a
•O cd
<D
cd *0 CM « x is
-G O O >• 8 £
<D
<D
s
-*-* ■Srt 2 3
"3
3 O
§
G x £ O 'S O h
^ 05
4)
0)
•72 X 05 <U X ° <D ed -3 43 13 4= u S X)

I was quite old

3 <D <D e 3> £> 3
O
43
co
T3 73 3 §! U ^ u
'3 o fi
X
M <U cd
n-k ^

opposite sex
f-§,2 f 3 <« 8 « £ <u
Sug
00-3 ^ !l s
Si *3 O o
Cd •—' O c»
d) 05
M'S _ » ^
Q, O
g § g»a
C
4) T3 cm
o 6 X '33
o « -X -G w 05
r* m
•—
05
-m
O
X>
SJ5
O *3 M
D
^
o 52
G
o cj ‘G
1a bO
d a) ^

3 3 ^ M -4_J G M 1) c .S U
O_ O G ^ 73 s JS °
x «*2 D © oo 52 « J3 «
43
<J
O £ G “ B » oo xS .2
M O X) H H 3
I«
H 3 HH G
*3
C3 IG ^
^ 7H >>
HM ^ Sm E
6 «
HM O hH

00 ON O ^ CM f<N Tt «TN vo -h CM m o-> 56

57
r- oo on o
m Nf Nf Tt Tf Ti¬ ’•d’ Tfr Tt- Tf Tt V-5 <o «o >o V-5

p m oo vo VN
4

NO On CM
47
r- ^ O .
CM Tf CM oo 1M rr, CM

vo vo —I m
4

CM ON 00 00 On NO O CM
53

m CM vo CM ^ r- CM NO m ’-h

D
People of my own sex frequently

There are some things I wouldn’t

x X G
D
X >» G O
<D <D O <D ^ <D
O O
(D
05
G
C
X2 E O
G ^3 o
<D o S 3 CM D, ^ E E
CG Oh w «D <D X
<D 3
rt X
M <D S o Oh -3 <D 60 G
<D
<D O <D a CM 05 O E
X) x: g x u
a o - S X
^■as 43 cd lx
G U
is 3
want to do with anyone

e « X
G > <u
o ^ g oi 8 JS a
X
o <D
E G 05 J3 ^ M ■B ° 3 O
. , G
(3 •3 X
>* h
>•’3 v'B
D
D
E D O -O M > l-H *3 « 3 3
G -G u
C J 3 60 c3 -M « y D
5 D
E >> O M 3
O
X2 c3 n O 73
-G u a
t- 05 3 5tj -xcd £l 3 3
X D D 00 c« cd O O
^ V A >» 05 ^
X X
xt
cj
3 3 ^ ^ aO ’S >4 G >> 3X *®°
75 ;^ cd V)
attract me

O o £ £ ^ ^ Cq w G M E ^ d G
X2 -M cd 05 >1
“ S T3 —, CS ^ G 05 d
O 05 g -3^ 43 * 05
M 3 75 D
M « « S3 73 cdco
<D •8 3 -E -o ^ ^ £ « S) 8 <D X
O ■MD G D
£ •S3
05
>
<D
-1
3 43 S 9 5 g
g w hfi ” o e
G O G, ? E ^ G 4) O XJ G
X ^
o >> U rj cfl
U G
43 _ W H H H ^ Ph 43 < S cm O D X £S Eh5
r- *A5
36

37

m rf »T5 X 00 ON O CM m
CM CM CM CM CM CM CM m m m m

33
 *> fl)
CO ^

?J Tf
»n
Ov VO
»n
»n O in
m
O O 00
Tf hH

» S3 L
GO >
£ i
a c «
© cd -G 00 O r- m
o <3 VO fM m m r- vo Tt
© is
Qh
n3 o BO on c <u a a X D 00
© O
£ ~
C a c
o 00
a
£
cd E ox <L>
<D
O C
c/d ,C3
(D
O 2 8 S a ’■o o'®
Gh GO t> * O ^43 V
Uh ftx G
O G
cd #G
*53 a
3
<D CO

O pG
C/D 2» O
Cd 3 O
E t>'
G
© 2
G
01 &
X Uh h-h
>> X E- cd <D O o -g
u a'S o
Uh C
« g
H-» a a G
cd >.d
"O 3 X © h—» © G M
•d o >>s c T3 -G
o •s ©
a 13 > _ X
§ a * © ' « Co C/D H-»
<D O
INVENTORY OF ATTITUDES TO SEX—cont.

o © pX
o-3
XI -a IS T3 ^ _. j> O
pG .d <D > cd
O *■*
1 °§ <_, a 13 __ CO
2 00 a S ,
•g o -X “ o cd P 13 H-* ^ G
0>
^ ©
> Uh 4- O cd o> t» o
Uh S Uh
c-gS a • hH ^ *G C c >>
Oh "G £ € ft S
o o m m <u a "1—1 cd i-^h 00 0) • i Uh
pX ’
O'M ^ CTj
0> Qh c^ 2 Uh ,
<L> 2 x M t3
“a C Uh
«H O >?X> <d c3 d > O <D , cd
_H ^ a d g oj w x (D G
G X T3 J3 y « O czi E O
► ’g
. c- o ^
00 03 ^ O © O
*G G
0
5 Uh
« 1 fli oo
5J^o
>
cd
c Dh CO
G
O 1)
Uh *2-*
CD
3
u
G _ © Uh £2 LX « E O ^ g* d
o g X si O
>H O uD >. HH O
2 T3 » 60 g
(/) Cd HH > m o K 8, « ^ o
pG u.
° S-
^ cu — 3
-
Q &m oa is
<N m n vo r- oo ov O <n
r- r- r- r- 0© 00 00 00

CO ©

i cd
vo oo m
r- 4-t
in
VO
m
8 n
m
r4 3 ^
CN
m
,'d-
<u e s
SO u Uh

H co
<3 c u
o
Uh
c«d
ca ■d- r) N vo ov tj- 00 Tf CN
© 5 fS ON N 00 *n CN ro 00 Tf in
Cl <<

G © c«
6 p^
© © © © G pO ©
X
cd
2 Uh pX
E O ©
Uh
©
to
^

C/3
Uh
© cd
G pO G
CO
cd
G
a>
o
>
G ©
6 © G S rs£
-G
£
'3p X
©
©
Uh
cd ©
>
^
00
o E'‘c
cd
2
© pG
X Cu «>
■ft 5 *
cd oo
JD © G •73 3 © 2 x >>
X <u O O g s > © G E
3 © ^
C/5
dO pG ft E cd 2 2 G cd u- G
a ©
cd cd to O pG
f
©
^X 3
p- ? E 00 -2 E
T3 00 H-> O O o G
Td *c
CO
2 G £ ©
§23 G
« ij G -G pG . ©
3 a,
G ©
C/5 ^ Uh
S JD 00 P v ?d E ^G a
*© ©
G 3 “X G x
-X
o G i ux 2
■5 o M O 9 _ 'S O jG ©
o «Z) > g a pG cd *7^ •*-» -G 55
Uh
3)
CO
Dh ^ 00 ‘pG G ^
D Uh
h x o «
G fc O G ♦pH TD CO >»
Uh © E cd pG
O ft "J ~ © Dh S
G ^ ft
w co ptH pG
© CD co oo “
cd co ©
G
cc G ^
G< G
G ©
O g O pG4
6 >. £> cu 3 00 QJ U-H
Uh
b g 00 b -S ^pX Gh CO r-H
•—I
© -H
^ Uh
aj pG ’c
*—< 9 <U 3
Cd
G
© ©
>
iig
^
E«
c 5 © 3
G G
©
.a ,o ©
Uh
©
Uh Cd
© 3
Gh'O
© ^ pG 5
’§ § a O b © O ^ I ° >< § © >- 3 © g © pG cd
H h cd co E E £ ftOh G >—( So (/) oo C/0 HH ^ < ob
00 av o i-h <N PO in vo r- 00 ov o
in in vo vo VO VO VO vo vo vo VO vo r-

34
 b c
o 0)
S a (D cd
X
cd « cd
>> c
g O <3 5-h
a cd 2 >*
<D b
<D
■o X cd -73 T3
a) p*» <u o
<D > ^ -a <D o >>
cd <D G <D g £ d»b £ >> n
<u
cd cd _
<U <D t- 2 -o «
eg 3 >
<L> , ■* Si o o ta £ 5 C3 ^ 1c/d s<U o S| „
cd
C* 5 « 3
^ G co
^ O
oS z KZ Xi
b
<u
T3 <D oo T) <U M-j o
>»
a>
>
4) .3
52 '•s& G
3
3
o b ..
<D tao d>

intercourse?
cd O <D
3 a. «M G 2P
£
O <D
K
<D
°
<D
£.2 ° c "
> o *g
V > g « tS C-
Q
<U G <U G (j « <M a X
zo zo> "So
"b b
>02 £
cd <3 4-t
a§50 cd X sc.s «je Btr.3 3SS <
Tf <TN VO oo
ON ON ON ON ON

<N ra t*- r4 Os t^
m
VO
CN m m §

ON CN (N (N O VO
»o VO ‘/'N VO 05 00
c
Q
•«—»
<l> >> a> u 0) G O 1)
>> «
C/D ^ P
X — XI X 00
cd t- ^ O - C
b
cd
00 3, ’C b O O
T3 <L> 0> 3 O fcn o a
3 3 cd P o x«
I s
(b) Reject it

fc ^
11“ 5-°
c o
£2 2
O 3 a>
G X
O X C/5
C/5 cd QJ
'S -j
-|J! G Wh G r.
0) b
3 £.
« g
O
3 S’ 3rrt g.
M a_, tn
O U > o ^5
a
a> x O ;>> X> >, cd o o a T3 "O
u “^3 X
■S2 > X 0)
G wO
S J5 3 X> 3 £ <U
!- O cd ’O g (-> o
3 3 rt Q 3 D (U o <D
£ O X
^ ■<-> US g E ^ ° 2 60 X - X
■rt o 3xo « 45 Ig O
.2 « S3 g 5 « x 3 'u X ^
>> > ui —; 3 2 jj*
(a) Accept it

0) ^
■s e
CZ) <U O
X iS C H 3 cd 3 3
« 2 ■
0-00, </D O<D S a 3 2
S3-13 6 2 2 M
co \3 _
o) 1,3
3 _,-
3
O >.6 ^ g
■-
u c3
00 3 <
C > •3 rt X 3^ * “
ScatH<u H
r, tH C« 0> <L> u (D 13 § l. g-l
G Td co G g O S3 b
*-h o g-s
O
<u <D
v »2« ■M D n ^ in X ^ s
° a Jr &Q O X C 3 5 > 3 £ u M-( X
p-i «+h a ^ H o H 0> !> «) r* & ^ X . a a

b- I/") 00 ON o fN m
oo 00 oo oo oo 00 ON On Ov.

35
 SEX AND PERSONALITY

male matches, and one had none. Indices of factor comparison will
be given in relation to each pair of factors presented. For each factor,
items and loadings will be given in such a way that all loadings
exceeding the arbitrary figure of 0-3 for either men or women are
quoted (when indices of factor similarity are low, this rule has not
always been followed in order to avoid swamping the tables with
masses of irrelevant figures). The factors have been given names
which are of course rather arbitrary; however, they help to identify
factors and may serve a useful purpose in making possible verbal
reference. Alternative labels might be preferred by many readers,
but the actual names used are not of course of any great import¬
ance.

Factor 1. Sexual satisfaction

The index of factor similarity of the very clearly marked factor of

sexual satisfaction is 0-97; in other words, men and women agree
almost perfectly in the structure of this factor. Table 2.1 gives the
numbers of the questions involved and the loadings for both men
and women. The highest loadings are for questions indicating ‘I am
satisfied with my sex life’, ‘I have not been deprived sexually’,
‘Nothing is lacking in my sex life’, ‘My love life has not been
disappointing’, and ‘I don’t worry about sex’. There is only
one reversal of sign between the sexes: men who are satisfied
sexually agree that we should do away with marriage; women
don’t.

Table 2.1 Loadings - sexual satisfaction

Question Males Females

4 0-91 0-82
11 -0-60 -0-58
17 0-41 0-30
20 -085 -0-88
22 -0-68 -0-59
32 -0-39 -0-32
35 -0-33 -0-23
60 -0-43 -0-23
89 018 -0-36

36
 STUDENT POPULATIONS

Factor 2. Sexual excitement

The index of factor similarity for sexual excitement is 0-94; here

too there is great similarity in factor structure between the sexes.
Table 2.2 gives all the details regarding the loadings for this factor.
The highest loadings are for items indicating ‘It doesn’t take
much to get me excited sexually’, ‘I get excited sexually very easily’,
‘When I get excited I can think of nothing else but satisfaction’,
‘I get very excited when touching a woman’s breasts’, and ‘Some¬
times sexual feelings overpower me’. For such people, ‘conditions
don’t have to be just right to get me excited sexually’, and they don’t
think ‘only rarely about sex’. It is interesting that the item ‘I find
the thought of a coloured sex partner particularly exciting’ has a
high loading only for women; for men this is quite negligible.

Table 2.2 Loadings - sexual excitement

Question Males Females

3 -0 31 -0-53
6 -0-43 -014
7 0-28 0-30
33 0-77 0-86
39 0-36 0-50
41 0-38 0-29
43 0-81 0-86
46 001 0-34
50 0-36 0-30
52 0-32 0-23
53 0-43 0-28
76 0-39 0-39

Factor 3. Sexual nervousness

Sexual nervousness too is clearly marked, with an index of factor

similarity of 0-92. Table 2.3 gives the details concerning the factor.
High-loading items are ‘I don’t have many friends of the opposite
sex’, ‘I feel nervous with the opposite sex’, ‘I don’t feel at ease
with people of the opposite sex’, ‘I feel more comfortable when I am
with my own sex’. There is no reason to assume any tinge of homo¬
sexuality in this factor; we will encounter a proper homosexuality
factor later on.

37
 SEX AND PERSONALITY

Table 2.3 Loadings - sexual nervousness

Question Males Females

16 006 0-32
31 -0-49 -0-66
51 0-76 0-75
54 -0-88 -0-89
56 0-77 0-81
58 0-58 0-77
66 0-22 0-36

Factor 4. Sexual curiosity

The index of factor similarity for sexual curiosity is 0 90. The factor
itself is very clear (table 2.4); high-loading items are ‘I like to
look at sexy pictures’, ‘Sex jokes don’t disgust me’, ‘I like to look
at pictures of nudes’, ‘I would take a chance of seeing people make
love’, ‘I would agree to see a “blue” film’, ‘I would read a highly
pornographic book’. These questions are all concerned with a liking
for pornography and voyeurism and a desire to have vicarious sex
experiences.

Table 2.4 Loadings - sexual curiosity

Question Males Females

10 0-34 0-37
47 0-78 0-77
62 -0-40 -0-45
69 -0-35 -007
81 0-84 0-77
82 0-41 0-27
83 0-34 0-49
92 0-32 0-54
93 0-27 0-54

Factor 5. Premarital sex(l)

The index of factor similarity for premarital sex is 0-79, indicative

of some diversion in factor loadings between the sexes. The items
which load highly for both groups are ‘Virginity is a girl’s most
valuable possession’, 'It would bother me if the person I married

38
 STUDENT POPULATIONS

were not a virgin’, ‘One should not experiment with sex before
marriage’, ‘It is better not to have sex relations until you are
married’, ‘I have not had intercourse’. Other items, loading not
so highly, are ‘The Pill should not be universally available’, ‘It is
not all right to seduce a person who is old enough to know what
they are doing’, ‘Women should not be sexually aggressive’. Details
about this factor are given in table 2.5. Growing older is associated
with a shedding of these beliefs for women, but not for men.

Table 2.5 Loadings - premarital sex

Question Males Females

Age 003 0-45

5 0-57 0-70
23 -0-04 0-30
26 -0-72 -0-75
42 0-54 0-80
45 0-71 0-86
61 -0-34 -0-27
65 0-32 -002
68 -007 -0-40
79 -0-26 -0-44
90 012 0-34
98 0-42 0-81

Factor 6. Repression

The index of factor similarity for repression is 0-75; this is not

very high and suggests that somewhat different items make up this
factor for males and females. It can, therefore, not be expected to
be as clearly interpretable as previous factors. High-loading items
are ‘Children should not be taught about sex’, ‘I would particularly
protect my children from contact with sex’, ‘I think only rarely about
sex’, ‘Masturbation is unhealthy’, ‘I don’t think about sex almost
every day’, ‘My religious beliefs are against sex’, ‘Self-relief is
dangerous, even when done in a healthy way’, ‘Men marry to have
intercourse, women have intercourse for the sake of marriage’,
‘I have strong sex feelings but when I get a chance I can’t seem
to express myself’. This factor has an almost Victorian flavour;
it seems to be more prevalent with the younger members of our
sample, although this is more true of men than of women. Details
are given in table 2.6.

39
 SEX AND PERSONALITY

Table 2.6 Loadings — repression

Question Males Females

Age -0 31 -013
6 0-30 0-51
8 0-37 0-28
32 0-33 010
38 -0-61 -0-59
41 -0-26 -0-32
42 0-35 Oil
49 000 0-60
74 016 0-66
75 -0-52 -0-32
90 0-38 013

Factor 7. Prudishness

The index of factor similarity for prudishness is 0-72; what was

said about the previous factor is obviously equally relevant here.
High-loading items are T don’t like to be kissed’, ‘I don’t enjoy
petting, ‘The thought of a sex orgy is disgusting to me’, ‘Sex
jokes disgust me’, ‘It is disturbing to see necking in public’, ‘Sexual
feelings are sometimes unpleasant to me’, ‘I consciously try to keep
sex thoughts out of my mind’, ‘I had some bad sex experiences
when I was young’; the remaining three items seem to load only
for men, and refer to a preference for homosexual over heterosexual
company. Table 2.7 lists the relevant items and loadings.

Table 2.7 Loadings - prudishness

Question Males Females

Age -006 0-32
16 -0-56 -0-30
18 -0-24 -0-74
19 -0-22 -0-42
24 -008 -0-39
27 001 -0-33
30 0-40 -003
36 -064 -005
44 -004 -0-32
55 -0-75 -0-45
59 0-74 014
69 -0-23 -0-53

40
 STUDENT POPULATIONS

Factor 8. Sexual experimentation

Sexual experimentation has an index of factor similarity of 068,

but the meaning is reasonably clear-cut as only three items have
loadings on it. These are shown in table 2.8; they are ‘Young people
should learn about sex through their own experience’, ‘A person
should learn about sex gradually by experimenting with it’ and
‘Young people should be allowed out at night without being too
closely checked’.

Table 2.8 Loadings - sexual experimentation

Question Males Females

67 0-77 0-76
71 0-69 0-75
72 015 0-37

Factor 9. Homosexuality

Homosexuality has an index of factor similarity of only 060, but

again the meaning is unmistakable. Items with high and congruent
loadings are ‘Homosexuality is normal for some people’, ‘I under-
stand homosexuals’, ‘People of my own sex frequently attract me’.
‘I have had conflicts about my sex feelings towards a person of
my own sex’, ‘I am embarrassed to talk about sex’. Items are listed
in table 2.9.

Table 2.9 Loadings - homosexuality

Question Males Females

27 000 0-30
30 0-54 0-27
35 -007 -0-42
36 0-36 000
40 0-54 0-73
57 009 -0-33
61 -0-32 -012
66 0-39 0-28
78 0-64 0-76

41
 SEX AND PERSONALITY

Factor 10. Censorship

Censorship has an index of factor similarity of 0.52, but is again

not difficult to interpret. Items with high and congruent loadings are
‘There should be censorship on sexual grounds of plays and films’,
‘There are too many immoral plays on T.V.’, ‘Prostitution should
not be legally permitted’, ‘Pornographic writings should not be
freely allowed to be published’, ‘Young people should not be
allowed out at night without close supervision’, ‘The Pill should
not be universally available’, ‘It is disgusting to see animals having
sex relations in the street’. Details about loadings are given in table
2.10.

Table 2.10 Loadings - censorship

Question Males Females

12 0-61 0-19
14 -0-35 010
61 -018 -0-45
72 -0-35 -0-29
74 0-54 012
84 -0-47 -0-65
85 -0-33 -0-41
86 012 -0-37
87 0-54 0-45
91 -0-50 -0-65

Factor 11. Promiscuity

The index of factor similarity for promiscuity is 0-50; nevertheless,

interpretation is not too difficult. Items with high and congruent
loadings are ‘Sex without love (“impersonal sex”) is not highly
unsatisfactory’, ‘I do not need to respect a woman (man), or love
her/him, in order to enjoy petting and/or intercourse with her/
him’, ‘The thought of a sex orgy is not disgusting to me’, ‘I believe
in taking my pleasures where I find them’, ‘I have been involved in
more than one sex affair at the same time’, ‘I would see a “blue”
film, read a pornographic book, or take part in a sex orgy’. Details
regarding this factor are shown in table 2.11.

42
 STUDENT POPULATIONS

Table 2.11 Loadings — promiscuity

Question Males Females

Age 0-38 006
2 -0-24 -0-56
13 0-33 050
44 -0-76 -0-48
70 0-23 0-50
77 010 0-51
92 0-48 009
93 0-44 006
94 0-70 0-37

Factor 12. Sexual hostility

Sexual hostility has an index of factor similarity of only 048, and

its interpretation is not very certain. Details about loadings are given
in table 2.12. Items with high and reasonably congruent loadings
are ‘I have felt like humiliating my sex partner’, ‘I have felt hostile
to my sex partner’; these are the only items that are both high
and congruent. Two further items, with much lower loadings, are ‘I
have been involved in more than one sex affair at the same time’, and
‘Sexual feelings are sometimes unpleasant to me’.

Table 2.12 Loadings - sexual hostility

Question Males Females

7 000 0-32
19 0-24 0-33
21 -006 -0-47
25 Oil 0-35
60 0-04 0-45
63 -005 0-32
73 0-69 0-62
77 0-36 0-20
80 0-74 0-66
83 -002 0-38

Factor 13. Guilt

Guilt has an index of factor similarity of 0 82, but nevertheless

43
 SEX AND PERSONALITY

produces some problems in interpretation. These are due to the

fact that one item was duplicated (28 and 65) in an attempt to check
on the care with which the inventory was filled in; the very high
correlation between these two identical items (0-78 and 0-82, res¬
pectively, for males and females) suggests reasonable reliability,
but does of course produce spuriously high loadings for the two
replications of these items. For the purpose of interpretation, these
very high values should therefore be disregarded. High and con¬
gruent loadings are then found on the following items: ‘I have felt
guilty about sex experiences’, ‘At times I have been afraid of myself
for what I might do sexually’, ‘My conscience bothers me too much’,
‘Sometimes sexual feelings overpower me’, ‘Sex thoughts drive me
almost crazy’, ‘I worry a lot about sex’, ‘Sometimes thinking about
sex makes me very nervous’, ‘Perverted thoughts have sometimes
bothered me’, ‘Sometimes it has been a problem to control my sex
feelings . The general picture is clearly one of guilt-ridden neurotic
reactions to sexual thoughts and experiences; details regarding
loadings are given in table 2.13.

Table 2.13 Loadings - guilt

Question Males Females

7 0-49 019
19 0 31 004
25 0-49 0-31
28 0-70 0-90
29 0-55 0-23
30 010 0-64
34 0-35 002
35 0-37 015
36 006 0-30
48 0-45 0-23
50 0-32 005
52 005 0-32
60 0-41 007
63 0-41 0-14
65 0-67 0-92
82 0-32 013

Other factors

These thirteen factors exhaust the clearly replicated and inter¬

pretable factors; the remainder are difficult to interpret and do not

44
 STUDENT POPULATIONS

have obvious duplicates when male and female results are com¬
pared. One female factor has indices of factor comparison of 0 71
and 0-69 with two male factors; we shall call these two male factors
14 and 16. One of these factors may be called inhibition (factor 14);
items having high and congruent loadings are ‘My parents’ influence
has inhibited me sexually’ (0-25; 0-40); ‘I didn’t learn the facts of
life until I was quite old’ (0-33; 0-34), ‘My sex behaviour has caused
me some trouble (0 44; 0T8), ‘Conditions have to be just right to
get me excited sexually’ (0-35; 0-14), ‘Sex contacts have been a
problem to me’ (0 07; 0-38), ‘Virginity is a girl’s most valuable
possession’ (0-32; O'M), ‘Self-relief is not dangerous so long as it
is done in a healthy way’ (0 34; 0 03). Factor 16 is essentially a
doublet, consisting of items 9 and 37; these are worded almost
identically, with the exception that one is worded positively, the
other negatively. This factor is therefore an artefact of no interest.
Factor 15 exists only for the females, having no match among the
males; it might be called a dual standard factor, as item 88 has the
only high loading (0-74). Other items loading on this factor are
‘The opposite sex will respect you more if you are not too familiar
with them’ (0-39), ‘There are things I wouldn’t do with anyone’
(0-37), ‘I have strong sex feelings but when I get the chance I
can’t express them’ (0-31). This factor is of marginal meaning and
interest.
The relations obtaining between these sex attitude factors on
the one hand and P, E and N on the other were investigated along
two lines. In the first place, a method was used of reflecting the
personality variables into the factor space, thus giving the estimated
factor loadings for the three personality variables on the fourteen
sex factors outlined above. Computations for the extension analysis
utilised a fortran programme fax.(2) The matrix of loadings for
the extension variables on the factors is computed as the triple¬
product Ret Rtt-1 Ptf, where Ret is the matrix of correlations between
the extension variables and the variables or tests in the factor
analysis, Rtt is the matrix of the correlations among the variables
or tests in the factor analysis and Ptf is the matrix of primary factor
loadings for the variables or tests in the factor analysis. (This analysis
exploits the fact that the factor analysis has been carried out ‘with
unities in the diagonal’ and thus bypasses adjustment owing to com-
munalities.)
In addition, a detailed analysis was made of the differences be-

45
 SEX AND PERSONALITY

tween high, medium and low scorers on P, E and N for each item
loading above 0-3 on each of the factors. The final results of these
analyses are incorporated on table 2.14, which shows the obtained
relations by direction (+ or —) and by strength (number of + and
— signs). A ‘0’ means that there is no relation either way, single
+ or — symbols indicate a weak relation and two or three +
or — signs indicate a moderate or strong relation. It will be seen
that persons scoring high or low on any of the personality factors
differ from each other and that the three personality traits have
quite different patterns of sexual attitudes. On the whole, these
patterns seem to bear out expectations quite well. A second method
of analysing personality-attitude relations will be discussed after
considering the superfactors arising from the correlations between
the primary factors.

Table 2.14 Relations between personality types P, E and N and sexual

attitude factors*

Factor P E N

1 Satisfaction —
+ —

2 Excitement + + ++
3 Nervousness 0 —
++
4 Curiosity ++ 0 +
5 Premarital sex ++ + 0
6 Repression- - 0 0
7 Prudishness + —
+
8 Experimentation + + 0
9 Homosexuality + 0 +
10 Censorship - - 0
11 Promiscuity +++ ++ 0
12 Hostility +++ 0 +++
13 Guilt 0 0 +++
14 Inhibition + 0 +++
*+ and - signs indicate positive or negative relations; 0 indicates absence
of any relation

Oblique factor analysis, using Promax methods of rotation, results

in factors that are themselves correlated and thus enables the inter¬
correlations between factors to be in turn factor-analysed. A few
examples may make the position clearer. In women, the premarital
sex factor is correlated 0-37 with excitement, -0-38 with nervous¬
ness, —040 with prudishness and —0-32 with inhibition. In men,
satisfaction is correlated -0-41 with nervousness and -0-26 with
guilt, premarital sex —0-30 with inhibition, repression 0-26 with

46
 STUDENT POPULATIONS

inhibition. Five second-order factors were extracted for the men,

four for the women; these resemble in many ways the primary
factors from whose intercorrelations they are inferred. For both
sexes, the first factor is one of satisfaction, combining items from
primary factors 1 and 3 reversed. When the three personality factors
are reflected into this new sex factor, N has the highest loadings
(—048 and —0-45 for men and women, respectively), with E also
having respectable loadings (0-38 and 035); P has slight loadings
of —0 07 and -0-19. This is not unexpected, as high P scorers
also score low on nervousness and satisfaction; thus they score
in opposite directions, on the two primary factors that go to make
up this second-order factor. The index of factor similarity for the
two sexes is 0-88.
Table 2.15 presents correlations among primary factors for the
male sample, and table 2.16 the same data for the female sample.
The factors are presented in order of extraction, but are identified
by the labelling used in the body of the paper. Factors preceded
by a minus sign have been multiplied by —1.
Male factor 4 and female factor 4 have an index of factor
similarity of 0 87; this factor represents a combination of excite¬
ment and guilt. When the three personality factors are reflected
into this new factor, P (0-22 and 0-21) and N (0-20 and 0 49) have
positive loadings, E (—0T2 and —0-15) low negative ones. Male
factor 3 and female factor 2 have an index of factor similarity of
0 80; this factor presents a combination of curiosity and promis¬
cuity. Loadings of the three personality factors are P (0T5 and 0-46),
N (0T7 and 0-23) and E (0 05 and 0 04). Extraversion is thus not
related to this factor, but both P and N are.
The last female factor finds no ready match among the male
factors; it resembles primary factor 6 (repression). Correlations
with personality factors are 0-45 with N and —0 35 with E\ with P
the correlation is 0-19. Male factors 2 and 5 are not clearly enough
defined to make interpretation meaningful, and no discussion will
therefore be given on them here. It is not unusual to find that primary
and third-order factors are more meaningful than second-order
factors (Eysenck and Eysenck 1969), although why this should
be is not clear. In any case, the intercorrelations among the second-
order factors have again been factor-analysed, and for both
men and women the analysis resulted in two clear-cut factors.
These are almost identical for the two sexes, indices of factor

47
 8

vo
8 r4 vo
6
I
oo
8 8 o
6 6
1

* Factors preceded by a minus sign have been multiplied by - 1 in order to give the direction
8 S6 »-h
6
o
m
6
l l

m
8 O o o ON
6 6 6 6
Correlations among primary factors for male sample

i i

ON n CJ
o 8 o o
6 6 6 6 6
i i

n
6
s
6
o
6
8
6
CnJ
6
s6
i 1 1

ON 00 ON
O o <N o
6 6 6 6 6 6 6
i

of the factor as presented in the body of the paper

i i l

C4 m vo VO
.-H
co
V—t r*
© o o fS
O 6 6 6 6 6 6
i i

o ra o on m vo r- 00
o o © o O o o
6 6 6 6 6 6

VO VO vo oo Tt r- »o
o o o ^ ^ ^ o
6 o 6 6 © 6 6 6 6 6
Table 2.15

i i i i
Tt- 00 vo os
o © *-h © o o 2 (N
o 6 6 6 6 6 6 6 6 6 6
I i i i
o oo O VO m r^J oo vp
p o o O o <N C4 o o o CN o
6 6 6 6 6 6 6 6 6 6 6 6
1 i l 1 i l

o On VO V-N ON OO vo o m 00
o O O o 8 O o O o i-H o
6 6 6 6 6 6 6 6 6 6 6 6 o
1 i l i 1

o vo r- o oo m fO VO
o o T—^ O i-H o o o o
6 6 6 6 6 6 6 6 6 6 6 6 6 6
i I i i 1 i i l
1"H CN o <N VO m o\ 00 v©
1—H
 Table 2.16 Correlations among primary factors for female sample

Factors preceded by a minus sign have been multiplied by — 1 in order to give the direction
of the factor as presented in the body of the paper
 SEX AND PERSONALITY

similarity being 0-95 and 0-93, respectively. We would expect these

factors to be reasonably meaningful, and this expectation is not
disappointed. Factor 1 combines all the aspects of pathology and
deprivation that have found expression in the various primary
factors listed; this factor may be named sexual pathology.(3) The
second factor combines all the aspects of permissiveness and active
sexuality that have been found in various other primary factors; this
factor may be called sexual libido. These two factors are uncor¬
related for both males and females.
The three personality factors were reflected into the sex question¬
naire item space, and the results bear out what one would have
expected from the detailed consideration of the various items
reported in previous sections. Sexual pathology is correlated
primarily with N, correlations being 040 for women and 0-47 for
men. Correlations with P are also positive, but too low to be of
much interest; they are 0-08 for women and 0T2 for men. Correla¬
tions with E are negative, as expected, and not very large; values are
—0-23 for women and —0-27 for men. Thus dysthymics (high N, low
E) would be most likely subjects for sexual pathology. Sexual libido
is positively correlated with P, the correlations being as high as 0-40
for women and dropping to 0T9 for men; this difference may be
related to the facts that men tend to score much higher on P than
women and that strong sexual desires traditionally characterise
men rather than women. N is also correlated with libido, but the
correlation is only 0-29 for women and 0 06 for men; the latter
correlation is clearly insignificant. E also shows only slight correla¬
tions with libido, 0T3, for women and 0T5 for men.
The data here given seem to establish that different personality
types differ profoundly in their attitudes to sexual issues, as well
as in their sexual behaviours. Such differences can be observed
for most of the fifteen or so factors into which sexual attitudes
were analysed. It is interesting that these factors appeared so clearly,
and in so similar form, in both the male and the female samples,
and that the personality differences in one sample tended to re¬
plicate those in the other. Of particular interest are perhaps the two
main higher-order factors, pathology and libido; these make in¬
tuitive sense, as well as emerging so clearly from the analysis:
and the relations with personality too are very much in line with
expectation. It will of course be realised that our data apply only
to a rather narrow sample of unmarried young undergraduates.

50
 STUDENT POPULATIONS

and cannot be extended much beyond these limitations; later work

with older and particularly with married groups has led to a modi¬
fication of some of the conclusions that at present appear indicated.
Nevertheless, as a pioneering study into these difficult and complex
issues our results may certainly be regarded as suggestive; only
replication with other samples will show how restrictive the limita¬
tions of the sample may have been. In spite of these limitations,
there are certain reassuring trends in our data that fit in well
with psychiatric experience; thus the combination in high N scorers
of sexual pathology and strong sexual libido may be responsible
for sexual conflict so often observed in neurotic patients.
Relating personality to sexual attitudes through factors is one
way of describing the relationships obtaining; a more direct way
is by way of subdividing the total sample into high, medium and
low scorers on each of the three personality variables, and then
calculating the percentage ‘Yes’ answers given to each question by
members of these various groups. Tables 2.17, 2.18 and 2.19 give
this information; also given there are correlations between items
and personality scale score (rp). The results are of course similar
in essence to those contained in table 2T4, but may nevertheless
repay some detailed description.
The first personality factor to be discussed is P. This presents
an interesting combination of promiscuity, premarital sex and
curiosity with hostility and lack of satisfaction; the picture is of a
‘lady-killer’ who has little love or kindness towards his victims, and
who is on the whole dissatisfied with his sex life. Individual item
correlations should be seen in the context of the pattern of factor
loadings given in table 2.14. None of the loadings of P on these sex
factors are very large (03, with premarital sex, is the highest, fol¬
lowed by 0-25 for promiscuity), but they form a meaningful pattern,
and show congruence for the two sexes.
The highest loading individual items refer to lack of concern
with virginity (items 5 and 26), liking for impersonal sex (2 and 13),
premarital sex (42 and 45), libertinism (44 and 79), liking for
pornography (47, 81 and 84), liking for prostitution instead of
marriage (85, 89), dislike of sexual censorship (91, 92, 93, and 94),
promiscuity (77), voyeurism (83) and strong sexual excitement
(52, 3, 6, 7, 33, 37, 41, 46, 53, 82). These items indicate an intense
preoccupation with sex in its biological aspect; other items indicate
the morbid and indeed pathological aspect of the high P scorer’s

51
 O' CO CO CO i-H CO wo CO O On O 04 ON ^H to ON CO
5
o © O 8 04 ~H i-H *H O 8 O P O O o *7* 04
. a
6 6 6 6 6 6 © 6 6 6 6 6 6 6 o 6 6 6 O 6 6
+ + 1 i + + + + + + + + + + + i + i + + i

00 ,
1 co hH Tf NO O' O' r- O NO Tj- NO NO CO
1 04 1-H NO 04 00 04 04 ON NO CO 04
§

_ ^ O' Os Tf CO o ON Tf O' NO O' vo in in O' NO ON NO 04 CO to

Oh 04 04 Tf O' *“H 04 O' CO 04 ON O' CO to 04

«o to> 0* 04 04 NO CO OO in NT) 04 i-H 00 CO 00 O ON OO NO NO

+ co 04 § ON 04 04 CO 1-H 00 i-H CO CO O NO NO
Table 2.17 Percentage of ‘yes’ answers for P + , P= and P— subjects for

O i-h 04 04 O O 04 04 NO O' 04 04 CO t-' ON 04

a o O o i-H ^h o O o o o o 8 iH O 8 o o O
94 items of inventory; also correlations between P and each item

6 6 6 6 © 6 6 6 6 6 6 6 6 6 6 6 6 6 o 6 o
l i i I i i 1

| Tj- 00 04 On CO CO VO ON wo NO 04 NO NO CO CO OO CO ON CO C4 On
m CO 04 NO 1-H 04 O' 04 NO 04 ^H in On 00 00

g ON NO 04 CO ON in r' VO 00 CO ON 04 O 04 00 NO NO 00 NO
co CO 04 NO 04 NO 04 in co i-H "3- ON 00 co 00
ft-

»o 04 04 o CO O NO 04 NO co in in O C4 i—H
+ CO CO O' 04 co 04 iH NO 04 O' co in ON ON »o ON

04 CO •on NO O' OO On o 04 CO tJ- in NO 00 ON o 04

C4 04 04 04 04 04 04 04 CO CO CO CO co co CO CO CO CO Tf

O' NO O O in CO On O' 04 04 NO 00 NO
04 i-H CO *“H 1-H i—< O o 04 © 04 o O O § O o o o
ik 6 6 6 6 6 6 6 6 6 6 6 © © 6 6 6 6 6 6 6 o
1 1 1 i 1 1 + l + + + ! + + 1 + 1 l 1 + +

li O' t/N in in CO NO NO CO ON n ■'t 00 1 in CO On to CO O'

NO 00 NO CO T“H CO 04 CO CO C4 1-H 04 CO

__ to> OO in wo 00 CO 00 T—H NO NO 00 NO 04 ON 00 00 O' 04

ft- NO 00 3 in 04 ■*fr 04 Tf r' CO 1-H 04 CO

Tj* CO NO CO ON in in CO NO O' in CO in [ NO 00 o
+ NO CO in 04 O' rf C4 iH 04

NO o- in CO O NO On 00 in ON o 04 tj- NO o 00 NO O' 00
k.O, o i-H © <—h o O o o © O o o o o o o o o
© 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 o 6
1 i i 1 i 1 i l 1

| 04 00 in OO CO in CO On CO ON NO in n n CO On 00 »o NO
Tf tr> 04 Tt- 1-H Tf in in 04 co NO co Tt*

g NO o> in 00 00 O' ON no CO On ON NO On O NO NO Tj- Tt 1-H co

co Tf i—i CO VO n NO CO l-H to

«o O' i-H ON 1-H in ON in 04 n NO Tf NO ON CO C4 to 00

4- co 04 04 04 hH in in NO CO NO 04 1-H 1-H to CO

1-H 04 CO vn NO O' co ON o 04 CO in NO 00 ON o
i-H i-H i™- i-H 1-H 1 04 04
00 ON ON 00
o o
NO
o
NO CO 00
o
04 CO O' CO 04 w-> O' of- Q 04 NO 04 o wo oo
O O O O *—• CO 04 04 *—< 04 o P 1—4 o 04 04 04 T—1
© 6 6 6 6 6 6 6 6 6 6 6 6 © © 6 6 6 6 6 6 6 6 6 6 6
1 + + + + 1 + + + + + + 4- + + + + + 1 + -f i + + + +

TO W“) 04 o co ON ON oo NO O' wo O' ON 00 O' 04 v-H CO CO t-H co co

04 ■•d- m wo CO NO 04 04 04 CO Tf rH 04 CO CO CO

o 00 04 r- ON o 04 00 00 CO o NO CO Tt O' 04 wo wo CO O' CO
04 Tf »n wo wo CO xf- CO CO TT 04 CO CO CO
,""1 ,"H

o o Tj- CO NO ON wo CO CO o NO O' wo CO
l CO O' 04 O' O'
04 Tf in 04 wo Tj- 04 O' wo NO 04 wo CO wo wo wo

, ,
oo W^S o NO NO NO 04 oo o 04 04 00 04 04 CO r—4 NO o’ On
O 04 o O o o 1—4 04 o o t-H o o o 04 04 o 04
6 6 6 6 6 6 6 © 6 6 6 6 6 6 © 6 6 6 6 6 6 6 © 6 6 6
1 1 1 1 i 1

co ^t NO Tf- 04 wo wo 04 co wo r- CO o CO NO ON wo NO CO co
co ^t NO O' NO 04 O' NO 04 NO Tf ^f- V~) NO CO WO O' O' wo
*“H

co 04 *Ti 00 CO NO r- 04 00 00 wo wo CO wo NO 00 CO CO O' WO 04 04 wo Tf-

m NO 04 NO wo CO O' O' rf- NO CO w-1 NO wo CO NO CO O' NO

00 04 NO W"> O wo On 00 ON 04 04 CO 04 as On ON 00 On WO CO ON ON 04 O' 04 rO
NO VN f'- CO wo CO O' 00 ■*fr NO CO NO wo wo CO O' 00 oo O'

ON o —• 04 CO wo NO O' 00 ON o 04 co Tt- wo NO O' 00 ON O 1-H 04 co Tf

NO O' O' O' O' 0' O' o* O' O' 00 00 oo oo 00 oo 00 oo 00 00 ON ON ON ON ON

00 ON w^ 00 o 04 NO O' W“> 04 00 O' Tt NO 00 O NO Tt 04 NO CO co Tt- NO ON

o 04 04 o 04 o o o o 04 ’—1 O o o o o 1^4 O i-4 o o o
6 6 o 6 o 6 o 6 6 6 6 6 6 6 6 6 6 O 6 6 6 6 6 6 6 6
+ 1 1 + + i 1 -f + + + i + -f i i + + + 1 + + + i -f -b

( »n »n in w->
ON ITS CO w^ o* 04 co CO ON CO CO O' 00 ON m 00
04 NO CO 04 04 oo 04 f' NO m r—H in

o CO O' 04 00 O' W-) O' o NO NO 04 NO Tf oo CO 04 O 04 00 m Tf in m

m r- CO 04 04 O' 04 O' i-H NO Tf 04 w-i 04 NO

o cn CO O' w-i 04 NO w-> co o •n O' 04 NO NO OO CO O NO 00 00 On On

cn rt T"H 04 CO 04 oo 04 00 t' CO 04 in 04 04 O'

,
Tf O' 00 Tf 04 co 00 04 00o o »n co w-> CO 00 co ON CO O' T“H
o o o o O o O 04 o o o o o o O o O o O o o o o
6 6 6 6 6 6 6 6 6 6 6 6 6 o 6 6 6 6 6 6 6 6 6 6 6 6
I 1 I 1 i 1 1 1 i 1 1 i

CO ON 04 o 00 as O' W-N »n 04 Os NO V"H »n •n CO •n co o oo

NO 04 04 NO 04 04 04 NO 04 04 as 04 00 m 04 CO 00

ON Tf- 04 04 NO in Tt- O o 04 ON m o o O'

r*H
04 o CO NO T^t O'
v~) 04 NO CO 04 04 O' 04 04 ON 00 T-H NO 04 CO 00

00 r-H W-l 04 00 V~i O' co ON t-H »n »n co »n CO o »n O' in 04 ON r'

NO 1 CO NO 04 CO 04 04 NO y—1 04 OJ ON CO 00 04 wn CO CO ON

CO Tf- m NO O' 00 ON o ~4 04 CO Tf* in NO O' 00 ON O *-i 04 CO Tf* in NO O' OO

^1* Tj- m «n m <n in in m in in in NO NO NO NO NO NO NO NO NO
 o- VO vo vo in 00 in 04 04 04 O' ft-H 04 00 vo co
ft.
CO o 8 8 < o s 8 © o r-H O O p r^H o r-H r-H
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
1 l I i l 1 i i 1 i i i 1

1 04 vo O' 04 00 Tf Tf o 04 04 Tf in ON o CO vo 00 Tf ON vo O'
co Tf Tf vo 04 04 1 VD 04 04 ,_H Tf Os vo 04 Tf 04

O 00
T-H
o m r-H r-H in 00 in VO 04 CO vo r- Tf vo O' 00 CO 04
tel 04 CO O' ’”H 04 00 r-M CO 04 vo Os vo CO vo 04

+
o-
r-H
04 Tf 00 o 04
r-H
Tf o CO 04 04 vo CO CO oo vo 00 vo
Tf 00 CO r-H ON CO 04 Tf Os co vo
Table 2.18 Percentage of ‘yes’ answers for E+, E= and E— subjects for

O' o vo o 00 in CO ON 00 in Os Os 04 O' CO 04 vo ON
k. r—t Ol p o o o o © o 04 r-H © O r-H o o o o o o
94 items of inventory; also correlations between E and each item

o 6 © © 6 6 6 © 6 6 6 © 6 © 6 6 6 6 6 6 ©
1 i 1 1 1 l l 1 I i 1 1 1 1

i O' Os CO o- 04 in CO 04 00 Tf r-H O' Os O' vo Tf ON CO CO oo

Tf Tf 04 vo CO 04 rH VO CO vo CO T"H vo Os 00 Tf 00

r s vo CO r-H co in vo 00 Os in vo CO Tf oo CO 04 04 vo vo vo 04 oo
tei co CO CO vo r-H 04 T-H 04 vo C4 r-H vo ON 00 Tf 00

co CO 04 in O' Tf 04 vo Tf CO
t CO CO 04 o CO o 04 vo
+ co 04 04 O' 04 r-H 00 r-H O' 04 vo ON Os Tf 00

04
Ol
CO
Ol
Tf
04
in
04
vo
04
O'
04
oo
04
ON
04
O
CO
r-<
CO
04
CO
CO
CO
Tf
CO
vo vo 00 ON o 04
CO CO CO CO CO Tf Tf Tf

vo Os O- co 00 vo 00 r-H in 04 CO Os oo
o o o o o
r-H On Tf 04 —* vo CO
*—• *7< O r-H o o s o r-H o r-H r-H r-H o o
ft. 6 6 6 6 6 © 6 6 6 6 6 6 6 6 6 6 © 6 6 6 6
i 1 1 I i i 1 1 1 1 1 1 1 1 1

Tf 00
I
1 VO
Os
CO
Tf
vo
O'
04
vo CO
CO 04
00
CO
Tf
CO
04
ft-H
00 Tf vo Tf o ON vo O'
o CO CO r_l 04 CO

P* co VO o VO CO Tf Tf O vo 04 Tf Tf Tf vo 04 CO o vo 00 o
tei vo oo in vo 04 Tf 04 Tf Tf r-H O' Tf 04 04

o O' © oo 00 Tf co vo vo 00 vo Tf vo Tf o o Tf CO
+ vo O' Tf *n 04 in 04 vo co T-H O' vo 04 04 CO

04 oo VO Os Tf CO CO Tf CO vo oo o vo
o o o o o o
ON O' O' o O' ON o
p O r-H r-H o O o T7H o o o r-H p
o 6 O 6 © 6 6 6 6 6 6 © 6 6 6 6 6 6 6 6 6
i 1 1 l 1 1 i 1 1 1 1 1 1

| vo Os 04 O' o vo 04 00 vo Tf Tf Tf Os CO vo o 04 vo 04 04
Tf Tf 04 CO 04 Tf Tf vo co CO vo CO NO ^f

r a oo
CO
Tf
vo
CO 04 CO CO Os O' in o Tf o- 04 vo r-H vo 04 On O' o
tei 04 Tf Tf vo vo 04 8 VO co T-H Tf Tf

r-H Ol O' 04 Tf CO in vo co 00 vo VO O' O' 04 04 Tf o

+ co Tf Tf Tf vo vo T-H vo vo Tf r-H
0-
Tf
00
Tf

(N CO Tf in vo o 00 ON o r-H 04 CO Tf vo NO O' OO ON o r-H

r-H r-H r-H r-H r-H r-H
- H ^ 04 04
o VO00 00 ra Os __ r- ON <N Tf o rj- OJ
cN
©
o o o O o ' o O O
6 6 6 6 6 6 6 6 6 6 6 6
O o
6
o
6
o
6
Os
o
6
ON
O
6
r^-
o
6
ss
6 6
1—4
O
o

6
vo
o
NO
O s
6 6 © ©
1 i i 1 l

o co »o r- o o ON p <N r- o vo ON NO o ON ON o Tf o cs <N VO
co co ''fr ^4 ,"”1 VO co CO i—4 <N <s xf- <N co
oo
CO
oo
CO

o
<N
»o vo
in **
cs ON 00
wo
^t
^t
i—h
r-
rj- VO o 00 Tf -vi- o <N 1—4 CO 00 <N 1“H CO
co co ^“4 co CO »o CO CO CO

vo co T;t ON © 1—4 On o o <N _

wo TH NO in ,“H r- 0- wo 1-H co
co
00
CO in r—4
VO
fN
^t NO rj* NO

Q
VO ^J* cs in »o r4 co Tf co 3: n m ON <N 00 T}- in
i—4 o o o O O © (N o O © o O o o O O o O o CN
6 6 6 6 © 6 6 © 6 © © 6 6 6 6 © 6 6 6 6 6 6 6 6 © ©
i i i i 1 1 i i i i i

r^ r- NO VO r^* 00 ON in fN o VO co ON CN rf 00 Os 00 ON 00 CO 00 r-
co Tf VO r- wo 1—4 r- NO -3* VO VO ''t CO in r-' NO

Tf ^t oo
, 00 o VO 1—4 »o (N CO 00 vo ON VO O o NO Os ON 00 CO
«o VO <N wo co CO NO co «o VO CO 1—4 in t" VO

<N Tf O co NO o in 00 VO ■<3- 00 o <N vo 00 ON CO Tf o in CO 00 fN in

»o »o <N in co in Tl- NO NO VO co r- oo 00 r-

o\ o ' CN co Tf vo NO r- 00 ON o —i (N CO »n NO 00 ON © i-H CO Tf
NO r*** r- r- r- r- r- 00 00 00 OO 00 00 00 00 00 00 ON ON ON ON Os

1-4 1-4 n in CO ON
00 ON
i^
Tf CO © CO
i—(
r-
CN
r-
©
CO
©
r-
CN
©
’—I
VO
CN
CO
<N
VO
i—i
CO 00 o
© © © © <r> © © © CS ©
© © © © © © © © © © © © © © © © © © © © © © © © © ©
1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 1

© © in ■'d- © © CO © •o ra 00 NO fN CO © NO (N in in 00 NO vo VO o
cs r- co CN <N <N vo i^ CO (N VO (N (N NO
T”<

00 © n tN vo vo 00 © <N vo CN © © ON 1—4 NO n VO <N

CN co CN CN i-H 1—4 oo <N i—4 00 VO (N ■'t 1—4 CN vo

CN co
t VO oo Os © <N ON CN CO i—4 00 NO in ON ON in CO CO NO i—4
CO in <N CN CO —H 00 i—H 00 1—4 VO CO i—4 NO 1—( <N
’"“l

co On Tf CO © i—4 <N oo ON r- CO 00 in r- i—4 n CO vo NO CO

© 1—4 1—4 © i—1 © © cs © © <s CO *—i r-H *—< © © 1—4 © © ©
© © © © © © © © © © © © © © © © © © © © © © © © © ©
1 1 1 1 I 1 1 I 1 1 1 1 1 1

CO CO © CN «o 00 © r- n in in © vo © r4 Os CO ON CO
NO CS NO r3 <N rt* CN Tt cs co CO 00 co 00 in (N 1—4 CO 00

© © Tf NO CO © 00 (N 1—4 © 00 CO VO ON (N ON 00 00 CO CO
NO <N (N VO cs CO CN CO (N <N ON fN 1—4 T—1 in <N CO Os

r- i^ CO OO NO
A r-' CN © NO © CO © <s vn Tf i—4 © 00 (N CN NO r- 1-H
i—4 CN 00 ’"V )£
On T—4
*n i“H cs vo CO On NO (N CO ON

co VO Np r- 00 ON © ^-4 <N CO *o VO 00 On © ^ <N CO -vi¬ vo NO r- 00

Tj* VO vo VO «o »o »0 »o vo »o «o vo NO vo NO vo NO vo NO vo
 SEX AND PERSONALITY

attitude. He considers himself deprived sexually (11) and dissatis¬

fied with his sex fife (4, 22), in spite of the fact that he has had
more sexual experience than the low P scorer; he feels hostility to
his sex partner (73, 80), is troubled by perverted thoughts (28, 29)
and has homosexual leanings (16, 30, 36, 40). Taking one’s pleasures
where one finds them (70) has clearly not brought him much hap¬
piness; the libertinism is marred by a pathological streak which
may justify the clinical connotations of the ‘P’ label.
The high E scorer is also characterised by the promiscuity factor,
but in him it is allied most prominently with lack of nervousness
and with satisfaction. The highest loadings are with lack of ner¬
vousness (0-35) and with promiscuity (0-27); here apparently we
have a happy philanderer, who derives satisfaction from his sexual
behaviour. The individual items having the highest loadings
emphasise the extravert’s social facility with the opposite sex (23,
31, 56, 51, 54, 58, 66, 17), his like for sexual activity (59, 69, 9, 18,
19, 32, 41, 55), his contentment with his sexual life (11, 15, 20, 22)
and his lack of worry about it (60, 63). He too is easily excited
sexually, (7, 33, 39, 43, 46) and endorses premarital sex (26, 42, 45);
he too is promiscuous (77, 44), but he lacks the pathological element
of the high P scorer (28, 35), and his liking for pornography is very
slight (84, 85, 91, 92, 93, 94). Homosexuality (36) is no problem to
him, and offers no attraction.
High N scorers show a different combination of excitement and
approval for premarital sex with the other factors; they are charac¬
terised by low satisfaction and high guilt feelings. Loadings are
highest on guilt (0-30) and lack of satisfaction (0-25); excitement
loads more highly for the men (0-20). Individual items emphasise
the same features; particularly prominent are the lack of satis¬
faction derived from sex (4, 20, 22), the guilt feelings associated
with a strong conscience (48, 25), the worry about sexual activities
(60, 63), the problem of controlling sex thoughts (35, 7, 28, 29) and
the fears and difficulties with contacts with the opposite sex (56,
17, 54, 56, 31). Blame is attached to the inhibiting influence of the
parents (34), religion (49) and ‘bad experiences’ (27). Sexual be¬
haviour is seen as both troublesome (21, 19, 66) and disgusting (11),
and the high N scorer stresses his inability to contact members of
the other sex (15, 23); in spite of all this he has strong sexual drives
(33, 41, 43, 50, 52) which he finds it difficult to control (32, 53).
Homosexuality is a problem (16, 36, 40). There is some evidence

56
 m i ON W~> ON cn f" 00 V~i N n >n 00 r- »n Tf m T—< m m
<N 1 P P ’7H O cn ~ O O O -H O o o O O o o
o © o o o o o o o o o o 6 666666
1 1 i i

r- »o NO ON Tf (S no V“l 00 »ZN On ON Tf cn On Tj- IZN fN

oo CN Tt on CN Tf CN CN NO CN

;7

2
00 NO »/“> o

4
r- hh cn (N
cn cn 't ON cn cn CN NO (N

Tf ON cn 00 On CN NO on NO *Q Tf cn 00 Tf m 00
+ t-H cn cn cn 1""1
o
Tf ON NO Tf >ZN cn (N 1-H Tf

r- CN NO Tf 00 cn rf Tf H00 00 »zn tT 00 r-
♦3 c CN hH T“H cn T-. o o o i—>i T—1 o T-H o o O O <N
o c
vi v>
666666 6 6 6 6 6666666
o o
I
IS1
i 00
2
4

4
2

5
o NO CN T—i NO Tf Tf r-
I § HH r-
,"H NO T™1 *0 ON 00 Tf IZN (N *n

Z ^3
•g I Tf en cn Tf cn hh ra ON VN 00
(
•AN ON
( r-
T-H r* CN NO cn rf OO oo ot 00
NO cs (N NO
5Z
II s + 00 Ol rf
no xo ts
m *ri no 0\ rf
CN NO cn r-* cn CN
*n *ri o
»n On On Tf ON
00 »ZN ON 00 »ZN O
NO (N NO m

- *
+ vj
z-°
s- C
•2.1 O
m m
cn m Tf *0 NO r- oo on ©
cn cn cn cn cn cn cn cn Tf Tf
CN cn ■^t »n NO
Tf Tt
oo

5 -2
V Vi
* fc
a o
§ °
> § r<N IZN NO o >o r- r4 O Tf cn 00
^"3 O O p CO o T-H O O ^ (N p p o t-H O (N o »—H fS <s ss
o © o o o o oooooooo 6 6 6 6 6 6 6 6 6 6
1 1 1 1 1 ! 1 i
Vi>
V, 2
Oo c NO ON ON r- CN NO <N o r^- -h <n <N th ON m Hi fN oo o m rH o rs no
a vj vr> r- m CN (N Tt cn t—i r- »T) T-H '■t CN
a >
•*- h*
^ .z:
m <S rf- m OO NTl
20

44

39

4
9

r- cn m NO »o ON NO NO oo r-
V o Z NO 00 Tf (N y~~' m r-H r- ■<* (N T-H (N cn <N CN
to
E
+ O On
no r-
ON O NO m VN 00 M ri 00 t-H t-H 00 Tf »o *n
’’T (N M Nf Tf H ^H ^
>ZN O Ti¬ r- On
<S CN CN cn ro en CN
<N ~
"H-
M O. ,
o r- M^^hON^Tj-Nh IO T-H NO oo T-H 00 On CN On
3aj
^ ° O hH ?N O O O <N o o o T“H hH o Hi cn o fS hH O

o o o 6 o o o o © o © 6 6 6 6 6 6 6 6 6 6 6 6 6
H i 1 i i l

NO NO NO m m no Tf ^ oo no <n —■ \o r- > <N r- 00 On NO O i-i

m rf U~) m Nt Ti H rf \© m fS cn

OO TH fS
6

44

NO o ON rt r-" W-i CN NO NO o r- t^» NO

61

z m «ri T-H
0\
tj- *0 NO (S cn T-H T-H cn cn
hh

+ m o 00 IT)
IO (N CS
m r-
*T)
oo on t*
<n no n
vn «n Tt OO NO 00
^ NO <s
Tf
T-H T-H
o ON m CN rf
cn »o Tf

*—< (N »o NO r- 00 ON © 1-1 <S m rf vi vo 00 On o rH cs cn Tt

rH i-H T-H 1H T-H T-H Hi T-H T-H T-H <N (N CN CN CN
 i^NmNviN'otsooi£;g\r'«NO I O O O O O *7* *7"
>t -I M N O O
o o O O O ^—< o O O O 'r~< ^
V. 66600000066000606666060
I I I

1 fninO'oONhoo^cocO' ONONOsm'Orsravof^ri-^t
M N M n fntn
in th >n Tf ih rs m

Si o\vio\'n^'ON«aiO\t-o\-'00^'03;N'ot'^oo^
xj- ,—1 ir, Tf — cn cn m (N nf *~* CN m m m

OMOcy\r-(N'^rtS'; mN«'O'O'H»0N co N m in x-0\ cn

+ vo *— in n « in rx rn m Tt >n C*l Tj- Tj-

moooi'oi'iMHiOrtO'ci't"-:CJ"C3!C!S!i32iS!Q
O.—'OOOOOOfS^'—1 ’—1000*7*0060000
X. 66666666666666666666060
I I
r'in'tfnino\'OVD io h n m On N moovnovr-'n'^;
I vo —i r~ vn t~- t~~ N in n n vo io in m *—1 vn t— *0

a (Ortt'-rnmN** vo t~~ vo «—1 vo —< Ov vtNMVfNOvVOOV

vo cs i— vo cn r— n vo ^ ^ in vo vj- cn vo m

Tt in Tft'^fin'n'HOinr'Ov mro^movOoomr^*—1
+ r- <n hinnhMnvovfin vo >0 — cn *-ivooor^vo
Table 2.19—cont.

Nnvt'n'or'ooovOrHMn Tfin'Or'OOOvOHNn’t
t'l't'r-M'f't'XM*® OOOOOOOOOOOO Ov Ov Q\ CT\ ON

m'tiHcoov't'vf'-i in-1 (S m ov 5 m fNOV>vo>/~)Tt->nnfri

0(S000006
X. OrtMO-i <—iO*-i >—1 *—i O cn O O cn
66666666 6666666 66666666

O(s)voor~r'(smmoofsm^ovfso>oomor~0
(S OO HMOO VO-*t V) I CvIVOCN Tf

«i'niniHinon'OninhHVOninnioo\OOOv>N
.—I C-4 ■•—I ~ 00 N « f.H 10 Mln*-I (S VO rl rj-

+ vOTtnmnNvtoovtMoooovDvt'OvmovOOMt'OOin
1-* m cs cSf' mNr'tsvo'tminN-ifnr'N ^

itoooivvo-voo(v|inmfxNtvco'vtvoo\oo«OooinN
«MnMH(v|^iS7NO,J99'ri9T"r9T'99?
66666666666666666666666

I nvotvN'or'OOiOt'vff'r'n'ONr'Vonnmnov
cv)*-* -*(x ts >-* on oo-* vn ^ cn a\ it

O\O'ONvt00rtt'cnNin'O'vt'tnrtavt^''fO\O'tNfn
m*-* N vo NNo\**oort«in(v| m on -vj-vn

+ OvN«'tnn'tO>nNoo'tr*OvOO'Ci'00|nvo^n
Tt *3- t—1 cn 10 N n n 00 vj » ro vo co *—1 m 00 rj- >r>

Ov O -i<Nn->Tfriovot'~oocr\Q'-*<Nr,inl-iovor~'000\0'-i
vfininin'nininininininiovo'ovo'o'ovovovovot't'
 STUDENT POPULATIONS

of liking for pornography (83, 85, 93, 94), but much less so than
in the high P scorer; it almost seems a substitute for the unattain¬
able sexual contacts with real life partners. Lastly, there is a
tendency to be hostile to the sex partner (80, 73), but again in the
context suggests a different interpretation to the hostility of the
high P scorer; here the hostility may spring from the failure to
i acquire a sex partner in the first place!
Taking an overall view, one might say that, as expected, high P
and N scorers show a distinctly pathological pattern of sexual re¬
actions. Both are characterised by strong sexual drives (the former
less so than the latter), but whereas the high P scorer ‘acts out’
his libidinous, promiscuous and perverse desires, the high N scorer
does not; instead he is beset by a whole set of inhibitions, worries
and guilt feelings which effectively prevent him from consummating
his desires. Yet both groups are dissatisfied with their patterns of
sexual performance, although presumably for different reasons; this
dissatisfaction constitutes the strongest evidence for the hypothesis
that both are to some degree ‘pathological’. (It would clearly not
be adequate to justify this term on the grounds of either statistical
infrequency of occurrence, or of moral and ethical undesirability of
the conduct in question; it is because both groups are so dissatisfied
with their behaviour that one may justly infer that it is not appro¬
priate.) Both groups are similar in that they view their sex partners
with some hostility, like pornography, and have homosexual lean¬
ings; yet as already pointed out, the different setting in which these
items occur suggests different interpretations of the motivation in¬
volved, at least for the first two points.
As regards the E factor, the evidence would seem to suggest that
here we have non-pathological ways of sexual adjustment, the
extraverted and the introverted, which are opposed in a very mean¬
ingful manner. The extravert endorses the ‘permissive’, promiscuous
approach to sex, with frequent change of sex partner and much
‘healthy appetite’ for frequent sexual contacts. The introvert en¬
dorses the orthodox Christian approach with fidelity, stress on
virginity and less purely biological factors as the prime contents.
Taken to their extremes, these approaches become the ‘libertine’
and the ‘puritan’ respectively, but if not taken to excess they are
probably both viable modes of adjustment. The extravert seems
more satisfied with his way of life, and is of course better able to
contact members of the opposite sex, but this may be an artefact

59
 SEX AND PERSONALITY

of the particular sample taken; at twenty, unmarried youngsters

quite naturally have some difficulties in living up to introverted
ideals. At forty, the happily married introvert may show better
adjustment than the extravert suffering from the ‘seven-year itch’.
This is of course merely speculation, but it may serve to emphasise
the restrictions imposed on interpretation by the specific nature of
the sample studied.
The data presented in tables 2.17, 2.18 and 2.19 enable us to say
something about the consistency of the personality-attitude relations
between sexes, and also about the similarity or dissimilarity of
attitudes held by different personality types. Given in these tables
are six columns (rp, rE and rN, each replicated for males and
females) which report the correlations of each of the ninety-four
items with P, E and N. These six columns were themselves cor¬
related, in the hope that the results would throw some light on the
two problems mentioned above. First consider the male—female
correlations within the personality type, i.e. PM v. PF=0-69;
Ek v. E¥=0-80\ Nmv. Ap=0'77. These demonstrate that personality
scale-attitude item correlations that are high for one sex are also
high for the other; there is clearly a considerable amount of con¬
sistency here, particularly for the E scale, slightly less so for the N
scale, and least of all for the P scale. This is not unexpected, as
the P scale is the least reliable and has had much less experimental
work associated with it than the other two scales. When we turn from
these intra-scale correlations to inter-scale correlations, we find
results which may be set out in the form of table 2.20.

Table 2.20 Correlations between columns of tables 2.17, 2.18 and 2.19

PxE PxN ExN

Male v. male 0-28 0-23 -0-61

Female v. female 0-39 0-38 -0-25
Male v. female 0-22 0-33 -0-39
Female v. male 0-32 0-14 -0-44

Clearly, the sexual attitudes of high P scorers are a little like

those of high E scorers, and also a little like those of high N
scorers, the degree of similarity does not amount to more than
about 8 per cent or 9 per cent of the variance. High E scorers are

60
 STUDENT POPULATIONS

somewhat more markedly unlike high N scorers; the degree of dis¬

similarity amounts to something like 17 per cent of the variance.
The within-sex comparisons are no different on the whole from
the between-sex comparisons, and all are in good agreement with
each other. The between-scale correlations are clearly lower than
the within-scale correlations, demonstrating that our results are
consistent across sex. On the whole these figures are very en¬
couraging; they suggest that different personality types do indeed
have different attitudes towards sex, regardless of the sex of the
respondent.
Having thus briefly discussed the sexual attitudes associated with
P, E and N, it may be worthwhile to devote a few sentences to
a discussion of the observed differences between male and female
attitudes. The items showing the greatest differences are given in
table 2.21 (Eysenck 1971b).
Overwhelmingly outstanding among items giving marked differ¬
ences between the sexes are items relating to pornography (47, 81,
84, 91, 92, 93), orgies (44, 94), voyeurism (83, 62) and prostitution
(85), closely followed by impersonal sex (2, 13). Sexual excitement
is close behind (33, 41, 43, 46, 82, 3, 39); in all those of course males
have higher rates of endorsement than females. Premarital sex is
also favoured more by the males (45, 70, 79, 42), as is promiscuity
(77). But contentment in their sex life is more marked among
women (4, 20, 11, 22), perhaps unexpectedly. Masturbation is more
a male pastime (10, 8), and men are also less prudish in general
(18, 68, 69, 59) and feel less guilt (25).
Most of these differences are not unexpected, although one should
not over-interpret them; some of the replies may represent little
but widely held views unthinkingly endorsed. The only unexpected
feature of the study is the apparent satisfaction of the women with
their sex lives; it used to be thought that the ‘permissive’ society
favoured men, as did the Victorian era. Possibly the clue lies
in the greater sex drive apparent in the men, and the difficulties
that this strong drive must give rise to when confronted with the
stark reality that over half the women in our samples were still
virgins and apparently intent on holding on to this status. In this
sellers’ market, women clearly have the upper hand, and may enjoy
this status; again the nature of our sample may be responsible for
a finding that is not likely to be duplicated for older men and
women. There is an interesting finding in Schofield’s book (1968), in

61
 SEX AND PERSONALITY

Table 2.21 Percentage differences in endorsements of twenty-seven items

on the sex inventory between men and women

M-F differences

If you were invited to take part in an orgy, would you: 57

(a) accept it? (b) refuse?
I like to look at pictures of nudes 53
I like to look at sexy pictures 53
The thought of a sex orgy is disgusting to me -47
If you were invited to see a ‘blue’ film, would you: 43
(a) accept? (b) refuse?
It is all right to seduce a person who is old enough to know 38
what they are doing
I believe in taking my pleasures where I find them 37
If you were offered a highly pornographic book, would 36
you: (a) accept it? (b) refuse it?
It doesn’t take much to get me excited sexually 35
Buttocks excite me 34
I get excited sexually very easily 33
I think about sex almost every day 32
Seeing a person nude does not interest me - 32
Sex without love (‘impersonal sex’) is highly _ 31
unsatisfactory
I do not need to respect a woman, or love her, in order 31
to enjoy petting and/or intercourse with her
Prostitution should be legally permitted — 30
If you had a chance to see people making love, without 29
being seen, would you take it?
Pornographic writings should be freely allowed to be 27
published
It is better not to have sexual relations until you are -25
married
Sometimes the woman should be sexually aggressive 24
I get pleasant feelings from touching my sexual parts 24
Something is lacking in my sex life 24
There should be no censorship, on sexual grounds, of 34
plays and films
Conditions have to be just right to get me excited sexually — 22
I find the thought of a coloured sex partner particularly 21
exciting
The opposite sex will respect you more if you are not too - 21
familiar with them
All in all I am satisfied with my sex life _20

62
 STUDENT POPULATIONS

which he showed that female adolescents who had had intercourse

were not very attractive on the whole, while male adolescents who
had had intercourse were; the explanation presumably is again in
terms of the sellers’ market - men must be attractive to get a girl,
but a girl who is attractive does not need to trade her virginity
for male attention. Specific research devoted to a clarification of
these relations might be of considerable interest. Figure 2.1 shows
the distribution of scores on a scale made up of the most dis¬
criminating male-female items.

Number of Subjects

Figure 2.1 Scores of men and women on masculinity-femininity in sex

attitudes inventory

It is of some interest to note that male-female differences on

our sex attitude inventory are similar to the high P v. low P differ¬
ences, as shown in table 2.17. In other words, the pattern of male
v. female response differences is similar to that of high-P males
v. low-P males, or of high-P females v. low-P females. This is
not unexpected; in our previous work with P we have noted that
males have much higher P scores than females, and that criminals
(who are of course predominantly male in society) have higher
P scores than do non-criminals. This suggestion deserves investiga¬
tion and the following calculations have been carried out accord¬
ingly.
For the purpose of analysis, the percentage of ‘Yes’ answers on

63
 SEX AND PERSONALITY

each question was established for men and women separately, and
that for women subtracted from that for men; this difference score
will be called the M-F score. Next, product-moment correlations
were established, separately for the two sexes, between each per¬
sonality dimension (P, E and N) and between each of the ninety-
four items in the inventory. There are thus six sets of correlations:
three dimensions, two sexes. These six sets of correlations were in
turn correlated with the M-F score, our expectation being that
M-F score would correlate positively and highly with the P factor
correlations, for both men and women. In view of the fact that
masculinity-feminity has been observed to load positively on ex¬
traversion it also seemed likely that positive correlations would be
found between the M-F score and E. No predictions were made for
N. Table 2.22 gives the actual intercorrelations between our seven
variables. Interest centres on the last column, which gives the cor¬
relations of the M-F score and the sex personality-sex inventory
correlation scores.

Table 2.22 Correlations between M—F score differences and sets of

personality inventory score-sex inventory item correlations. Asterisked
values significant at < 0 01, doubly asterisked values significant at < 0001
levels of p

3 4 5 6 7
P M 2 0-70** 0-31* 0-23 0-23 0-33* 0-54**
F 2 0-34** 0-38** 0T4 0-38** 0-74**
E M3 081** -0-56** -0-34** 0-31*
F 4 -0-45** -0-24* 0-30*
N M 5
0-77** 0-10
F 6
0-21*
M-F 7

As expected, the correlations of the M—F score are very high

and positive with P, for both sexes. They are considerably lower,
although still significant at the p < 0-01 level, for E. For N, the
correlations are for all practical purposes negligible, although both
are positive, and that for females is significant at the p < 0-05 level.
It is clear that our hypothesis receives strong support from these
data; male attitudes towards sex, as compared with females ones,
are similar to those of high P scoring males (or females) as com¬
pared to low P scoring males (or females). These correlations, in
fact, are to the same order as the correlations between males and
females on the psychoticism scale.

64
 STUDENT POPULATIONS

One further item of interest may support our view. Eysenck

(1971c) found that high P scorers, as compared with low P scorers,
had indulged significantly more frequently in minor perversions
(fellatio, cunnilingus, ‘69’). So had men as compared with women,
when only respondents having had intercourse were included in the
tabulation. However, taking part in an activity is not identical with
enjoying it, and it seems likely on our hypothesis that of those who
had actually indulged in these perversions, men would have enjoyed
them more than women. Some evidence is furnished on this point
in a study by Kolaszynska-Carr (unpublished) in which male and
female subjects who had indulged in these activities (between 40
and 50 of each sex, depending on the activity in question) reported
on their enjoyment of their experiences. Taking fellatio as typical of
this category’ of conduct, 39 out of 41 male subjects reported en¬
joyment, but only 24 out of 42 females. Even for such items as
manual genital manipulation, male by female, the percentages are
quite different (94 and 64 per cent); for mutual manual genital
manipulation they are 96 and 68 per cent. It would seem that for
even slightly unusual practices women are much less likely to
report enjoyment; no such sex differences were found for petting
behaviour of a more innocent kind (fondling of the female breast,
kissing of nipples, etc.). This whole area of enjoyment of sexual
activity deserves to be studied in much greater detail.
It is of some interest to note that in some unpublished research
imprisoned sex offenders scored very significantly higher on the
P scale than did other prisoners (who in turn of course scored very
significantly higher on the P scale than did matched non-prison
controls). The mean P value for the sex offenders was 1T07, as
compared with a prisoner mean of 6-25; p <0 01. The offences in
question included rape, indecent assault and buggery.
Sexuality thus seems to constitute a continuum that is largely
colinear with P and with masculinity; it would seem important to
investigate possible hormonal and other biochemical correlates of P
in suitably chosen samples. But even without such obvious biological
assays, it may be said that our data furnish important support for
the view that the P scale appears to be closely related to the mas¬
culinity-feminity dimension.(4)
Two questions in the inventory are related to sexual reactions
that might be considered medically pathological, although use
of this term is of course somewhat arbitrary in this context. The

65
 SEX AND PERSONALITY

questions relating to male subjects are numbers 95 and 96; they are
concerned with impotence and ejaculatio praecox respectively. For
the women, these two questions refer instead to frigidity (from (a)=
never to (f)=always) and orgasm during intercourse (from (a)
=very often to (f)=never). The actual wording of the possible
answers ((a) to (f)) is identical to that of the male questions. The
wording of the female questions is: ‘Have you ever suffered from
frigidity?’ and ‘Do you usually have orgasm during intercourse?’
These questions are only meaningful for respondents who have in
fact had intercourse, and were only answered by them; in con¬
sequence they could not be included in the factor analysis, and
results are discussed separately in this section.
The distributions of replies, as expected, are very asymmetrical,
and in order to make possible the use of t tests an attempt was made
to divide the distribution at a point that would give as nearly
as possible groups of equal size; this aim was not accomplished
with any very great success, owing to the piling up of data in certain
categories. Nevertheless, the results are suitable for statistical treat¬
ment. The male results will be discussed first, followed by the
female results. In each case, the P, E and N scores of the groups
that showed or did not show the pathological behaviour in question
were calculated and compared, significance of differences being
assessed by means of the t technique.

1 ^ Male, impotence. The great majority of men gave answer (a),

i.e. never to this question (n= 164); consequently all other answers
were grouped together to form the ‘pathological’ group (n=120).
Mean scores on P, E and N are shown in table 2.23; it will be seen
that impotent men are somewhat (non-significantly) higher on P,
more introverted and significantly (p <0 05) higher on N.

P E N
1 Male Impotence Non-pathological 4-37 1309 10- 58
Pathological 4-82 12- 65 11- 84
2 Ejaculatio Non-pathological 4-62 13- 04 10- 54
praecox Pathological 4-48 12-70 11- 70
3 Frigidity Non-pathological 3 00 12-59 12- 05
Pathological 2-80 11-58 13- 41
4 Orgasm Non-pathological 3-06 11-87 12- 29
Pathological 2-77 11-84 13- 75

66
 STUDENT POPULATIONS

2 Ejaculatio praecox. A majority of men gave answers (e) and

(f), i.e. never or hardly ever (n—152); consequently all other
answers were grouped together to form the ‘pathological’ group
(ft = 132). Mean scores on P, E and N are shown in table 2.23; it
will be seen that men suffering somewhat from ejaculatio praecox
are slightly (non-significantly) lower on P, slightly more introverted
and significantly higher on N.
3 Female frigidity. The great majority of women gave answer
(b), i.e. once or twice; this was grouped with answers (c) to (f) to
constitute the ‘pathological’ group, with those answering ‘never’
(a) constituting the non-pathological group. Mean scores on the
personality dimensions are given in table 2.23; frigid women (using
this term somewhat inaccurately for our ‘pathological’ group) are
somewhat more introverted, but not significantly so, and score
higher on N, but also not significantly so. Numbers are only 49 in
the non-pathological group and 122 in the pathological group; had
the numbers been as large as those in the male groups, these
differences might have reached significance. Clearly repetition of the
study with larger numbers is called for.
4 Orgasm. Many women gave answer (a), i.e. ‘very often’ or (b),
i.e. ‘often’; these were combined to form the non-pathological
group (ft=83). The other answers were combined to form the
‘pathological’ group (ft=86). Neither P nor E seem to be related
to orgasm frequency; N, however, differentiates the two groups at
the 0 05 level of statistical significance: higher N scores go with
lower orgasm frequency.

The results of this analysis are not unexpected; it is found

that sexual pathology as defined here is associated with neuroticism
(significantly in three cases out of four, and almost significantly
in the fourth case). Introverts show slightly greater pathology, but
these differences never reach significance. High P scorers do not
differ significantly from low P scorers, and may in fact have slightly
less pathology as regards these indices of behaviour.
It is doubtful if the behaviours called ‘pathological’ really deserve
this name, in view of the frequency with which they occur in this
normal group, and it seemed of some interest to study the per¬
sonality correlates of the much smaller more extreme groups giving
more definitely pathological reactions. Five males admitted to having
suffered from impotence often, more often than not or always; they

67
 SEX AND PERSONALITY

showed a markedly elevated P score of 7-00, which is significantly

higher than average. The E score of this group fell to 1T9, and
the N score rose to 12-8; these changes are in line with expectation,
but not significant in view of the very small size of the sample. Six
women admitted to frigidity often, more often than not or always.
Their P scores went up to 4-58, and their N score reached the very
high value of 17-83; the latter value is significant beyond the 1 per
cent level, but the former is not significantly different from average.
The other extreme groups do not add anything of interest to
the data already presented. It is interesting that, in spite of the
small size of the sample of frigid women, the greater degree of
pathology involved has now made the relation with N significant.
We may conclude, therefore, that all four types of sexual pathology
are related to N, but that P is involved significantly only with
high frequency of impotence.
Discussion of the results will be confined to two main points:
4 #

1 the problems of sampling and

2 the problem of veridical report.

The correlations here established between sexual attitudes and

personality variables are meaningful only in so far as they can be
considered to transcend the particular sample on which they were
established. Correlations are not as subject to sampling distortions
as are population parameters such as means, but nevertheless some
evidence is required to show that our sample is not so highly selected
with respect to relevant variables as to make the conclusions of
doubtful generality. Eysenck (1972) has shown that the sample is
very similar to unselected population samples of similar age with
respect to percentage of men and women with experience of coitus,
and also with respect to the scores on P, E and N. In other words,
our sample is representative of the population of unmarried adoles¬
cents of eighteen to twenty-two years of age with respect to the
two main variables we are concerned with, i.e. sexual experience
and personality; it seems unlikely that our data are entirely idio¬
syncratic and unrepresentative. No doubt some distortion of
sampling has taken place through the act of volunteering and other
associated factors, but these are probably not so serious as to
invalidate the results.
As regards veridical reports, we have two lines of argument.

68
 STUDENT POPULATIONS

The first relates to internal evidence; thus, duplicated items gave

very highly correlated results, which suggests that items were not
filled in randomly or with intention to deceive. The large number
of comments written on the questionnaire returned suggested that
respondents took the task very seriously. Most important, meaning¬
ful factors are not likely to arise from an analysis of correlations
between items that were not in fact completed with some degree of
honesty. Furthermore, the higher correlations between sexual
behaviour patterns and personality in males to be discussed presently
are unlikely to have arisen from faked data.
More convincing perhaps are various bits of external evidence.
If the relations established in this paper are real, then it should
be possible to find evidence in the literature of factual consequences
of these relations. Eysenck (1971a) has argued, for instance, that
if extraverts are in fact more promiscuous, then V.D. patients and
unmarried mothers should be particularly extraverted; Eysenck
(1961) and Wells (1969) have found evidence in favour of these
predictions. Sex differences in line with our results have been dis¬
covered in an experimental investigation by Sigurt and others (1970).
Psychiatrists have repeatedly found a relation between neurosis
and sexual pathology; our data are very much in line with these
suggestions. Ultimately, of course, there can be no absolute proof for
the veridical nature of the answers given, but such evidence as has
been quoted makes it unlikely that the data seriously misrepresent
the truth. After all, respondents were assured of anonymity and had
no motivation to tell lies; furthermore, much concentrated work was
required to fill in the various questionnaires properly and post them
back to the author, and few people would be likely to undertake all
this just in order to mislead.
It might also be pointed out that other writers, using different
methods, have reported results that, where comparable, were similar
to ours. Mention has been made of the work of Giese and Schmidt;
we might also mention the interviewing studies of Schofield (1968),
and of Bynner (1969). There is thus beginning to build up a set a
of findings linking personality factors with sexual attitudes and
behaviours that seems to hang together and be reproducible from
study to study, even when different methods of information¬
gathering and different samples, of different nationality, are involved.
Finally, it should be noted that these results are for the most part
in excellent agreement with prediction from theory; respondents

69-
 SEX AND PERSONALITY

could hardly have known these theories, or filled in their inventories

in such a way as to support prediction!
We must next turn to the results of our sexual behaviour question¬
naire. This is reproduced in table 2.24.(5)

Table 2.24 Sexual behaviour inventory

Here are brief descriptions of sexual behaviour patterns which people

indulge in. Indicate by putting a cross (X) in Column 1 whether you have
ever indulged in this type of behaviour.

(Note: Manual = by hand /Oral = by mouth) (column 1)

Have done

1 One-minute continuous lip kissing

2 Manual manipulation of male genitals, over clothes, by female
3 Kissing nipples of female breasts
4 Oral manipulation of female genitals
5 Sexual intercourse, face to face
6 Manual manipulation of female breasts, over clothes
7 Oral manipulation of male genitals, by female
8 Manual manipulation of male genitals to ejaculation, by female
9 Manual manipulation of female breasts, under clothes
10 Manual manipulation of male genitals, under clothes, by female
11 Sexual intercourse, man behind woman
12 Manual manipulation of female genitals, over clothes
13 Manual manipulation of female genitals to massive secretions
14 Mutual oral manipulation of genitals to mutual orgasm
15 Manual manipulation of female genitals, under clothes
16 Mutual manual manipulation of genitals
17 Oral manipulation of male genitals to ejaculation, by female
18 Mutual manual manipulation of genitals to mutual orgasm
19 Mutual oral genital manipulation

Details of this study are given in Eysenck (1972), where a factor

analysis of the intercorrelations between the individual items is
reported. Factor 1 has high loadings on items relating to petting
(kissing, manipulation of female breasts over or under clothes,
kissing nipples of female breast). Factor 2 has high loadings on
items related to intercourse and manual manipulation of sexual
organs, such as would unavoidably be part of intercourse. Factor
3 has high loadings on items related to oral manipulation of partner’s
sexual organs (fellatio and cunnilingus). For the sake of convenience,
we may perhaps speak of these factors in terms of petting, inter¬
course and perversion. Factor loadings, for males and females
separately, are given in table 2.25.

70
terms of normal curve
Percentage scores in
/—s /—■s /-N /—V
ON /^-^s 00
,—s 0 r- wo NO ^^ CO y—K ^—s
t-H t—< 00 NO Tf ONv—1 r-H wo CO
>- > ^ -/ j .Z S- s w
te —4
t r- is' 0}- 'vf CO CO ON NO CO O O T"H (N
rf co (N 00 NO Tf 4 1 Wo NO CO p ^r 00

deviate O 0 6 0 000 O 0 6 0 O 6 0 6 6
1 . 1 1 11 1 1 1 1
z-s Z-s z-V z—s
ON CO NO r-* 00 /—s z»"s r- t-H Z-V /^S wo /-s
< NO 1 H wo OO r-H 1—( y—i o\ 1—H ■*—1 CO r- r4
'wz' N^/ ’w’ |-Z >- . J
S 0 NO »—< OO oj- wo CO (N On 00 CN On 00 T“H
t—H wo
NO wo OO co 0 (N OO NO CO Xj- Tt- q\ CN ON
6 6 6 0 0 0 0 6 6 6 O 0 0 O <6
1 1 1 1 1 1 1 1 1 1 1
00 NO (N wo r- wo CO 0 wo wo ON tj- NO wo
Factor loadings (F)

0 O O NO co O r- 0 0 -H NO 0 ’-1 1 O O 00 ’—i OO
0 O O O O O 0 0 0 0 0 O 6 6 O O 6 6 6
1 1 1 . 1 1 1 1 l 1

(
<N *0
wo 00 00 00 On 00 wo 0 00 r- T“H
O OO *—< 0 <N ▼—< 00 O OO '-H p <N ’—i <N On 0 0
<N
O O O O 6 O 0 0 0 6 6 6 6 6 6 0 0 6 6
1 i 1 1 1 1

l 1
ON 00 (N WO (N Tf On wo wo wo NO OO wo
NO 0 00 O O O O On 0 00 t-H r- O 0 (N 0
6 0 6 6 O 6
O O 0 0 0 6 6 6 6 0 6 0
i 1 i l i 1
Factor loadings (M)

— c4 00 r- Q <N
co 00 CN Tf 0 NO (N 0 ON
m O NO 0 O 0 NO 0 0 r- »—1 0 OO
to O 0 0
6 6 6 6 6 6
00600 0 0 0 0 0 6 6
O

1 1 1 1 1 1
On wo 00 00 wo On NO r- (N r- (N 0 <N (N NO On O
00 tj- 0 co O <N OO 0 00 0 Tf 00 O r^- 1
6 0 0 O O 0 O 60 0 0 0 6 6 6 6 6 6 O

1 11 1 1

wo r- r- r- 00 wo 0 CO 00 r- 0 CO
On O 0 cn ON 0 (N CO 0 tj- 0 0 CN 0 0 CO 0
O 6 0 0 O 6
O O 0 0 6 O 0 0 6 6 6 6 6
l 1 1 1 1
CN 00 O O' NO C/C OO O ri NO Tf ON <N
»—H CN »—1 (N <N O (N <N »—< CO (N t—H i—t *—1 CO 1 <N
O 0 6 O 6 6O O 0 6 6 0 O O 6 6 6 6 6

co 0 0 r**- •—H fN (N NO ON 0 wo 00 NO w^»

p *—1 <N —• O O (N p t-H i-H 0 »—< 1
O 0 O 0 0 O O O 0 6 6
O O 0 O 0 6 6 6
1 1

00 rf <N Tf cn r- CO 0 r4 Tf CO wo CN
O 0 O O 0 O O 0 O O 0 p O O 0 p 0 O O
O O 06 6 0 O O 0 O 60 0 6 6 6 6 6 6
11 1 1 1 1 1 1 1

00 wo co r*- co Tf 0 (N O tT NO On <N 00 CO CO
0 *—* O 0 0 T“H 0
Significant at 1 per cent level: 013

O <n '“i O
Significant at 5 per cent level: 010

O i-H O 0
O 0 O 6 O O O 6 60 6 6 <6 6 6 0 0 6 6
1 l 1 1 1 l1 I l 1 1 1 1 l l 1 1 l l

NO wo (N CO n r^* wo 0 00 00 0
O 00 00
<N r—4 t-H —« *—< 1—1 (N Ol 0 T—< cs r-H r-H

O O 6 6 O 6 6 6 6 6 6 6 O 6 6 0 6 6 0

co CO On wo 0 <N 00 (N 00 ON 00 ■'t wo CO NO CO
>—1 CO r—1 <N t—i t-H CN rJ ’—1 »—H
fN (N t-H <N CO
% O 6 O O O 6 6
0 O <6 6 6 6 O O O O 6 O

00 O O wo NO ON 00 NO 00 CO OO (N ti- NO (N
p ’—1 <N O *—1 0 1 0 T-H O t-H 0 *—1
ten O 6 6 O 6 6 6 6 6 6 6 0 6 6 6 6 6 6 6
ft, 0 , <N (N p Tf wo S 00 On CO CO CO (N CO NO
O 0 0 O O1 0 0 0 0 0 O 0 0 0 0 O
S O 0 6 6 6 O 6 0 0 0 0 6 6 6 6 6 6 6 6
1

, n co Tf wo NO r- 00 ON 0
, CO Tf wo NO 00 On
T—1 T—1 1—1 »—1*—1
 SEX AND PERSONALITY

The first 8 columns in table 2.25 give the correlations, for males
and females separately, of P, E, N and age with the nineteen items
of the Sex Behaviour Questionnaire. Correlations of 0T0 are statis¬
tically significant at the 5 per cent level, correlations of 0T3 at the
1 per cent level; only two-tailed tests are used in this comparison.
Before discussing the three personality variables, it may be worth¬
while to look at age as a factor; this has not been partialled out as
it does not correlate with P, E or N within the narrow range of
ages here sampled. It does however correlate with sexual experience;
as one might have expected, all the correlations (with two trivial
exceptions) are positive, and the tendency is clearly for females
to have rather higher values than males. The highest values, for both
sexes, are with items loading on factors 2 and 3; correlations of age
with factor 1 items are uniformly low, or even negative. Petting, in
these groups, is probably so universal, even at seventeen to eighteen
years, that there is little increase; it is the more serious items in
factors 2 and 3 that show such an increase. Even so, the values
are perhaps somewhat lower than one might have expected; age
never seems to account for more than 10 per cent of the variance.
Correlations with P are almost uniformly positive; high P scorers
have more experience in all forms of sexual conduct. Figures are
pretty uniformly higher for females than for males; the most notable
difference is in connection with item 5 (intercourse), where the
correlations are 0T2 and 0-25. Correlations are highest, for both
sexes, in connection with items loading on factors 2 and 3; they are
insignificant for items loading factor 1. Hence P influences sexual
conduct only when this is going beyond simple petting; in terms
of our theory it seems possible that petting involves a less imper¬
sonal element of interaction, although this hypothesis may be quite
beside the point. The variance contributed by this dimension of
personality is not high, but it should be remembered that the
distribution of P scores is rather J-shaped, with very few students
having high scores; this would tend to lead to an underestimation
of the importance of P.
Correlations with E, as predicted, are all positive, and, as also
predicted, are higher for males than for females as far as the
majority of items are concerned. Particularly large differences are
observed in relation to items 5, 6, 9, 12, 14 and 15; slight inversions
of the general rule occur with a few items. The ‘perversion’ factor
items on the whole tend to have lower correlations with E, although

72
 STUDENT POPULATIONS

still positive; petting, intercourse and manual manipulation of

partner’s sexual parts tend to give higher correlations. It is not
clear why this should be so; possibly the small number of subjects
who indulged in ‘perverted’ practices of this kind is responsible
for the observed facts.
Correlations with N, as predicted (although without much con¬
fidence) tend to be negative - uniformly so for the men, less so for
the women. In addition, the correlations for the men tend to be
bigger than those for the women. This again is understandable
in terms of the more active role that the male traditionally assumes
in this field; the high N male may have difficulties in initiating
sexual meetings and practices, while the high N female may have
less difficulties in simply responding. Indeed, she may have some
difficulties in not responding - to react in a negative fashion may
require anxiety-provoking self-assertion and independence! The
general run of the correlations is not high; they tend to be inter¬
mediate between those for P and those for E, perhaps somewhat
nearer the former than the latter.
The studies described so far have dealt only with students; would
differently selected groups of the population show similar patterns
of sexual attitudes, and similar correlations with personality? In
another study (Eysenck 1973a) the same questionnaires were ad¬
ministered, with the exception of the sexual behaviour one but with
the addition of a questionnaire on conservatism, to 241 male
prisoners, all of whom were adult volunteers. The conclusion
reached was that;

... this study has demonstrated that the patterning of sexual attitudes
is very similar in prisoners to what had been found in younger
samples of males of higher social class, education and intelligence.
Furthermore, the relations observed between the clusters of sex
attitudes and the personality factors measured were very similar in
size and direction. This seems to justify the conclusion that the
results obtained from unrepresentative samples (students, prisoners)
may have greater generality than one might have thought; the fact
that the results fit in well with the personality theory developed
by the writer also supports this view. [p. 305]

Another study suggested the possibility that, in addition to P,

E and N, there might be another personality variable that could
exert an influence on attitudes towards sex. This suggestion arose

73
 SEX AND PERSONALITY

from a study reported by Eysenck and Eysenck (1971c). Our

previous work had left somewhat open the question of the truth¬
fulness of responding, and the study in question sought to answer
this question by using a Lie scale, taken from E.P.I. (Eysenck and
Eysenck 1964) and consisting of eighteen items. It was hypothesised
that if sex attitude questions were answered truthfully, then they
would have low loadings on the Lie scale; high loadings would
suggest defensive or ‘fake good’ behaviour. A random sample of the
available population, consisting of 228 male and 263 female subjects,
filled in the questionnaire; these subjects had been assembled by a
market research firm for other purposes and had filled in the
questionnaire between other, commercial, duties. The population
thus contrasted with that previously used, which was comprised
of university students. Apart from the Lie scale the questionnaire
contained 47 psychoticism items, similar in kind to those used on
previous occasions, and 12 extraversion and 12 neuroticism items,
taken from the E.P.I. Interest centred on the possibility of using
sex attitude items as part of the Psychoticism scale, as in our
previous study (Eysenck 1971c) correlations with psychoticism
had been the outstanding characteristic of many sex-attitude items.
Many of these items were too explicit to be used with a random
audience, but the following eight items were used in the present
investigation;

1 Do you think that sex without love is highly unsatisfactory?

2 Do you think people must necessarily be in love to enjoy love-
making with each other?
3 Would it bother you if the person you married were not a
virgin?
4 Do you think it is all right to seduce a person who is old
enough to know what he is doing?
5 Would you object to pornographic writings being freely pub¬
lished?
6 Do you think that marriage is old-fashioned and should be
done away with?
7 Do you disapprove of censorship of films and plays?
8 Do you think that ‘wife swapping parties’ are disgusting?

The last of these items had not been included in the original
questionnaire; some of the others were slightly reworded, so that
items 2, 3 and 5 now require an answer in the opposite direction

74
 STUDENT POPULATIONS

in order to give correlations with psychoticism having the same

sign as before. Results of a factor analysis (principal components,
followed by Promax rotation) disclosed four clear-cut factors,
labelled psychoticism, extra version, neuroticism and he; the sex
attitude items had sizeable loadings only on psychoticism and he,
and these loadings are given in table 2.26, together with the cor¬
relations of these items on the original study with psychoticism
(no Lie scale had been used then, so that these correlations are not
available).

Table 2.26 Loadings on P and L of eight sex-attitude items

Items Original P P loadings L loadings

correlations new sample new sample
Male Female Male Female Male Female

1 (2) -017 -0-26 -001 -0-28 0-49 0-53

2 (-13) 012 0-22 -007 -0-22 0-59 0-56
3 (-26) 010 0-23 -018 003 0-25 0-42
4 (79) 018 0-23 005 014 -0-45 -0-26
5 (-84) 0-18 015 -0-12 -014 0-45 0-48
6 (-89) 0-23 012 0-43 0-24 -018 -0-03
7 (91) 016 0-22 0-24 0-21 -0-38 -0-17
8 (- ) -0-27 -0-43 0-56 0-42

It will be seen that the sex attitude items have loadings on

the psychoticism factor that are similar in size and identical in sign
to the originally observed correlations between these items and
psychoticism. The numbers of the items in the original scale are
given in parenthesis, so that readers can check details in the original
paper (Eysenck 1971a). The minus sign preceding numbers 13, 36
and 84 is intended to remind the reader that these were reversed
in the rewording, so that the signs of the correlations are opposite
for the two samples. Considering the difference between the two
samples, the agreement between the two sets of correlations is
reasonable.
The last two columns give the loadings of the eight items on the
Lie scale; it will be seen that almost all the items have very high
loadings on this scale and that these loadings are much higher
than the loading on psychoticism. This suggests that dissimulation
(‘faking good’) may have played a considerable part in the process
of answering these questions. There are reasons to doubt this inter¬
pretation. Table 2.27 gives the intercorrelations between the factors:

75
 SEX AND PERSONALITY

it will be seen that these are quite independent, and in particular

that Neuroticism and Lie scales are orthogonal.

Table 2.27 Intercorrelations between personality factors N, E, P and L,

for male and female subjects

NL NP NE LP LE PE

Men 009 001 -008 -o-oo -009 002

Women 0-03 009 -012 0-14 -0-15 -005

This observation is relevant to the question of the interpretation

of the Lie scale. Michaelis and Eysenck (1971) have shown that,
as motivation to dissimulate increases in a population, lie scores go
up, neuroticism scores go down and the correlation between the
two becomes more negative. The failure of these scales to show
such a high, negative, correlation suggests that dissimulation played
little part in the answering behaviour of this group and that the
correct interpretation may lie rather in the personality structure
of high lie scorers when motivation to ‘fake good’ is low. We have
surmised that such people are conformist, orthodox and conserva¬
tive in their beliefs (Eysenck and Eysenck 1976, Wilson 1973), and
there is evidence that the L scale does correlate with conservatism
in the studies mentioned.
Some direct evidence in support of this view is available in
the work of Joe (1976). He administered the Wilson Conservatism
scale to 66 male and 64 female undergraduates, together with the
Zuckerman Heterosexual Experience scale, finding negative cor¬
relations between conservatism and heterosexual experience (r=
-0-44). ‘These results suggest that low conservative subjects on
the whole engaged in a greater number and range of sexual ex¬
periences than high conservative subjects. For both male and
female, low conservative subjects have advanced to coital activities
while high conservatives have advanced to only foreplay activities
(e.g. manipulation of the genitals).’ High conservatives also had
fewer experiences of viewing ‘X’-rated movies.
These results suggest that our findings, as reported in table 2.26,
can best be explained not in terms of dissimulation, but rather in
terms of the conventional, conformist, conservative nature of the
high L scorer. Further data to support this interpretation will be
given in the next chapter.

76
 STUDENT POPULATIONS

L Strictly speaking, this should be ‘Lack or absence of premarital

sex’. For ease of understanding, the loadings have been multiplied
by — 1 for later analyses, so that this factor will be properly named as
in the text.
2. The programme was written especially by Dr O. White, and is
available from this Department.
3. It might also be named by its opposite aspect, ‘sexual satisfaction’;
we have used this second name in later chapters.
4. The relationship between the masculinity-femininity scale and the
libido scale would find an easy explanation if it were true, as Money
(1961a, 1961b), Foss (1951) and others suggest, that ‘the clinical evidence
... supports the view that, in adulthood, androgen is the libido hormone
for both men and women. The androgens of eroticism in normal,
untreated women may be of adrenal origin or may derive metabolically
from the closely related progestins’ (Money 1965, p. 76). However,
as Davidson (1972) has pointed out, ‘the fact that testosterone in the
doses administered in these studies causes clitoral hypertrophy vitiates
the worth of this evidence ... Doses of androgen which produce
clitoral hypertrophy would be clearly supraphysiological and indeed
this change may be responsible for the reported increase in “libido”.’
(p. 96). He concludes that ‘there is no adequate evidence that the
effects of androgens on feminine sexual behaviour represent any im¬
portant physiological function under normal circumstances.’ (p. 96).
The possibility that some such relation may nevertheless exist must
remain open. The evidence is more decisive as far as the importance
of androgens on aggressive behaviour is concerned (Conner 1972), but
even here there are many complications in the way of arriving at any
clear-cut statement of the position.
5. The inventory is based on the pioneering work of Brady and Levitt
(1965) and Bentler (1968).

77
 3

Sexual attitudes and behaviour in

adult populations

Our first series of studies, using the ninety-eight-item questionnaire

and a student population, was followed by another series, using a
rather longer and more detailed questionnaire and a more mature
population, many of them married or living together. It seemed
likely that these changes in measuring tool and population would
make for certain minor changes in the conclusions, but that on the
whole similar findings would re-emerge. Such at least was our hope.
None of the data collected has been published previously (unlike the
data in the previous chapter) and consequently, a more detailed dis¬
cussion will be given.
The sample for this investigation of the major factors in people’s
attitudes to sex consisted of 427 males and 436 females, ranging in
age from eighteen to sixty, with an average of around thirty years.
The questionnaires for men and women were almost identical, but
occasionally questions had to be reworded to make them applicable
to the other sex; e.g., a question put to the males, regarding whether
they felt excited when feeling a woman’s breasts, when put to the
women asked whether they felt excited when men felt their breasts.
The actual questionnaire used is given below (table 3.1). When there
were differences between the questions asked of the men and women,
the male form is given first, the female form second; in addition,
the letters ‘M’ and ‘F’ respectively are attached to the number of the
question. There were 158 questions for the males and the same
number for the females. The percentage of ‘Yes’ answers for each
question for the males and females respectively is given in each
case after the actual question. These questions, when repeatedly

78
 ADULT POPULATIONS

referred to in the text, or in the tables giving the factors extracted

from the factor analyses, have been paraphrased in order to save
space; the numbers are given in each case so that reference is possible
to the actual form of the question given in the questionnaire.
The sample of men and women is not representative of the
population, but it does include reasonable numbers of different
ages, as well as a reasonable selection of people of different social
classes, income group and educational backgrounds. An effort was
made to attract volunteers in many different ways, including circu¬
lating student groups, approaching individuals and groups (such as
the Salvation Army) personally, advertising and generally making
the existence of the inquiry known as widely as possible. The
results clearly cannot be used to estimate population parameters,
but this is not the aim of the study; we are concerned rather with
the degree to which questions are correlated, the factors they give
rise to and the degree to which these factors correlate with person¬
ality variables. Such data are much less vulnerable to faulty samp¬
ling, and in addition we have two safeguards against errors arising
from this factor.

1 We can compare the factors and correlations obtained with

those obtained from other samples selected on other principles; if
the factors and correlations agree in broad outline, then clearly the
different ways of obtaining the samples cannot have given rise to
large errors, nor can the resulting factors be pure artefacts.
2 In the second place, we can look at the scores of our group in
respect to the personality inventory used, and compare the results
with population estimates obtained with these instruments. We shall
see that, with respect to these personality variables, our sample
is not dissimilar to that used for standardising the personality in¬
ventory used, and consequently does not differ too much in one
very important respect from the general population. These con¬
siderations encourage us to believe that our results have wider
applicability than just to the sample tested.

The number of items used in the questionnaire was too large

to make possible a factor analysis on the University of London
computer, and consequently a sub-set of 135 variables was chosen
for analysis. The choice was based on such considerations as avoid¬
ance of items having endorsements in excess of 90 per cent or
in default of 10 per cent, duplication of meaning and high loadings

79
 oconom^cotj- »-H VO V“) so vo vo c^i
^ Tf HHm VsO VO <N

>nmvvTtoo>riOTt,a\ m
Tt
o Tt
’—•
s xj- — vo »o VO GO (N

w w
lu UJ
05 05
ooOooa°oooog o o o o
z<zzz<zzzzzz z z z z
e
£
o c^. c^* c^* c^* O' c^* c^* o* O* c<. c^* C^« C^*
£
W
e X W cn cn w cn co CO CO CO CO CO
UJ Oi UJ W UJ 05 W UQ UJ LU LU W
q ^ ^ <D M P O >H p O P >h
c i tih c
£ £ <3 <D
< <

u
o
u.
*G <d
G
ctf
£ GO
Co a
C §
Q
UC
8, J
<3
„?s >> <D
O a
>>
2 _o
5 e*
o (D
s*. X 1
o3 a> S>
<u «+H
a CO 05 g|
■M
00 O s. GO S
C/5 O
M
o r' c o
.c • e 2 .« ' <D fl *o
O 3 3 M
M 3 .<d 5-2 o
s
"“-i aS o O 5
o g * 8
« .9 o
<L> s m **
M T3
X
c3 cs
C/5
C/5 g S
3 a, a o
o > a
G 43 _ •33 O 2
co
O §’g 6 ~ H cS o
W >> >. u 43 3 >
4J O ■y X £ o
co o 2r! g
Diu'B •S 4J
5 w
JO
*> < .5? S
o
e o a M
Frankly, I prefer people of my own sex
3 43 w ao >• 03 0S o
• 2 WJ
t> o
.s s 60 9
a o a
a O 5 m ctf
>. e •P ^
X ’*"*
I am sexually rather unattractive

o D •*-» •o 9
u S 2Q
S3
43 «1
c/5 43 al
-H £0 3 n ?
£ C3 'C 3 o o O
'-h a3
o
■M a
> cl >.-a g i b 8 8CL
c °«
*> I M M * CO
4) •'*' C/5 Zt S3 3 <D '1
03 D
c O M
Cl o c g o
s is « -g GO
■ ■ G
rr-J

1 8 ^ 2
V
M
o o : y* >•
m il) "B TS ■°c«
§ G 7?
O v
0) *M
4> ^
o3 c/5 I* fl
course with him

> " M S •a §^£ 5

<D
■s.l
Lh
.3 £ ■M C
3 +-> J= 53 •- o G ^ ^ 2
oj
•s-a 03 c/5 S a <D G
H o § 3 8’S
g. ■S.° >> ° sj| ^J8 co
5 O
Cl, s-e O CL o ’*S
G "O
O
O i:is « 2
£ '■S M ^
(U 03 co
43 g§ X3
H co U ler
iNn^vnohoo ono-nsi
14
15

80
 CO — o r~ o >o so CO r- Tf 00 Q\ n h n -h oo o <n — vo o Tf r-~
fr, l-H CO VO <N (N CO 00 \o Tf co f>n0N>n'O Tf ^ ^ "t r) <s
<N SO VO Tf Tf VO CO Tf 00 so o x
CO VO CM Tf CO CN VO

w W W UJ w w §
cfl O o o O
O 2 § 2 o 2 Q o o o o o o
z < z
fe
<Z? Z Z <; z z Z z Z Z Z Z Z
tt. h tu

c^« o- c^ c^* c- c^* o* o- c^* e^ o* c*-* C**« c^* c^* O* C- C- c^» CN» c^* c^* o« c^* c^* o-

(fl C/3 CO w CO • VD g co co c/d c/5 co

w 1 s g » g B4 U3 D
(3*
w UJ
g | s g S g
>
£ >* S > P >
H
>« >H P

00
a

T3
C
X)
S3
o >»
4> (H
■*-» E
<D
6
s >»
— — *-i 1-3
* * g-s o
^

ccd
« 3 g B 6

a fl 4> 00 1>
C/5
co c3 -*-*
^3 0
K O --S
C3
C/5 s —c«Ei>5
-
c/5 ’H
o> t3
<-(
cd
l_,

33 S OT3
fO D, O
QhX «« » *s!-b-p s*
o< q 3 glsi6£
c
oj.- 3 uX y
a! ** u r£ « a
* _ 1>°>,0S£'cS2
SflS>.S
V *3 H C >
^ O1” - O-^.M *
E C 01 £ u to
« » -g ^ U c *
o
_, _ u t£ 'C »
-*-* G O > C
>»§)j=^ §
S3
C/5 11^5-S
£ to 'c <B 3

_ <D CO ? | 1-S.ij
*.S2 * £
U „_, D O
i/3 hh co co

\£)h*000\O^NfO,,t,0'oF'00^0’^N!?,t{n'^)l>OOONQ’-jNm5»0'0
I I ' I I I—J I—I

81
 i—i co co wo VO wo o 00 r-i 00 wo vo 00 o os os vo CN co r- OO CT\
Ph co oo CO oo oo CN T-H CO 00 TH wo WO 9
QO 9
V£> CO 00 >n «o

O N Tt V£> co 00 ,
Tf WO CO wo VO ,
Tf VO ON wo r^ OO On CN VO <n r- r~- ■<*
S (N (N t" CO Os oo m <S\ CO wo vo c4 oo r- ■Tf 00 \C v-i

w W w
o c o o co O o O O O CO o o o O o o COO 0 Q o o o O O 2 2
z z z z 5j Z z Z z Z z z z z z z z z z z z z z z z z z z
u- Uh

o. c^* c^* c^« e^« o* <>• c^* c^« e- c- c^* o- o* e-- o* c^* c^* c^* c^* c^* C<. c^- c^* c<. c^* o- c^-

(/DC/)^C0Sc/)^i(^C0C/DC«S(/)C0C0^(Zl(Z)CZ)Cnc/5C0C0C/5(/17DC«C/JC/DC0

sssslsgssggisssgssgssssisgsssss
00
c
'o
*0
T3 <D
o u»
cd
x: >*
o X3

co
jo
3
«S
H

•r ^ >
§‘Q^
s S 5
.S
g S
^ > <U o
^-» <L>
0) * X> «a rci
«
C4) &
00 O o « «
“ £ E w St
D >
1
H (Z t> M M serf
■'n«t'00
t?22!9'HNn7'riV0r'(!0O'Q'HNnt^.^. — W1
0\o««Nn'tio
w „ „
TfTi-Tfv-iv-i>ov-iinv-iv-4>^i>o>^i'$'Ovo'o'o\ovo'0'Ovor;<t~t^t<Ht;-r^

82
 44
37
55
63
68

20
00

26
Tt"

53
oo CO VO Tf
04 wo *o
84 O' <N 5*
67
74
82
56
C\ 00 m

35
73
ON o\ o o* wo oo oo h« - 'O VO
_ Tf
o~ —- co «o
Ti¬ ^ O' M M ^O Tt

W
I
Uh
UJ bi w w
NO

CO
NO
NO
9
2 o NO
CO

NO
NO

NO

NO
NO

NO
NO
2 o b b 0 O
c/5 co
< < < z z
CO b b b b

o* o* c^* O' O* o- O' C-* O' O* O' O' o* C^' O* O' O' O' O' O' O' O' O' O' O' O' O'

C/5 C/5 C/5 C/5 CO C/5

g g g £ £ £
SfBfESiSS 'iBBgBSSBSBSBil

vo r- oo o\ o —i ts mT}-<r>'Ot^ooONO — NWTtio«r'coftOHNt<iTt-
f' oo oo oo OO 00 00 00 00 00 00 C\ C\ ftap'O'O'O'jionoopoq

83
 m vo »o r- on vo o o vo ON to r- _ O <N oo (N o o *—1 oo vo OO
Ph n <n ^t «o t-H vo r- vo <N VO vo to *"• r- tN CO (N CO »o to

m rf On xf r4 ON «o Os to to to to c4 xj- vp 00 vo ON vo ON ON oo CO CO
S "d- (N TJ* to VO vo m to to 1"H to to <N VO vo (N y"^ to VO to

S W w w w tu PJ
tn co CO CO cfl C/5
0 o o o O O o o o O O O o o O O O O O o O o o
z z b-4 z Z Z z
z z Z Z z z z z z z Z z z Z z z < z
n. £ tt< Bh u-

P- c- c^* c^* c- c^. c^» o. o* c- e-~ c^* »• e^« o* c^« c^« c^* e- c^* c^« c^- o- c^* p- e-

C/5 C/5 CO C/5 CO C/D c/5 c/5

W UJ
s s gg g g JH ><

co

D
3
ctf
H

:0»HNff)fnTtI2^or^ooaNO*H(snTtin»o^t^oooN2^(S

84
 r- m vn m lo oo

58

55
44
tL,

32

39
*-H T}- VO i—i oo cm <N
r-
cn m *o VO

16
cn i—i i"" vo

36

55
—« rf m i—« oo 8 Tf cm oo VO
OO
W W
SSfta • W W w w W
oo^^oo^o cn
^OOOCOOO
$
co co £ zz^zz^z $ ?z;zzzzzz
fr< bn bn
S 3
O- c-- o- C- c^. O'* C-* c^» c^« c^* c~* c~* o- C^* JS-. c^. c^. c^. fs..

S 9 CO § CO cocoSScocogco §
Oh Oh 113 ^ tu
o o 5 £ S
r* H H H gsssgsss

-C

3*H =$
o
£
Co
c
o
<D 3-~
o a Ol 9 y
fi § 3 * 8
<u - 57
o * C* o Pi
O T)
SJ 3 ^ 2 —
•« o .£, Z> Z5
W O w
? *
2
,5 6 Oh
<U <a
a00
o
ss u
a p 6
<*>
§
a
>.
s: g 2
M
2

•a
u «V
« 5 37
“ te o
S 8
<u
u ^
cS bl
<D '—s
c3
w ^ w
£
3 3
O O
f>> >>

^1* ^ Tf ^ Tj- »0 cm
*n

85
 (N <N »n
<D to I *n OO qs
6s vb ra

if m oo
00 <D to 13 oo rn in
s 6v r- Cvl

C/5

00 P
c tO
*5b
<D
©
u -
to
a o
o <D
to
to 03
<D
>
5 o «
p to U ’O
3
o d d
.p s ©
G ’aT
13 s3 s
o o <d
*-
e c a P J3 1 *
o c3 P4 w
<D G a 3
O G G c
G ■g to
■+-» g <d
O -H—*
’5b■"
g
Oh 'to ^ six
If you were invited to take part in an orgy, would you:

G P a '—x O <D
>» s <D a> E .-
§ 3 m x o
w G 2
G « 13 rS * © S' > £o»“
o
c 5^
c/3 0)
>> <*-1
(b) Refuse

>> o go
si frg ccJ S Si cu 2 13 >■ 5 ► TP g w « o O 3
in > £ £ o to
o a ^ So H > o J-l C3 K*v . ^_i
•1-H -♦-»

..OS< OS Zffi S Z EZ 1 - - to -3 jd
3 .“
Ideally, would you prefer to have intercourse:

<D
o /-—S — o Q^ ^
T3 u'C' 60 Si 5 13 <D
w*§ sS
c-9 <3 a) C o 3
g s—x - d II
S 00 « 3
a e X _.
l "5
’5b
to
O
.Q
5?
*5
p
.5 « «^ 3—. cu
d£ 5
2 -E
s
M-l 13 Im
s so to
o so (H
ctf p f-sll
00 o
13 13 >>
■8 U
CD
> ll'lf
<D <0 (D
£ u. M
o CD <D
CO © > G «u t
o ft
(D *g w G X! 3 „ Jm «
P g P ■“ c
• P co ■ O _ 3?
(a) Take part

6 I Ut u* 60 >> .£ p
bO o CO
CD O, g ^ ° to 0) o
£ c 3 !« ■ ■ f—S o--9
> > PP d O •S
<D <D
^ c <n 3 P v s ^33
Ui to TP “ C S S ■- to -r?
<d
& s> « 3 tn 4-» CD bo
o
P «D 4h •-3m cS^ o e»o .s
_Q Cj w
=> <D <D
zoo o& >» P
pzo
O, O >o (D O
S 3 > OS m 3 i M— ^ 3
H—S /-"“S ^ o ~ M
'3'2'i? to O © ' <d G X) O Pm d too ^ 6 ^ CH
' > S_^ •>-^ (D § <d G
as > G
O G to o
E Q pp *3 ^
<D __ W ^
•*-< >Tl -4-H
P g CCS O
Pm
cd P4 13
ctJ c3 X)
v~> »n
153

154

*n vo r- oo
in »n »n «n

86
 ADULT POPULATIONS

of the items on previous analyses. The items omitted from analysis

are enclosed in square brackets in the questionnaire. Phi coefficients
were used for intercorrelating the items. The use of this coefficient
has been criticised in connection with factor analytic work because
it is affected by the marginal proportions to such an extent that
its maximum possible value is restricted where proportions vary
from one another; this particular phenomenon has been thought
to be responsible for the introduction of spurious ‘difficulty’ factors
(Cattell 1952, Ferguson 1941, Wherry and Gaylord 1944). Tetra-
choric coefficients and a corrected phi coefficient (phi/phi max.)
have often been preferred. The whole problem has been examined
empirically by Comrey and Levonian (1958); they state that ‘it
seems reasonable to conclude that the phi coefficient is the method
of choice in point correlation work where factor analysis is to
follow. This thesis is strengthened by the fact that the number
of significant centroid factors obtained is at least as great with
phi-over-phi-max and tetrachoric r as with phi. Hence, if spurious
factors exist with factor analysis of phi coefficients, they may be
no less evident with phi-over-phi-max or tetrachoric coefficients.’ (p.
753). Phi coefficients resulted in analyses relatively free of exces¬
sively high communalities encountered when using the other two
types of coefficients. It may be concluded that the use of phi co¬
efficients in this study is not contra-indicated by the best available
opinion.
The analyses were done by Promax rotation of principal com¬
ponents, using the method of Hendrickson and White (1964). Fifteen
factors were orginally extracted, although eigenvalues in excess
of unity were found for some forty factors. However, in this as in
our previous studies, we found that only about a dozen factors
could be interpreted with any confidence, so that the number ex¬
tracted is if anything in excess of the number of ‘true’ and mean¬
ingful factors contained in the matrices. Factor comparison was
made of the fifteen male and the fifteen female factors and those
factors showing the largest comparison coefficients were then
matched. The method of factor matching was that described in
Eysenck and Eysenck (1969). The factors were interpreted and
named in terms of the items having the highest loadings; table
3.2 presents the results. Given are the extraction numbers of the
factors for males and females, the coefficient of factor comparison
and the tentative name given that factor. It is of course realised

87
 SEX AND PERSONALITY

that such names are inevitably subjective, but they have a useful
function to perform in making reference to a given factor easier
to follow. In any case, it is possible for the reader to check the
adequacy of the naming procedure by looking at the detailed
results for each factor given below.

Table 3.2 Factor comparisons for males and females on sex questionnaire

Male Female Factor comparison Name of factor

(Extraction factor number) coefficient

1 1 0-93 (1) Permissiveness

2 2 0-92 (2) Satisfaction
5 9 0-95 (3) Neurotic sex
6 5 0-97 (4) Impersonal sex
4 11 0-91 (5) Pornography
13 6 0-89 (6) Sexual shyness
7 13 0-76 (7) Prudishness
8 15 0-75 (8) Dominance-submission
15 14 0-71 (9) Sexual disgust
9 3 0-70 (10) Sexual excitement
3 7 0-66 (11) Physical sex
11 4 0-52 (12) Aggressive sex

We now proceed to give the actual items having loadings of 0-3 or

above that go to make up each factor. Factor 1 is called ‘Permissive¬
ness’, and the items seem to make this name quite appropriate.
The index of factor comparison is 0-93, suggesting that both sexes
are very closely in agreement on their interpretation of the items
that go to make up this factor.

Factor 1: Permissiveness
Item Loading Item
Males
5 -0-37 Virginity is a girl’s most valuable possession
17 -0-45 Disturbing to see necking in public
25 0-37 Wouldn’t bother if person married not virgin
38 -0-53 Shouldn’t experiment with sex before marriage
40 -0-38 Thought of sex orgy disgusting
41 -0-62 Better not to have sex until married
57 0-61 ‘Pill’ should be universally available
64 -0-42 Sex jokes disgust
67 0-54 Young people out at night not too closely checked
78 0-76 Pornographic writings should be freely published
79 0-44 Prostitution should be legally permitted
80 0-35 Abortion should be the concern of no one but the
woman concerned

88
 ADULT POPULATIONS

Item Loading Item

Males
81 -0-68 There are too many immoral plays on T.V.
85 0-72 There should be no censorship on sexual grounds
in films and plays
87 -0-70 Sexual permissiveness threatens to undermine
civilisation
93 0-34 Sex play among children is harmless
134 -0-49 Object to four-letter words in mixed company
157 -0-38 Age at first intercourse low

Females
1 -0-46 Opposite sex will respect more if not too familiar
5 -0-52 Virginity valuable possession
17 -0-51 Disturbing to see necking in public
25 0-40 Wouldn’t bother me if person I married was not
a virgin
38 -0-68 Should not experiment with sex before marriage
41 -0-74 Better not have sex before marriage
57 0-39 Pill should be universally available
64 -0-66 Sex jokes disgust me
69 -0-61 Would protect children from contacts with sex
76 0-30 Like to look at pictures of nudes
78 0-48 Pornographic writings should be freely published
79 0-38 Prostitution should be legally permitted
81 -0-73 Too many immoral plays on T.V.
85 0-38 No censorship on sexual grounds of films and plays
87 -0-72 Sexual permissiveness undermines civilised society
91 -0-36 Preoccupation with sex created by T.V., films,
newspapers
93 0-37 Sex among young children is harmless
134 -0-44 Object to four-letter words in mixed company
149 - 0 31 Reading girlie magazines failure of adult attitude
to sex
151 0-36 Blue films invitation accepted

Factor 2 is labelled ‘Satisfaction’, and again the meaning of the

factor is quite unmistakable. The index of factor comparison is
092, which again suggests considerable agreement in structure of
attitudes between men and women.

Factor 2: Satisfaction

Item Loading Item

Males
4 085 Satisfied with sex life
11 -0-48 Have been deprived sexually
16 0-32 Sex contacts have never been a problem
19 -0-82 Something lacking in sex life

89
 SEX AND PERSONALITY

Item Loading Item

Males
21 -0-73 Love life has been disappointing
56 -0-36 Worry a lot about sex
105 -0-32 Don’t always know for sure when had an orgasm
108 0-64 Partner satisfies all physical needs
113 0-52 Find easy to tell partner like/dislike about love-
making
114 -0-45 Like partner to be more expert, experienced
118 -0-51 Feel sexually less competent than friends
139 -0-41 Cannot discuss sexual matters with wife/partner

Females
4 0-78 Satisfied with sex life
11 -069 Have been deprived sexually
16 0-32 Sex contacts never a problem
19 -0-84 Something lacking in sex life
20 0-30 Sex behaviour never caused any trouble
21 -0-63 Love fife been disappointing
56 -0-36 Worry a lot about sex
108 0-77 Sex partner satisfies all physical needs
113 -0-59 Would like partner to be more expert, experienced
114 0-37 Easy to tell partner what I like/don’t like about
love-making
117 -0-50 No one been able to satisfy sexually
118 -0-36 Sexually less competent than friends
147 -0-32 Right that male should be dominant sex partner
155 -0-33 Suffered frigidity

Factor 3 has been called ‘Neurotic sex’, and the very high index
of factor comparison suggests that the items characterising this
factor are very similar for both sexes. Another name for this
factor might have been ‘Conflict over sex’; the items are suggestive
of a conflict between strong sex drives and conscience or some
other factor holding back the person from indulgence.

Factor 3 : Neurotic sex

Item Loading Item

Males
7 0-57 A problem to control sex feeling
18 0-54 Sexual feelings unpleasant
20 -0-43 Sex behaviour never caused trouble
24 0-57 Felt guilt about sex experience
26 060 Afraid for what might do sexually
32 0-38 Sex disturbs more than it should
44 0-52 Conscience bothers too much
46 0-45 Sexual feelings overpower
56 0-34 Worry a lot about sex

90
 ADULT POPULATIONS

Item Loading Item

Males
60 0-64 Perverted thoughts have sometimes bothered
68 0-61 Sometimes felt like humiliating partner
75 0-57 Sometimes felt hostile to partner
84 0-37 Had some bad sex experience when young
100 0-32 Sexual fantasy often involves flogging
155 0-33 Suffered impotence

Females
7 0-54 Problem to control my sexual feelings
18 0-41 Sexual feelings sometimes unpleasant
23 0-42 Try to keep sex thoughts out of mind
24 0-47 Guilty about sex experiences
26 0-64 Afraid for what might do sexually
32 0-38 Thoughts of sex disturbs
44 0-35 Conscience bothers too much
46 0-59 Sexual feelings overpower
48 0-40 Sex thoughts drive crazy
56 0-32 Worry lot about sex
59 0-39 Thinking about sex makes nervous
60 0-33 Perverted thoughts sometimes bother
84 0-45 Bad sex experiences when young

Factor 4, called ‘Impersonal sex’, is equally clear as the previous

three factors, and has the highest index of factor comparison, 0 97.
Very similar items load this factor in the case of men and women,
respectively.

Factor 4: Impersonal sex

Item Loading Item

Males
2 -0-50 ‘Impersonal sex’ highly unsatisfactory
40 -0-37 Sex orgy disgusting
42 0-34 Thought of coloured partner particularly exciting
65 0-46 Believe in taking pleasures where found
83 0-37 Should do away with marriage entirely
89 0-55 Faithfulness is nearly as silly as celibacy
92 0-75 Enjoy watching partner have intercourse with
someone else
95 0-57 Would vote for law permitting polygamy
97 0-37 Prefer new partner every night
119 0-73 Group sex appeals
120 0-45 Illicit relationship excites
135 -0-65 ‘Wife-swapping’ is distasteful
144 0-77 Would not be disturbed if partner had sex with
someone else
153 0-67 Accept an invitation to an orgy

91
 SEX AND PERSONALITY

Item Loading Item

Females
2 -0-35 Sex without love unsatisfactory
46 -0-39 Sexual feeling overpowers
65 0-35 Take pleasures where found
83 0-42 Do away with marriage
89 0-51 Faithfulness as silly as celibacy
92 0-60 Enjoy watching partner have intercourse with
another
95 0-53 Vote for law permitting polygamy
102 0-37 Sex more exciting with a stranger
119 0-70 Group sex appeals
120 0-38 Illicit relationships excite
135 -0-67 ‘Wife-swapping’ distasteful
144 0-60 Not disturbed overmuch if partner had relations
with someone else
153 0-75 Accept invitation to an orgy

Factor 5 contains largely items dealing with pornography, and

has accordingly been so named. The index of factor comparison
is 0-91, which is quite satisfactory.

Factor 5: Pornography
Item Loading Item
Males
10 0-30 Get pleasant feelings from touching sexual parts
43 0-76 Like to look at sexy pictures
58 — 0'40 Seeing a person nude doesn’t interest
76 0-70 Like to look at pictures of nudes
77 0-54 If had chance to see people making love, would
take it
83 -0-30 We should do away with marriage entirely
141 0-32 Naked body is a pleasing sight
151 0-59 Accept an invitation to ‘blue film’
152 061 Accept if offered a highly pornographic book
Females
10 0-44 Pleasant feelings from touching sexual parts
13 0-30 Need neither love nor respect to enjoy petting/'
intercourse
40 -0-31 Thought of sex orgy disgusting
43 0-77 Like sexy pictures
58 -0-38 Seeing a nude doesn’t interest
76 0-72 Like pictures of nudes
77 0-69 Would take chance to see people making love
unseen
96 0-42 Masturbation good for a change
151 0-44 Accept an invitation to ‘blue film’
152 0-55 Accept if offered pornographic book

92
 ADULT POPULATIONS

Factor 6, like the preceding ones, is largely self-explanatory; it

has been called ‘Sexual shyness’. The index of factor comparison
is 0-89, which is reasonable.

Factor 6: Shyness

Item Loading Item'

Males
22 0-42 Never had many dates
29 0-43 Have had strong sex feeling but can’t express
47 0-81 Feel nervous with opposite sex
50 -0-83 Feel at ease with opposite sex
52 0-79 Hard to talk with opposite sex
59 0-30 Thinking about sex makes nervous
61 0-39 Am embarrassed to talk about sex
124 0-35 Afraid of sexual relationships

Females
47 0-81 Nervous with opposite sex
50 -0-87 Feel at ease with opposite sex
52 0-78 Hard to talk with opposite sex

These six factors constitute the backbone, as it were, of this

analysis; the remainder, although not difficult to interpret, have
somewhat lower indices of factor comparison, and there is con¬
sequently a suggestion that men and women may not entirely agree
on the nature of the factor in question. Nevertheless, they are
similar to factors unearthed in earlier analyses, and are sufficiently
clear-cut to make it permissible to introduce them here. Factor 7
is called ‘Prudishness’, and has an index of factor comparison
of 0-76.

Factor 7: Prudishness

Item Loading Item

Males
15 0-41 Prefer people of own sex
51 0-57 Don’t like to be kissed
55 -0-48 Enjoy petting
58 0-50 Seeing a nude person doesn’t interest
64 0-31 Sex jokes disgust
71 -033 Get excited when touching woman’s breast
76 -0-30 Like to look at pictures of nudes
124 0-33 Afraid of sexual relationships
126 0-55 Can’t stand being touched
141 -0-43 Naked body is a pleasing sight

93
 SEX AND PERSONALITY

Item Loading Item

Females
51 048 Don’t like to be kissed
55 -0-48 Enjoy petting
97 -034 Prefer a new partner every night
112 -0-41 Enjoy lengthy pre-coital love-play
122 -051 Sexual activities are average
126 041 Can’t stand being touched

Factor 8, labelled ‘Dominance-submission’, has an index of

factor comparison of 0-75, but the label is applied only very tenta¬
tively; the items leave open many other interpretations, particularly
of the male factor.

Factor 8: Dominance-submission

Item Loading Item

Males
31 -0-36 Parents’ influence inhibited sexually
65 0-30 Believe in taking pleasures where found
93 -050 Sex among children is harmless
99 061 Prefer partners several years older
110 0-31 Most men sex-mad
158 0-33 Strength of sexual desire great

Females
10 030 Pleasant feelings from touching sexual parts
82 -039 Dual standards of morality should continue
128 -0-36 Prefer partner to dictate rules of the sex game
147 -051 Right that males should dominate in sex relations

Factor 9, entitled ‘Sexual disgust’, is much more clear-cut

although the index of factor comparison is only 0-71. Nevertheless,
the items defining it are similar for the two sexes, and their mean¬
ing is unmistakable.

Factor 9: Sexual disgust

Item Loading Item

Males
9 -0-54 If loved a person could do anything with her
34 0-58 Some things wouldn’t do with anyone
104 0-37 Some things do only to please partner
133 0-33 Female genitals are aesthetically unpleasing
146 0-58 Some forms of love-making are disgusting

94
 ADULT POPULATIONS

Item Loading Item

Females
9 -0-35 Love a person, could do anything with them
34 0 31 Some things wouldn’t want to do with anyone
48 -0-36 Sex thoughts drive crazy
105 -0-30 Don’t know whether had an orgasm
112 -0-33 Enjoy pre-coital love-play
121 -0-37 Usually feel aggressive with partner
129 -0-47 ‘Straight sex’ unsatisfactory
146 0-35 Some forms love-making disgusting

Factor 10, too, is very explicit; it is called ‘Sexual excitement’,

and has an index of factor comparison of 0 07. There is little doubt
about the meaning of this factor.

Factor 10: Sexual excitement

Item Loading Item

Males
3 -0-61 Conditions have to be right to get excited sexually
30 0-72 Doesn’t take much to get excited sexually
39 0-68 Excited sexually easily
155 -0-36 Not suffered from impotence
158 0-33 Strength of sexual desire great

Females
3 -0-41 Conditions have to be right to get sexually
excited
6 -0-43 Think rarely about sex
9 0-64 Love a person, could do anything
18 -0-37 Sexual feelings sometimes unpleasant
30 0-79 Doesn’t take much to get sexually excited
34 -0-58 Some things wouldn’t want to do
37 0-46 Think about sex every day
39 0-78 Sexually excited easily
58 -0-30 A person nude doesn’t interest
61 -0-32 Embarrassed to talk of sex
71 0-40 Excited when men touch breasts
86 0-48 Sex greatest pleasure
109 -0-56 Sex not all that important
125 -0-33 Wish men were less sexually demanding
133 -0-53 Male genitals aesthetically unpleasing
141 0-30 Naked human body is pleasing
146 -0-61 Some love-making is disgusting
154 0-60 Preference for having intercourse often
157 0-64 Strength of sexual desire great

95
 SEX AND PERSONALITY

Factor 11 is one of ‘Physical sex’, and has an index of factor

comparison of 0-66; this is rather low, but the items leave little
doubt about the interpretation of the factor.

Factor 11 : Physical sex

Item Loading Item

Males
48 0-30 Sex thoughts drive nearly crazy
49 0-36 Excited, can think of nothing but satisfaction
71 0-32 Get excited touching a woman’s breast
72 0-32 Have been in more than one sex affair at a time
86 0-54 Sex is greatest pleasure
106 0-57 Few things more important than sex
109 -0-32 Sex not all that important
111 0-62 Being good in bed terribly important in partner
121 0-36 Usually feel aggressive to partner
127 0-60 Sex is the important part of marriage
131 0-52 Physical attraction important
145 0-31 Men more selfish in love-making than women
158 0-30 Habitual strength of sexual desire great

Females
31 -0-30 Parents’ influence inhibited sexually
53 0-36 Learnt facts of life quite old
86 0-46 Sex greatest pleasure
103 0-47 Never discuss sex with parents
106 0-62 Few things more important than sex
111 0-39 Good in bed with marriage partner important
127 0'66 Sex most important part of marriage
131 0-39 Physical attraction important

Factor 12, the last of the interpretable factors, is called ‘Aggressive

sex’; it has an index of factor comparison of only 0-52, but the items
having the highest loadings are sufficient similar to make the inter¬
pretation possible.

Factor 12: Aggressive sex

Item Loading Item

Males
13 0-33 No need to respect/love a woman to enjoy
petting / intercourse
68 0-34 Sometimes felt like humiliating partner
88 0-32 Sex should be for purpose of reproduction
101 0-41 Make lots of vocal noises during intercourse

96
 ADULT POPULATIONS

Item Loading Item

Males
116 0-50 Sometimes feel like biting, scratching during
intercourse
121 0-44 Usually feel aggressive with partner
132 -0-33 In sex, tenderness important quality

Females
68 0-53 Felt like humiliating sex partner
75 0-54 Felt hostile to sex partner
91 0-41 Preoccupation with sex created by T.V., films and
newspapers
116 0-46 Feel like scratching, biting partner
121 0-30 Feel aggressive to partner
145 0-35 Men more selfish in love-making

These twelve factors, being oblique, are of course not uncor¬

related and table 3.3 shows the actual correlations between the
factors, for males and females separately. It will be seen that these
correlations are similar for the two sexes, but not identical; particu¬
larly in the case of sexual excitement are there quite marked
differences between males and females in the pattern of correlations
with the other factors. However, there is clearly a tendency for the
correlations in the table to fall into two groups, which are largely
independent of each other. The first group, for men and women
alike, is made up of factors of permissiveness, impersonal sex, por¬
nography, dominance, physical sex and aggressive sex; for the
women sexual excitement also falls into this cluster. The second
group of correlations joins dissatisfaction, neurotic sex, shyness,
prudishness and disgust; for the women, again, sexual excitement is
(negatively) correlated with this cluster. For the men sexual excite¬
ment is correlated with the other factors in much the same direction,
but correlations are all much lower. This is the only difference
between the two sexes that stands out from the analysis; apart
from this, the two patterns are quite similar.
When higher-order factors were extracted in the next stage of
the Promax programme, two terminal factors were found for both
men and women, and these bear a close relationship to the two
clusters identified by simple study of the matrices of factor inter¬
correlations. We shall call the first of these factors ‘Sexual Libido’
and the second ‘Sexual Satisfaction’, again without wishing to pre¬
empt any discussion about the true nature of these factors. Loadings

97
 8
"o5
%

Os ON VN VO ON © m cm OO
CM O y—> CM CM © © p
6 6 6 6 6 © © © © © © r
i 1

co o ON CM oo Tf <N vo m
i-H © o CM © CM ©
6 © 6 © © © © © © © 1 ©
i 1

VO ,
CM ON o NO Tt oo
o O O y—i p © © ©
twelve factors for males and females separately

6 6 © © © © © © © 1 © ©
1 1 1

o CM VO 00
t cm
ON o o CM o © CM P © ,—c
© © 6 6 © © © 1 © ©
© ©
i i i i i i i

© © VN 00 OO OO vo ©
oo rn © © CM 1-H *—1 © © © ©
© © © © © © © I © © © ©
1 1 1 1 1 1

•o vo Tf CM oo »o On m
© CM © © © CM © © ©

Females
r-* ©
© 1 © © ©
© © © © © © ©
1 1 [ 1 1

m cm oo CM © OO ©
vo © m © s © CM © m i S
© © © © © 1 © © © © © 6
1 1 1 1 1 1 1 l

oo © *T) C" Os oo oo M5
CM © © © © © CM © O
© © © © 1 © © © © © © 6
1 1 1 1 1 1

cn 00 CM 00 Tj- 00
rt- ’-H © CM i-H s © © CM © CM
© © © 1 © © © © © © © ©
1 1 1 1
between

vo CM vo © *o
m © cn © CM © s CM ©
© © 1 © © © © © © © © ©
1 1 1 1 1

vo Os VO VN vo vo CM Os
CM © CM © © CM s CM r-H CM ©
Correlations

© 1 © 6 © ©
© © © © © ©
ii i i i i ii

0 \>oI~-ro(s)ot'~ooP-'ooO'\
OOOOOOOOOoO
Table 3.3

fM m
 ADULT POPULATIONS

of more than 0-3 again are used to sort out those items that best
characterise the factors and items and loadings are printed below.

Superfactor 1: Sexual libido

Item Loading Item
Males
1 -0-34 Opposite sex will respect more if not too familiar
2 -0-45 ‘Impersonal love’ highly unsatisfactory
5 -0-34 Virginity is a girl’s most valuable possession
6 -0-32 Think rarely of sex
10 0-30 Get pleasant feelings from touching sexual parts
17 -0-30 Disturbing to see necking in public
25 0-38 Would not bother if person married was not a
virgin
30 0-30 Doesn’t take much to get me sexually excited
34 -0-35 There are some things I wouldn’t do with anyone
37 0-45 Think of sex almost every day
38 -0-48 One shouldn’t experiment with sex before marriage
39 0-37 Get excited sexually easily
40 -0-57 Thought of sex orgy disgusts
41 -0-62 Better not to have sex until married
42 0-38 Coloured sex partner particularly exciting
43 0-43 Like to look at sexy pictures
46 0-42 Sometimes sexual feelings overpowering
48 0-40 Sex thoughts nearly drive crazy
49 0-32 When excited can think of nothing but satisfaction
55 0-32 Enjoy petting
57 0-35 ‘Pill’ should be universally available
64 -0-41 Sex jokes disgust
65 0-55 Believe in taking pleasure where found
72 0-38 Have been involved with more than one affair at
time
74 0-55 All right to seduce a person old enough to know
76 0-33 Like to look at pictures of nudes
77 0-48 Chance to see people making love, would take it
78 0-49 Pornographic writing should be allowed to be
freely published
79 0-37 Prostitution should be legally permitted
81 -0-35 Too many immoral plays on T.V.
85 0-40 There should be no censorship on sex grounds
of plays and films
86 0-38 Sex far and away greatest pleasure
87 -0-34 Sexual permissiveness threatens to undermine
civilised society
89 0-47 Faithfulness to partner is as silly as celibacy
92 0-32 Enjoy watching partner have intercourse with
someone else
95 0-35 Would vote for a law permitting polygamy
96 0-35 Though having regular intercourse, masturbation is
good for a change
97 0-31 Prefer a new partner every night

99
 SEX AND PERSONALITY

Itei Loading Item

Male
106 0-37 Few things more important than sex
109 -0-31 Sex not all that important
119 0-53 Group sex appeals
120 0-45 Thought of illicit relationship excites
125 0-39 Wish women more forthcoming sexually
127 0-31 Physical sex important part of marriage
134 - 0-37 Object to four-letter words in mixed company
135 - 0-51 ‘Wife-swapping’ extremely distasteful
146 - 0-37 Some forms of love-making are disgusting
151 0-54 Accept invitation to see ‘blue-film’
152 0-54 Accept pornographic book if offered
153 0-60 Accept invitation to orgy if offered
154 0-40 Preference for having intercourse frequently
158 0'50 Strength of sexual desire great
Fern;
1 - 0-43 Opposite sex will respect more if not too familiar
2 -0-55 Sex without love unsatisfactory
3 -0-33 Conditions just right to get excited sexually
5 -0-41 Viriginity a girl’s most valuable possession
6 - 0-32 Think rarely about sex
10 0-37 Pleasant feelings from touching sex parts
25 -0-44 Not bother if person married not a virgin
30 0-33 Doesn’t take much to get sexually excited
36 0-31 Understand homosexuals
37 0-40 Think of sex everyday
38 -0-43 Should not experiment with sex before marriage
39 0-39 Get sexually excited easily
40 -052 Thought of sex orgy disgusting
41 - 0-58 Better not to have sex until married
42 0-40 Coloured partner exciting
43 0 44 Like to look at sexy pictures
46 0-31 Sexual feelings overpower
58 -0-42 A nude person doesn’t interest
63 0-32 Sometimes a woman should be sexually aggressive
65 0-53 Believe in taking pleasure where found
69 0-32 Protect children from sex contacts
72 0-52 Have been involved in more than one sex affair
at time
74 0-37 All right to seduce a person old enough
76 040 Like to look at pictures of nudes
77 0-31 If chance to see people making love, would take it
78 048 Pornographic writing should be allowed to be
freely published
79 0 43 Prostitution should be legally permitted
81 -0-32 Too many immoral plays on T.V.
83 0-32 Do away with marriage entirely
85 0-41 No censorship on sexual grounds of plays and
films
87 -0-39 Sexual permissiveness threatens to undermine
civilisation

100
 ADULT POPULATIONS

Item Loading Item

Females
89 0-54 Faithfulness in partner is as silly as celibacy
92 0-31 Enjoy watching partner have intercourse with
someone else
95 0-32 Would vote for law permitting polygamy
96 0-35 Though having regular intercourse, good to
masturbate for a change
108 -0-34 Partner satisfies all physical needs
113 0-30 Like partner to be more expert/experienced
116 0-34 Feel like scratching and biting partner
119 0-42 Group sex appeals
120 0-55 Illicit relationship excites
132 -0-37 In sex, tenderness important quality
134 -0-41 Object to four-letter words in mixed company
135 -0-54 Idea of ‘wife-swapping’ distasteful
144 0-45 Not disturbed if partner has sex with someone
else
151 0-47 Accept invitation to blue film
152 0-45 Accept pornographic book if offered
153 0-39 Accept invitation to take part in an orgy
154 0-36 Preference for having intercourse frequently
158 0-34 Strength sexual desire great

Superfactor 2: Sexual satisfaction

Item Loading Item
Males
4 0-52 Am satisfied with sex life
11 -0-47 Have been deprived sexually
15 -0-32 Prefer people of my own sex
18 -0-30 Sex feelings are sometimes unpleasant
19 -0-51 Something lacking in sex life
21 - 0*51 Love life disappointing
31 -0-37 Parents’ influence has inhibited sexually
32 -0-31 Thoughts about sex disturb more than should
44 - 0-31 Conscience bothers more than should
47 -0-43 Feel nervous with opposite sex
50 0-31 Feel at ease with opposite sex
51 -0-32 Don’t like to be kissed
56 -0-34 Worry a lot about sex
58 -0-34 Seeing a person nude doesn’t disturb
60 -0-30 Perverted thoughts have sometimes bothered
65 -0-30 Believe in taking pleasures where found
108 0-32 Partner satisfies all physical needs
113 0-33 Easy to tell partner if like/dislike her love-making
117 -0-37 No one has satisfied sexually
118 -0-39 Feel less sexually competent than friends
124 -0-35 Afraid of sexual relationships
126 -0-37 Can’t stand being touched
133 -0-34 Female genitals are aesthetically unpleasant

101
 SEX AND PERSONALITY

Item Loading Item

Males
139 -0-35 Can’t discuss sexual matters with wife
141 0-30 Naked body pleasing sight
149 -0-34 ‘Girlie’ magazines suggests failure of adult
attitudes
159 -0-34 Inhibiting sexual influences

Females
4 0-57 Satisfied with sex life
11 -0-51 Have been deprived sexually
15 -0-40 Prefer people of own sex
18 -0-32 Sexual feelings sometimes unpleasant
19 -0-62 Something lacking in sex life
20 0-42 Sex behaviour never caused trouble
21 -0-58 Love life disappointing
24 -0-32 Felt guilty about sex experiences
31 -0-44 Parents’ influence inhibited sexually
32 -0-36 Sex thoughts disturb more than should
44 -0-32 Conscience bothers too much
55 0-39 Enjoy petting
56 -0-34 Worry about sex
64 -0-33 Sex jokes disgust
68 -0 31 Sometimes felt like humiliating partner
75 -0-39 Sometimes felt hostile to partner
86 0-34 Sex is greatest pleasure
108 0-57 Partner satisfies all physical needs
109 0-31 Sex is not all
112 0-34 Enjoy lengthy pre-coital love play
113 0-^4 Easy to tell partner like / dislike about love-making
114 -0-45 Like partner to be more experienced and expert
117 -0-33 No one has satisfied sexually
118 -0-41 Feel sexually less competent than friends
124 -0-33 Afraid sexual relationships
125 -0-37 Often wish men less demanding sexually
133 -0-37 Male genitals aesthetically unpleasing
144 -0-30 Not disturbed overmuch if partner had sex with
someone else
145 -0-39 Men more selfish in love-making
146 -0-38 Some forms of love-making disgusting
150 -0-34 In sex, women come off second-best
155 -0-54 Suffered frigidity
157 0-39 Strength sexual desire great
158 -0-42 Influences inhibited sexually

When it comes to the correlation between the two superfactors,

there is a large difference between the sexes. For the men the correla¬
tion is very small (r=0-09) and essentially unimportant; it accounts
for less than 1 per cent of the variance. For the women, however,
the correlation is much larger 0=0-31) and certainly not negligible;

102
 ADULT POPULATIONS

it accounts for 10 per cent of the variance. This difference may be

connected with the ‘excitement’ variable mentioned above; for
women satisfaction and libido are much more closely connected
than for men, for whom there is practically no connection at all.
It would not be helpful to speculate on the reasons for this difference;
obviously further work is required before any confident interpreta¬
tion can be given. Here we merely note the fact of this difference;
it is of importance in trying to assess the contribution of genetic
factors to both libido and satisfaction.
It is possible to devise scales for the measurement of both the
primary factors and also the superfactors. To take the primary
factors first, scales have been devised for eleven out of the twelve;

Table 3.4 Reliabilities, means and standard deviations for eleven sexual
attitudes scales

Reliability Means
M F M F

1 Permissiveness 0-84 0-83 11 -69 ±2-66 10-92 ±2-84

2 Satisfaction 0-82 0-83 8-03 ±2-90 8-79 ±2-84
3 Neurotic sex 0-74 0-72 2-87 + 2-35 2-75 ±2-30
4 Impersonal sex 0-85 0-81 5-91 ±3-56 3-43 ±2-86
5 Pornography 0-78 0-78 6-82 ± 1-71 5-14 + 2-25
6 Sexual shyness 0-72 0-66 0 81 ±1-21 0-61 ±0-99
7 Prudishness 0-58 0-61 1-09+1-18 1-64 ±1-43
8 Dominance—submission — —

9 Sexual disgust 0-54 0-65 1-88 + 1-32 2-20+1-56

10 Sexual excitement 0-66 0-77 700± 1-69 5-47+2-35
11 Physical sex 0-65 0-61 5-23 ±2-01 4-61 + 1-97
12 Aggressive sex 0-47 0-51 2-36+1-42 2-62 + 1-40

there were not sufficient common items for males and females on
the dominance-submission factor to make construction of a scale
possible. Items for the scales were selected in such a way that both
size of loading and common loading for men and women were
considered important. The actual items for the various scales, with
key, are given in Appendix A to this chapter. Table 3.4 shows
the reliabilities of the scales (alpha coefficients), and the means and
standard deviations for men and women separately. Half of the
reliabilities are in excess of 0 7; half the scales have reliabilities
too low for serious consideration, except perhaps for group com¬
parison. It is interesting to note that scales 1 to 6, which were put
in this order in accordance with the factor comparison indices, also

103
 SEX AND PERSONALITY

have the highest reliabilities; thus we have two different reasons

for considering our first six factors to be better established than the
second set of six.
The means of the various scales show interesting sex differences.
Males clearly have higher scores on permissiveness, impersonal
sex, pornography, excitement and physical sex; this is very much
what popular wisdom (and previous studies) would have led one
to expect. Women have higher scores on satisfaction, disgust and
prudishness; these too accord with previous work, although the
high satisfaction scores of the women, as compared with the men,
would not perhaps form part of the popular wisdom (Eysenck
1971a).

Table 3.5 Means and standard deviations of masculinity-femininity scale,

and personality scales P, E, N and L. In brackets: standardisation data for
personality scales

Males Females
Mean: M-F scale 32-93 + 5-93 24-05 ±6-33
P 2-26+1-77( 2-74) 1-60 ± 1 -62( 1-57)
E 10-83 ± 4-46(12-67) 12-04± 3-87(12-43)
N 9-17±4-92( 9-59) 11-48 ±4-66(12-31)
L 5-76±3-66( 6-74) 6-68±3-47( 7-84)

Scales were also constructed for the superfactors of sexual libido

and sexual satisfaction; the items and keys are given in Appendix
B. Yet another scale was constructed to measure masculinity-
femininity, based on the items showing the largest differences
between the sexes; this scale and key is given in Appendix C. This
scale is very similar to an earlier one based on our original sample
of item (Eysenck 1971b). To take the M/F scale first, table 3.5
records the means and standard deviations of the scale for men
and women separately; it will be seen, as of course was inevitable,
that men are more ‘masculine’, their scores being on the average
well over one standard deviation (S.D.) higher than those of the
women. The reliability of the scale is 08 for both sexes; this is very
satisfactory as the scale was constructed by taking items giving the
largest differences between the sexes; there is no a priori reason
why such a method should produce a scale that would be reliable
within either sex. Also given in the table are the means and
standard deviations of the four personality scales used, i.e. psycho-

104
 ADULT POPULATIONS

ticism (P), extraversion (£), neuroticism (AO and the Lie or dis¬
simulation scale (L). The instrument used was an earlier form of
the E.P.Q. and is reprinted in Appendix D, together with a key.
The scale was standardised on a fairly random sample of the
population obtained through a market research firm and is not
likely to differ substantially from a true sample. The reliabilities of
the personality scales on the standardisation group, for males and
females separately, were 0-74 and 0-57 for P, 0-79 and 0-78 for E,
0-84 and 0-82 for N and 0-78 and 0-74 for L.
It will be seen that the sample tested with the sex questionnaire
is not very different, in point of personality, from the standardisa¬
tion sample, although because of the large size of both samples
almost all the differences are statistically significant. For P differences
are minimal. For E, males and to a lesser extent females are some¬
what less extraverted than the standardisation group. For N, males
and even more so females are somewhat less neurotic than the
standardisation group. For L, both males and females have lower
scores than the standardisation group. Our experimental group
consequently contains more stable introverts than would be found
by chance; its members also seem to dissimulate somewhat less (or
are more conformist!). None of these differences is large in absolute
amount.
Table 3.6 shows the correlations between the M/F scale and the
four personality scales. It will be clear that these correlations are
not very large. Masculinity correlates with P for both men and
women (0T0 and 0-20, respectively). Correlations with E are positive
but very small, while correlations with N are contradictory. Mascu¬
linity correlates negatively with L (—0T5 and — 0T8). This is
interesting in view of the fact that men usually have lower L scores
than women (see table 3.5). The direction of these correlations is
therefore in the expected direction, even though the size of the
correlations is rather small. The intercorrelations between the per¬
sonality variables are small throughout, although E and P clearly
show a negative correlation (—019 and —0T9), with E and N
also showing a negative correlation (—0 06 and —0-20) and N and
P a positive correlation (007 and O il). Correlations of L with
the other three variables are all negative. Again, these values are
statistically significant but do not amount to much psychologically;
it is the large number of subjects that accounts for the statistical
significance.

105
 SEX AND PERSONALITY

Table 3.6 Correlations between masculinity-femininity scale and

personality scales: males above leading diagonal, females below

M-F P E N L

M-F —
0-10 007 012 -0-15
P 0-02 - -0-19 0-07 -Oil
E 0-06 -019 - -006 -0-25
N -0-03 0-11 -0-20 - -o-ii
L -0-18 -0-22 -Oil -0-12 —

We may now turn to the Libido and Satisfaction scales. The

reliabilities of these scales are very satisfactory: for the males, they
are 0-90 and 0-82, while for the females they are 0-89 and 0-81.
Correlations are —0-11 and 0 03 respectively.(l) Means on the
Libido scale, as one might have expected, are higher for the males:
24-47, S.D. 6-82, as compared with 19-11, S.D. 6-84 for the females.
This difference is of course fully significant. On the Satisfaction
scale there is no meaningful difference, although the females have
a slightly higher score: the mean for the males is 12-40, S.D. 3-06;
that for the females is 12-70, S.D. 3-02. These results agree well
with our previous work.
Table 3.7 reports the intercorrelations of eleven of our twelve
scales (scale 8, it will be remembered, showed too little agreement
between males and females to make proper construction possible)
with each other, and with the Libido and Satisfaction scales. It will
be seen that both sexes give very similar results, and that the pattern
of correlations makes perfectly good sense. Scales for permissive¬
ness, pornography, impersonal sex, excitement, physical sex and
aggressive sex correlate positively together, and negatively with
prudishness, sexual disgust and sexual shyness, to form the factor
of Libido. Satisfaction, physical sex and sexual excitement correlate
positively together, and negatively with neurotic sex, sexual shy¬
ness, sexual disgust and prudishness to form the factor of Satisfac¬
tion. Figure 3.1 shows the pattern of correlation of these eleven
scales with the two superscales. Libido and Satisfaction. It is note¬
worthy that the two sexes show very similar patterns, without any
marked discrepancies between the positions of the scales in the two-
dimensional framework. Correction for attenuation would give even
higher correlations, but would not change the pattern.
There were 239 pairs of subjects who were married or were
living together on a semi-permanent basis. It seemed of interest

106
 o VO WO l-H VO On Os OO 04 ©
•g C'- 00 r- © 04 CO CO CO
’JS © © 6 © 6 © © © © © © © i
1 1 1 1
3
a
leading diagonal, females below

o
-*-» vo 00 (N CN
o 00 CO vo VO co
cd o On s O wo 04 04 © © 1 ©
cw © 6 © © 6 © © © © © © ©
*£3 1 1 I \ 1 1 1
IA

wo VO wo 1—1 OO © ON Tf co OO
CN (N O C4 (N © *—H
T—< 04 1 © CO
© © 6 © © © © © © © © ©
1 1 1 1

WO 00 o 00 CO wo CO VO Vp
© co fN © co 04 CO 04
V* CO
6 © o 6 © © © © © 1 © ©
©
1 1

OO co co O wo 04 vo
CN O CO VO wo 04 wo
1 04
6 6 6 © 6 © © © © © ©
©
Intercorrelations between sexual attitude scales: males above

! 1 1

ON O © © C4 vo Tf {v.
<N O cs c4 CO 00 CO
ON V1 co
6 © 6 6 6 © © 1 ©
© © © ©
1 1 i l l 1 1 1
1

(N wo OO © ON wo
(S O W~> wo CO o» 04
r-
© 6 © 6 6 © 1 ©
© © © © ©
1 i l i 1 1 1 1
i

O
VO
tT
On
(N
©
o ON © vo VO © vo
VO 1 © 04 © © Tl“ ©
© 6 6 6 6 © © © © © ©
©
i 1 1 1 1

wo r* 04 00 OO CO wo 00 VO OO
Tf s © -<* © Ol CO ©
wo i 04
6 © © 6 © © © © © ©
© ©
1 1 1

o W“> 00 ON Tf o 1^. 04 vo
CN ON »—< 04
.
CO 04 04
6 © o i
© © © © © © © ©
1 6
1 1 1

co 00 wo 00 CO wo © 00 vo
Tt* Tf
P © co © 3 8 04 *o
co
o © 1 6 © © © © © © © © ©
1 1 '

00 OO r- CO © © VO © (N
<N
o © CO 04 04 © ON ©
6 I © 6 © © 6 © © © © © ©
1 1 1 1 1 1 1

00
1 3 8 5 wo
ON
©
04 tT
co
04
04
00
04 ©
ON
Table 3.7

6 6 6 © © © © © © © © ©
1 1 1

G
O
"■4-*
o
a o
U-t
</) 3
"OJ '£
<N co wo VO ON © 04 c3
22 in £

107
 SEX AND PERSONALITY

•2 J
2 Satisfaction

11 Physical sex 10 Sexual excitement

1 Permissiveness
•11 q 5 fbrnographi
.4
- »5 •?_,_V
°12 Aggressive sex
LIBIDO
•1
° 4 Impersonal sex

7 Prudishness «g ■4
□
•7

9 Sexual disgust q Sexual shyness

■ Males
□ 3 Neurotic sex s Females
-.6 *3
SATISFACTION

Figure 3.1 Position of primary sexual attitudes factor scales on two super¬
factors: libido and satisfaction

to study the degree of homogamy present for the eleven scales

constructed to measure the primary factors, and also for the two
scales measuring the two higher-order factors. Libido and Satisfac¬
tion. Table 3.8 gives the correlations for the pairs for these scales.
Also included are correlations for the factor scores on the two super-
factors; this was done to compare the observed relationships for
scales and factor scores. It will be seen that factor scores give very
similar information to that obtained from the scales. There is clearly
a considerable range of assortative mating,(2) with correlations
between spouses ranging from 058 to a low of between 0T3 and
0T5, there being three scales at the low level. Both Satisfaction and
Libido seem to give rise to similar degrees of homogamy; there is
a low degree of homogamy for neurotic sex, sexual shyness, prudish¬
ness, sexual disgust and sexual excitement. These low correlations
between spouses may be of interest to marriage guidance counsellors;
they are likely to provoke difficulties in a close sexual relationship.^)
Our data make it possible for us to determine also the degree of
assortative mating for personality demonstrated by our subjects.

108
 ADULT POPULATIONS

Table 3.8 Intercorrelations between 239 spouses for scores on sex factor
scales

1 0-58 Permissiveness
2 0-47 Satisfaction
3 0-22 Neurotic sex
4 0-42 Impersonal sex
5 0-36 Pornography
6 014 Sexual shyness
7 0-13 Prudishness
8 — Dominance-submission
9 015 Sexual disgust
10 0-20 Sexual excitement
11 0-47 Physical sex
12 0-40 Aggressive sex
Satisfaction 0-46 Satisfaction
Libido 0-51 Libido
F 1 0-50 Factor satisfaction
F 2 042 Factor libido

As this feature is of great importance in genetic analysis, a brief

review of what is known about assortative mating for personality
will be given before we discuss our own results. (See Eysenck 1974
for more detailed analysis.) There are two main lines of thought,
each supported by some data. Early investigators favoured a theory
of homogamy (Burgess and Wallis 1944). In contrast. Winch (1958)
elaborated a theory of complementary needs in the marital rela¬
tionship. The heterogamy position has received some support
(Schellenberg and Bee 1960, Gray 1949, Tharp 1963, Roos 1956,
Reiter 1970); the homogamy position has equally received some
support, usually indicating that ‘there is homogamy for psycho¬
pathology’. Kallman and Mickey (1946), Kreitman (1964, 1968),
Murstein (1971) and Yom and others (1976) may be cited here.
1 Other writers again find that their data are not in conflict with
a theory of panmixia: Udry (1967), Bowcrman and Day (1956), Hill
(1968) and Gottesman (1965) might be quoted in this connection.
Cattell and Nesselroade (1967) and Pickford and others (1967) cite
data in support of the contention that happy marriages are charac¬
terised by homogamy - a point that is interesting but not relevant
to the genetic aspects of assortative mating unless happy marriages
produce more children (Eysenck 1974). A consideration of all the
available data suggests no departure from panmixia, except with
respect to psychiatrically abnormal persons where a small amount
of homogamy appears to be present.

109
 SEX AND PERSONALITY

Studies with the P, E, N and L have on the whole supported

this view. In our Department, Nias (1974) studied 700 pairs of
parents who had 1,200 children, and Insel (1971) made a thorough
study of 98 families, including 589 subjects in all, involving in
every case three generations. The correlations found by them were
quite insignificant (and indeed opposite in sign) for E (0 06 and
—0T0) and N (0T2 and —0-09). For P, however, both report signi¬
ficant positive correlations (0-22 and 0-36, respectively). The data
from the present study are as follows: P, 0T42; E, 0 061; N, 0-223;
L, 0-174. The correlations for the two psychiatric variables (P and
N) are both significant, as is that for L. (L is usually negatively
correlated with both P and iV; this may account for the observed
small positive correlation.) For E the correlation is quite insignifi¬
cant. There is thus the suggestion of a very mild degree of homo-
gamy in normals for P and possibly N.
We may turn to a discussion of the relationship between sexual
attitudes and personality. In the previous chapter we have shown
that there are consistent and predictable differences in sexual atti¬
tudes between subjects having certain personality scores on the
personality dimensions P (psychoticism), E (extraversion) and N
(neuroticism). These studies, as well as the German study by
Giese and Schmidt (1968), were carried out on students, many of
whom had only very little experience of sexual matters; it seemed
important to investigate in our present older and much more ex¬
perienced sample whether similar results to those observed in the
earlier study could be found. The method used in this study is
rather different from that used previously, when factors were corre¬
lated with personality scale scores. The reason for preferring a
different method in the present study is simply that not all the items
could be used in factor-analysing our results, owing to limitations
imposed by the capacity of the computer; consequently there would
have been a considerable waste of information had we correlated
the factors with the personality variables. What was done instead
was to correlate each item, for males and females separately,
with scores on the P, E, N and L questionnaire variables; in this
way all items could be considered. We also correlated all attitude
items with social class, occupation, age and marital status; in this
way it was hoped to discover attitude differences between different
class and age groups. The detailed findings are recorded in table
3.9, this should be consulted for all details. However, in order to

110
 ADULT POPULATIONS

| give a more cohesive account of attitude differences relevant to

i individual personality factors, and to age and social class, a verbal
[ picture will be given below of the highest correlations in each
) column. No correlations below the value of + 0T2 have been used,
5 although in cases where there were very many items exceeding this
value we put the lower limit of acceptance slightly higher. Correla-
f tions are low on the whole, but this is not remarkable in view of the
1 fact that single items are always rather unreliable; no very high
) correlations could have been expected. With the numbers of sub-
t jects involved the correlations included would nearly all be signi-
I ficant, although of course in view of the large numbers of items
I involved ordinary standards of significance cannot reasonably be
i applied to single items.

Table 3.9 Sex inventory item correlations with personality, class,

occupation, age and marital status

Item Class Occupation Age Marital P E N L

Status
Males
1 - 0003 -0059 0-198 0002 -0-131 0 007 0-036 0-210
2 0-125 -0149 0-078 0-067 -0-102 -0076 0065 0105
3 -0030 0-032 0-096 - 0-072 -0-010 -0-026 0089 0-039
4 0005 -0-016 - 0-084 -0-200 0013 0-017 -0-221 - 0029
5 - 0-047 0-015 0-088 0-157 -0-054 0004 0-169 0-113
6 0-064 0031 0-115 0102 0 111 0018 -0-016 0064
7 - 0-025 0-022 - 0-051 - 0-004 -0-074 0149 0-218 -0-169
8 -0-055 0-038 0118 - 0-050 0039 - 0055 0093 0-084
9 0-009 0196 - 0-039 - 0079 - 0-009 0011 0065 0014
10 - 0-039 0-037 - 0-151 - 0-092 -0001 0051 -0011 -0-212
11 -0-021 -0-055 0-150 0-008 0-144 - 0075 0173 0-024
12 0-020 0017 0-003 0074 -0169 0-014 0-118 0-020
13 -0050 -0-100 0003 -0068 0-097 0026 -0015 -0-142
14 -0092 0-030 0-030 - 0-092 0-080 - 0086 0133 0081
15 0023 -0011 0-016 0-158 0-088 - 0076 0 007 - 0-062
16 - 0-028 0065 - 0-062 0 026 - 0-038 0-209 - 0071 -0 005
17 0-008 -0175 0-140 0-125 -0-162 -0113 0-114 -0016
18 - 0045 -0111 0-009 0-048 -0-027 -0091 0-108 - 0-049
19 - 0-061 0001 0-080 0-070 0-027 -0-107 0-230 0-027
20 -0-066 0-181 - 0-079 - 0-095 - 0049 -0063 -0-120 0-237
21 - 0056 - 0025 0-129 0-130 0-085 -0 096 0-221 0-015
22 0-013 0021 0-076 - 0059 0012 -0-317 0-087 0-264
23 0 009 0-033 0042 - 0-030 - 0-079 0-035 0010 0087
24 0-017 -0-232 0-105 0-032 -0-028 -0021 0-238 -0-126
25 0-068 0-023 - 0-071 - 0-223 0-109 0015 -0051 -0-139
26 -0-030 0-032 - 0-075 -0033 0-026 0-097 0-261 -0-119
27 0-010 0-006 0000 0096 -0-032 0094 0014 -0132

111
 SEX AND PERSONALITY

Item Class Occupation Age Marital P E N L

Status

Males
28 -0-082 0-012 0 055 0078 -0033 0-297 0 048 -0-124
29 -0-026 0046 0063 - 0 041 0-115 -0-160 0094 0-112
30 -0-152 0030 -0-168 -0-068 -0-010 0-068 0039 - 0-042
31 0-055 -0-105 0-259 0030 0110 -0-128 0185 0077
32 -0-028 0-001 0-071 - 0079 -0031 - 0-028 0-340 0-085
33 -0-054 0-102 0083 -0021 - 0-078 0049 - 0059 -0-128
34 0-019 -0196 0-124 0-054 - 0-020 -0-013 0-062 0028
35 0031 0-067 -0-005 -0120 - 0069 0112 0-015 0017
36 0-006 0-058 - 0-099 -0088 - 0-034 0090 0018 -0017
37 -0016 -0-015 -0-233 - 0-098 -0010 0-173 0043 -0-092
38 -0-012 -0-106 0-148 0-129 - 0-057 -0063 -0012 0-212
39 -0-179 0-033 -0-217 - 0077 - 0-055 0-217 0166 -0096
40 -0-097 -0-167 0-185 0-214 -0-143 - 0068 - 0059 0-209
41 -0087 -0-133 0-185 0-132 -0-137 -0-137 -0026 0-239
42 -0119 0-218 - 0-074 - 0-020 0-043 0059 0-204 - 0-048
43 -0-009 0-115 -0066 -0118 0021 - 0-099 0118 -0-110
44 0-021 -0060 0-027 0-088 - 0060 -0016 0-275 -0097
45 0-073 -0-073 0-142 - 0-006 0113 -0-170 0-099 0130
46 - 0-070 0091 -0-215 -0-038 0048 0-155 0-267 -0-193
47 0-043 -0-053 -0-035 -0-058 0129 -0-187 0-233 0-032
48 -0-072 0-071 -0-040 0-020 -0071 0-013 0-227 0-042
49 -0-057 0-162 - 0-051 0009 -0-024 0-009 0-260 0011
50 -0-007 -0018 0-038 -0031 -0-041 0-199 -0-121 -0 073
51 0-009 0-067 0-120 0-149 0-083 - 0-079 0056 0093
52 - 0-006 0-071 -0-038 0-081 -0026 - 0-200 0 021 0-113
53 0-011 0-032 0081 - 0-055 -0001 -0127 0-158 0 005
54 0-003 0126 -0017 0-072 -0-001 - 0092 0-075 0111
55 -0 066 0066 -0161 0-006 - 0-042 -0-144 0-092 - 0028
56 0-121 -0-003 0 003 0-025 -0-073 -0-044 0-330 0039
57 0006 0-163 - 0093 - 0082 0-083 0-017 0-062 -0-136
58 - 0-037 0-013 0-136 0-085 0-104 0-051 -0031 - 0-005
59 -0-142 0-174 - 0003 0055 - 0-074 0059 0-234 -0068
60 0020 - 0-034 0-027 0-016 0000 -0-011 0-185 -0-094
61 0110 -0-103 - 0079 - 0088 -0-021 -0-175 0-029 -0013
62 -0-106 0-048 -0120 -0095 0066 - 0-001 0-053 -0-156
63 -0-061 0-067 -0-081 0028 0-067 0-056 0-031 -0-157
64 -0-019 -0079 0082 0128 - 0032 - 0-065 -0-002 0-102
65 -0-080 0-222 -0013 -0-158 0143 0-199 0-177 -0-158
66 -0-077 0-014 -0-018 0025 0016 0-024 0-122 - 0-090
67 -0-023 0-118 - 0137 -0149 0028 0-066 -0-087 - 0161
68 0-060 0-051 0023 0-002 0-011 0-095 0-229 - 0-202
69 -0-061 0089 0050 0-112 -0069 0-063 0-171 0-156
70 0-052 -0-154 0-032 -0-082 0-021 -0-047 - 0-020 0-067
71 -0-091 0-061 0-060 -0-037 -0137 0-112 0118 0-004
72 -0-008 0049 0-163 -0 056 0-091 0-186 0086 - 0-262
73 -0-015 0045 - 0-044 -0-058 -0 044 0086 -0-000 -0056
74 -0039 0-118 -0-177 -0-107 0141 0060 0-108 -0-210
75 0-021 0-000 -0-065 -0-043 - 0000 0-048 0-311 - 0-225
76 0-105 -0-037 - 0-042 -0-123 0004 - 0-078 0028 -0-012

112
 ADULT POPULATIONS

Item Class Occupation Age Marital P E N L

Status
Males
77 - 0082 0-113 -0-103 -0-141 0-063 - 0036 0-095 -0092
78 - 0-027 0-090 -0114 -0-084 0192 -0011 0-093 -0194
79 - 0151 0-092 0029 -0-073 0-082 0025 0039 -0121
80 0-055 0-200 0-206 0-005 -0033 - 0-026 0-045 -0-002
81 -0-081 -0115 0-110 0131 -0-111 0018 -0033 0071
82 -0-127 0-187 - 0-079 -0-014 0-011 0-160 0-086 -0061
83 0-044 -0033 -0-095 -0016 0-281 0-117 0058 -0 092
84 0-062 0-163 -0019 - 0069 0-106 0-117 0-211 - 0057
85 0 005 0-055 0-027 -0-108 0-107 -0-031 0030 -0131
86 -0-156 0-196 -0025 -0-102 -0-082 0070 0-251 0034
87 0-014 - 0-024 0137 0-208 -0054 - 0060 0044 0-200
88 - 0-092 -0-028 -0 009 - 0-024 -0073 0-045 -0 053 0-044
89 -0-033 0-122 -0-004 - 0-079 0-129 0168 0094 -0-139
90 - 0035 -0014 0-013 0042 0-099 -0055 0-000 -0041
91 0-003 -0-025 - 0-062 0-045 -0094 0031 0-121 0060
92 - 0-065 0-212 0-077 0-000 -0-019 0-046 0-043 -0025
93 - 0 046 -0-013 0102 - 0-044 0-143 -0028 -0180 -0010
94 -0-102 -0-002 0-069 - 0-026 -0024 -0019 -0012 0-143
95 0-046 0-099 -0-001 -0031 0-142 0 097 0 001 -0055
96 --0-005 0-212 -0-097 0029 0-036 -0017 0094 -0082
97 -- 0-007 0-125 -0-002 0-024 -0006 0098 0104 0047
98 0 000 0-000 0 000 0000 0 000 0 000 0000 0 000
99 -- 0 089 0-250 -0-158 - 0086 0-128 0-072 0-178 -0012
100 0018 0-085 0011 - 0041 -0086 -0-105 0-141 0-010
101 --0009 -0 042 0 007 -0-033 -0 090 0-145 0140 -0-088
102 0-019 0-144 -0006 -0-027 -0-034 - 0-022 0-099 0000
103 0-005 0001 0118 -0012 - 0-022 -0-186 0-129 0120
104 0-104 -0-149 -0-093 - 0-069 0-032 0099 0-174 - 0-055
105 0 000 -0-070 -0 063 -0 097 - 0057 -0-020 0-022 0-058
106 --0-186 0-054 -0-019 -0-070 0 006 0 004 0-183 0-029
107 0-027 0011 - 0-049 0-114 -0-067 0058 0-107 0-090
108 --0019 -0-064 - 0-030 -0059 0-040 0089 -0-142 0046
109 0111 -0-021 0114 0-112 0-108 -0125 -0149 0039
110 --0-121 0-060 - 0164 0-002 0-136 0044 0083 - 0-206
111 0-046 0.103 0-133 0-028 - 0045 0017 0-092 0080
112 -- 0-072 0-116 0-076 0008 -0-105 0144 0-101 0062
113 -- 0-069 0-088 -0188 -0-233 0070 0-120 -0015 -0-026
114 -- 0-054 0-112 0040 0-060 0078 -0013 0-048 0068
115 -- 0-023 -0163 -0 093 0056 -0012 - 0-093 0009 -0034
116 0070 - 0048 -0-249 -0113 0-041 0-216 0-094 -0-219
117 --0-063 0-082 - 0-075 -0-086 0094 -0097 0-102 0030
118 -- 0-076 0-117 - 0045 -0089 0019 - 0006 0116 -0 008
119 --0-163 0-324 -0034 -0060 0061 0-129 -0033 -0-132
120 - 0-131 0-117 0-048 - 0-040 0009 0-025 0-230 -0-083
121 --0-123 0-158 -0-144 -0086 -0-018 0-250 0170 -0-151
122 0037 0015 0-003 -0 021 -0066 0-076 -0-021 -0040
123 0-040 -0-151 0-145 0-128 -0-072 - 0002 0099 0049
124 0006 -0-068 -0014 0-003 0-048 -0-152 0-153 0050
125 - 0-058 0-138 0-033 0 073 0-083 0-122 0198 -0 026

113
 SEX AND PERSONALITY

Item Class Occupation Age Marital P E N L

Status

Males
126 -0-043 -0015 0-078 0026 0169 -0159 0101 -0016
127 -0-066 0-165 0-074 -0010 - 0020 -0013 0150 0-064
128 -0-074 0-248 - 0-136 0-053 0047 0-085 0-160 0-005
129 -0-109 0138 - 0067 -0-042 0-061 0-050 0-144 0-063
130 0 057 -0022 0119 -0106 0-082 -0019 - 0129 0-087
131 0-012 0-041 0087 0-098 0-020 0105 0-239 0-029
132 0-003 -0-087 0076 0-109 -0132 - 0-048 0101 0114
133 0-153 -0-112 0-047 - 0 069 0-044 -0-115 0113 0-031
134 -0100 - 0-078 0-075 0-046 -0-194 - 0078 0001 0-206
135 0-152 - 0-196 0 053 0-152 - 0 056 -0017 0-045 0055
136 - 0040 0-120 0-106 0-010 0146 -0-038 0096 0 099
137 0-025 -0036 0-139 0-208 -0014 0-013 0169 0051
138 -0-022 0-045 - 0-035 -0-124 0-071 0-015 0064 -0-090
139 0041 -0088 0-104 -0-012 0080 - 0036 0-175 0061
140 0-046 -0089 0-208 0-083 -0002 0 032 -0-093 0-039
141 -0 041 0104 - 0107 -0 071 - 0057 -0-034 0-054 -0006
142 -0-018 0-028 0-122 - 0-001 -0012 0049 -0123 0-069
143 -0059 0-066 0031 -0051 - 0073 0-094 0050 0-001
144 -0-016 0137 - 0 005 -0041 0120 -0014 -0019 - 0048
145 - 0063 0011 0-003 - 0032 -0-122 0-087 0-065 0-005
146 0-041 - 0-201 0080 -0009 0056 -0 076 0-025 0-003
147 0080 -0-127 0-155 0 001 -0088 0-041 - 0-098 0-135
148 -0-060 0121 0031 -0-044 0003 0-022 0-202 0-055
149 0-158 -0172 0-307 0-140 0064 -0-175 -0 005 0-254
150 -0-075 -0-015 0091 0067 0001 0-014 0-100 0062
151 0010 0-093 - 0-171 -0155 0-044 0-028 0100 -0-186
152 -0008 0-089 - 0-196 - 0109 -0-006 -0001 0-088 -0-177
153 -0056 0-193 0-016 - 0094 0-108 0148 0-091 -0-127
154 - 0083 0-065 - 0-300 -0107 0036 0121 - 0-036 -0129
155 0-129 -0-064 0-055 0-077 0103 0-085 0-101 -0-100
156 0119 -0-239 0-351 0-106 - 0-065 -0-310 0056 0-305
157 -0-168 0-211 - 0-265 -0-106 - 0-085 0-227 0-212 - 0-033
158 0-033 0-091 0-162 0 090 -0016 - 0-058 0102 0-132

Females
1 0-033 -0-021 0107 0066 -0-144 - 0-076 0-184 0-180
2 0010 -0041 0-201 0014 - 0 -109 -0-051 -0010 0-138
3 0-152 - 0066 0-153 0-042 -0-087 - 0-050 0-244 0-152
4 - 0-091 0-041 -0040 -0130 - 0-030 0-025 -0-137 0-084
5 - 0-005 0026 0-252 0-177 -0-117 -0-065 0-062 0-269
6 - 0029 0-061 0-275 0-094 -0-011 -0 004 -0-017 0101
7 0043 - 0092 -0-127 -0053 0-079 0-096 0-104 - 0-068
8 0-079 -0-119 0 099 0-023 -0010 -0-035 0-024 0-060
9 - 0-028 -0-018 - 0-201 - 0-057 0-139 0-156 -0 079 - 0-011
10 - 0061 0-115 0-033 0-007 0-044 - 0-079 -0027 -0051
11 0-055 -0-140 0-244 0091 0-042 -0-074 0-139 - 0-029
12 - 0-054 0116 0-039 0-113 0-107 -0-107 0173 0-127

114
 ADULT POPULATIONS

Item Class Occupation Age Marital P E N L

Status
Females
13 -0-101 0-134 0-002 0-064 0-015 0-035 0-051
- - 0-045
14 0-064 -0-007 0-007 - 0-061 0-109 0-203
- 0-135 0084
15 -0019 0057 0-069 0-012 0-042 0008 0101 -0031
jT6 -0-129 0-095 -0-059 0-004 -0060 0199 -0-178 0-133
17 - 0045 -0-041 0-140 0-127 0068 -0-136 0-139 0010
18 0-047 0-015 - 0 042 0-071 0012 - 0-057 0-263 -0101
19 0-140 - 0-085 0-143 0-218 0-039 - 0-033 0-192 -0-101
20 -0-087 0-052 -0056 -0-100 - 0066 0-095 - 0-130 0-092
21 0097 -0-014 0-100 0-032 0-092 - 0092 0-142 -0049
22 0-094 -0-137 0-072 -0-122 0-028 -0-197 0040 0-055
23 -0107 0045 0-205 0075 0-034 -0-051 0-101 - 0-003
24 -0-024 -0067 -0019 0-013 0-137 -0080 0-208 - 0066
25 - 0024 0-054 -0-180 -0-182 0-179 - 0-054 0-126 -0-182
26 0-023 - 0041 -0-097 -0-017 0100 0101 0-121 -0-089
27 - 0-072 0017 -0-067 0-021 0101 -0 028 0-071 -0-074
28 0-014 -0-147 -0-041 0-104 -0-108 0-345 -0-117 -0 044
29 -0-135 0-103 - 0-002 0-033 0-047 -0-102 0-260 -0095
30 - 0-064 0-054 -0-159 -0-075 0-076 0-128 0-050 -0-068

31 0-075 -0-148 0-128 -0-068 0-058 -0-171 0-203 0-018

32 -0099 0-048 0-057 0 005 0106 -0048 0-148 0 027
33 - 0-044 0050 -0096 0-019 0193 0021 0-063 0-029
34 -0-007 0-094 0-174 0073 -0171 -0193 0-167 0-050
35 -0-099 0055 0-090 0019 - 0-009 -0036 0018 -0059
36 -0-140 -0-023 -0-102 -0-114 0-038 0-070 - 0-069 -0-153

37 -0 067 0-039 -0-216 -0014 0000 0098 0-010 -0160

38 0-101 -0-036 0-269 0-164 -0 091 -0-088 0-064 0080
39 -0-102 0-096 -0-215 - 0-034 0-132 0166 0-082 -0101
40 0-027 -0-062 0-209 0045 0-007 - 0-054 -0-030 0-195
41 0-049 -0-132 0-331 0-158 -0-163 - 0-071 0-075 0-158
42 -0-087 0-182 -0-143 - 0-049 0 022 0-057 0036 - 0084

43 -0-152 0-260 -0-195 -0060 0-043 -0096 0-156 - 0-052

44 -0-003 0-092 0 090 - 0 029 0-047 - 0-042 0-295 0-065

45 - 0-084 0-046 - 0-057 - 0003 -0-015 -0-022 0-133 -0001
46 0-003 0022 -0-108 -0-123 0067 0-053 0-100 -0-138

47 0-068 0109 0044 -0-015 0142 -0-170 0-232 - 0032

48 0-094 0030 -0-139 - 0-004 -0023 0098 0089 0-022

49 -0-160 0-112 0-076 0017 -0-124 0 047 0-200 0110
50 0-020 -0169 0-003 -0012 -0-106 0-176 -0144 0066
51 0-026 - 0-039 0064 -0-034 0114 -0110 0095 - 0-007

52 -0-034 0-171 -0-017 0-003 0100 -0-232 0-190 0-037

53 0-063 0095 0-110 0068 0019 - 0096 0-200 0074
54 - 0 092 0-123 0-108 - 0025 0089 -0 098 0-101 -0012
55 -0005 -0-134 -0-190 0-026 -0024 0-124 -0-033 -0-112

56 0-067 -0-030 -0-030 -0-006 0-008 -0114 0-246 - 0-002

57 - 0-006 0-025 - 0-069 0-048 0010 0-027 0059 -0135

58 0103 -0-160 0-171 0-037 - 0-070 0-104 - 0050 0091
59 0025 -0-015 0019 0-068 0066 -0003 0-272 -0-121

60 0-010 0-060 0-051 0-036 0097 -0-115 0-181 0-014

61 0-116 0009 0-153 0-185 0027 -0-131 0-158 0014

115
 SEX AND PERSONALITY

Item Class Occupation Age Marital P E N L

Status

Females
62 -0-074 -0039 -0122 -0-160 0-079 0122 -0-118 - 0-185
63 -0-114 0032 -0136 -0-049 - 0068 0129 - 0090 - 0-052
64 0035 -0-125 0-157 0086 0032 -0130 0020 0-107
65 -0-135 0-135 -0-143 - 0027 0-109 0-121 -0014 - 0-054
66 -0-102 -0-046 -0-149 - 0-025 0045 0097 - 0063 -0-120
67 - 0-008 0-027 -0026 0016 -0078 0094 -0-127 0043
68 0-047 0-037 -0-109 0-055 0-208 - 0-058 0-261 - 0-233
69 -0-005 - 0-057 0-228 0-089 - 0067 - 0-002 0-065 0-134
70 -0-021 -0-138 - 0-046 -0144 -0-164 0-108 -0-187 0-004
71 -0034 -0061 -0039 0-060 0049 0042 0050 0-002
72 -0-042 0-038 0-047 0080 0-091 0069 - 0-026 - 0-068
73 -0099 0-007 -0-100 -0025 -0036 0-049 0-007 -0-096
74 - 0-054 0-032 - 0177 0-021 - 0059 0076 0026 - 0049
75 0-070 -0-115 0018 -0-014 0-111 - 0-069 0-250 - 0193
76 - 0148 0-237 -0-118 - 0-073 0003 -0019 0-105 -0-008
77 - 0132 0-277 - 0-220 0-023 0017 -0061 0048 -0015
78 -0029 0-033 -0-187 -0-082 0063 0053 0043 - 0-248
79 -0011 - 0073 0099 -0-058 0077 0-024 -0-127 -0-119
80 -0-065 0-136 0103 -0002 - 0-007 -0-008 -0038 0-167
81 0-076 -0152 0-288 0-128 -0041 -0001 - 0-004 0-082
82 - 0003 0-200 -0015 - 0-007 -0051 -0-068 0-056 0-279
83 -0038 0-097 - 0050 -0-009 0-385 0-059 0-217 - 0-079
84 - 0-023 0-087 -0-113 -0016 0084 0 007 0-181 -0-152
85 - 0044 0-018 - 0-048 - 0-042 0-088 - 0-029 -0-034 -0-192
86 -0-158 0-099 -0-027 - 0-095 -0057 0-017 - 0-034 0-049
87 -0 009 - 0-044 0-229 0-101 - 0-062 -0-012 0-103 0169
88 - 0090 - 0026 0-058 0-153 -0-165 -0032 0-082 -0-036
89 -0-137 0-124 -0027 0-047 0-088 0098 0038 -0-077
90 0-028 - 0-051 0-226 0-087 0010 -0-130 0-040 0-128
91 0019 - 0-028 0-078 0017 -0-028 -0048 - 0-022 0069
92 - 0-029 0190 -0046 0001 0 046 - 0-044 0-012 0-003
93 0028 -0003 0-024 - 0069 0032 0-049 - 0-094 0 000
94 -0006 0-041 0-186 0047 -0-151 -0053 0-039 0-243
95 0-083 0-077 - 0048 0-044 0-281 0-046 0-027 -0178
96 -0-104 0-154 - 0-029 -0-038 0004 - 0-024 0-093 -0061
97 -0128 0070 -0-044 -0 045 0-037 0032 0119 - 0-063
98 -0-013 -0019 0-181 0-136 -0014 -0-107 0-070 0-093
99 - 0-078 0-111 - 0-044 - 0006 0-036 - 0-024 0-175 0027
100 0-047 0-145 - 0039 - 0-003 0-052 0055 0-093 - 0-089
101 -0-138 0-013 -0175 -0-078 0-104 0-139 0-052 -0131
102 0-009 0-102 -0031 -0-006 0063 0-125 0-107 - 0-086
103 - 0058 0-000 0-257 -0-021 0 061 -0-114 -0021 0-133
104 0-024 0-056 0-097 0-084 -0-120 -0-139 0-178 0-140
105 -0-014 0-037 -0011 0-005 0040 0-032 0-000 -0129
106 -0-121 - 0002 -0-061 - 0062 -0-048 0-082 - 0106 0-064
107 0-101 -0030 0036 -0-088 -0-170 - 0-024 0-005 0-110
108 -0-134 0-115 - 0-043 -0123 -0073 0-094 -0115 0-138
109 0-093 - 0-052 0-185 0099 -0017 -0-050 0094 0150
110 0 009 0-132 - 0007 - 0 005 0075 -0-037 0-181 0048

116
 ADULT POPULATIONS

Item Class Occupation Age Marital P E

Status
Females
111 -0-072 0-094 0-169 0-025 - 0088
- 0-067 0095 0-100
112 -0-120 0-018 -0145 0-047 - 0-060 0095 -0046 - 0-071
113 - 0-022 -0010 -0-018 -0015 0029 - 0-003 0118 -0-123
114 - 0-078 0-094 -0-185 -0019 0-043 0-050 -0-143 0029
115 0-151 - 0-266 0-063 - 0-075 0060 - 0044 -0-121 0005
116 -0112 0-058 -0-357 -0-164 0-115 0051 0-142 -0199
117 0-064 -0066 - 0-009 0104 0-050 -0-168 0-088 - 0-084
118 0040 -0-107 0098 0-045 0-040 - 0-246 0-240 0-004
119 0-020 0-114 - 0 091 0-070 0047 0-017 0-032 -0-172
120 -0-137 0-058 -0-105 0038 0106 0-135 0-144 -0-136
121 -0-101 0-132 -0161 -0081 0093 -0-071 0-204 -0-095
122 - 0020 -0122 0-000 0-025 0013 0-050 0-055 -0-102
123 -0-015 -0-018 0-134 0-204 0014 - 0-020 0-038 0-023
124 - 0-024 -0011 0-060 - 0007 0108 -0-091 0191 - 0069
125 0-094 0-101 0-167 0-004 0-025 - 0-090 0-225 0-055
126 0-030 -0-038 0-006 0007 0053 -0-162 0-113 0-094
127 -0-067 0-047 0042 -0-152 - 0040 0000 0-096 0161
128 -0021 0-058 0-226 0-011 - 0-074 -0-018 0-008 0-097
129 0006 0-124 -0110 - 0-073 0089 -0-013 0-143 0-006
130 0030 -0013 0037 0-020 0-032 0-136 - 0046 0-143
131 - 0-072 - 0-005 - 0-025 0-004 - 0-020 0-111 0111 0045
132 0014 -0014 0-106 - 0-055 -0062 0-018 0-161 0-144
133 0-157 -0-019 0-208 0-125 - 0-023 -0046 0-069 0-119
134 0013 0-185 0141 0125 -0-145 -0090 -0039 0-249
135 0-002 -0-088 0-101 -0-009 -0051 -0-063 0070 0-114
136 - 0069 -0010 0083 -0-019 0-216 0010 0034 0097
137 -0-049 0-120 0-143 0-064 0-016 0-118 0-026 0012
138 -0-068 0022 -0036 0-012 -0022 0-184 0-023 - 0-079
139 0-058 0110 0151 0-136 0-058 -0-048 0-109 0010
140 0-111 - 0-058 0-251 0-077 0-024 0011 0-029 0046
141 -0061 0093 - 0-041 - 0-034 -0-032 0-142 -0148 0014
142 0-040 0-048 0-267 0-169 -0-082 0036 -0-012 0059
143 0-048 0-042 0-062 0-042 -0-121 -0-041 0-146 0-138
144 -0-025 0036 - 0-032 0030 0-130 0061 -0-012 -0095
145 -0048 0054 0-171 0-155 -0-004 -0029 0-124 -0-016
146 0-008 0 034 0-251 0110 -0-124 -0-100 0110 0-053
147 0-122 - 0-047 0-115 0-039 - 0024 - 0-070 0-084 0-129
148 0-144 -0041 0-106 0001 0009 - 0-094 0034 0-052
149 0113 -0-214 0-278 0-030 0073 0-012 -0-029 -0-067

150 0-035 0-013 0192 0-055 -0-009 - 0106 0-209 -0066

151 -0069 0-062 -0-143 - 0-053 0-105 0-097 -0 005 -0-150
152 - 0 110 0090 -0-126 -0-080 0-115 0007 0-022 -0-160
153 - 0023 0-093 -0-126 0-034 0-047 0-140 - 0002 -0-064

154 0015 -0057 -0-389 - 0-098 0-026 0131 -0-056 -0-163

155 0-133 -0-105 0-101 - 0-034 0-150 -0109 0-255 -0-203

156 0-104 -0-165 0-513 0-139 - 0-247 -0-112 -0-180 0-167
157 -0-062 0-109 - 0-196 -0-116 - 0066 0-225 -0015 -0-062
158 0-126 ■-0023 0-234 0022 0-013 - 0073 0-164 0015

117
 SEX AND PERSONALITY

In order to demonstrate that the patterns of correlations are

meaningful, the following method was adopted. Extraverted men
and women should show similar patterns of correlations over the
items; so should introverted men and women. The same argument
applied to the other columns; if men and women agree on the
pattern of item correlations, then we may confidently assume that
the results are significant and probably meaningful. Accordingly,
product-moment correlations were run between men and women
for each of the eight columns in table 3.9, with results shown in
table 3.10.

Table 3.10 M-F correlations for personality scales, social class,

occupation, age and marital status over columns in table 3.9

M-F
correlation

Social class 0-52

Occupation 047
Age 0-71
Marital status 0-44
Psychoticism 0-74
Extraversion 0-57
Neuroticism 0-55
Lie scale 0-64

It will be seen that all the correlation coefficients are highly

significant, indicating that men and women of given personality
have similar attitude patterns. This seems to justify our procedure,
and we will next proceed to give a coherent picture of the sexual
attitudes characteristic of different personality types. By ‘coherent’
in this connection is meant simply that an attempt has been
made to put together similar items, even though in order of correla¬
tion they might not be very close together; all details are, of course,
contained in table 3.9, to which serious readers are referred. Many
of the conclusions are borne out by items having correlations slightly
below the arbitrary value of 0T2 which was chosen to exclude items
with lower coefficients; while inclusion of these items would have
strengthened our case, no effort has been made to round out the
picture in this manner.
Let us consider first of all subjects with high P scores, as opposed
to subjects with low P scores. Male high scorers, first of all, are
clearly opposed to current morality and customs. They wish to

118
 ADULT POPULATIONS

do away with marriage;(4) they favour polygamy; they believe it is

all right to seduce a person old enough to know what she is doing;
they consider premarital sex all right. High scorers have a realistic,
not an idealistic, view of sex. They believe that romantic love is an
illusion, they don t think it disgusting to see animals having sex;
they believe in taking pleasure where they can find it; they believe
that tenderness in sex is not important, and that faithfulness in
marriage is silly, they would not be disturbed if their sex partner
had sex with someone else. High scorers are very permissive. They
believe that pornography should be freely available; they don’t
object to four-letter swear words being used in mixed company;
they are not disturbed by seeing necking in public; they don’t think
sex orgies are disgusting; they consider sex play among children
to be harmless; they do not believe that the opposite sex will res¬
pect you more if you are not too familiar with them; they do not
believe that men are more selfish in their love-making than women
- although they also believe that most men are sex-mad. When
it comes to personal history and feelings, however, there is a curious
contrast with this tough-minded facade. High scorers can’t stand
people touching them; they claim to have been deprived sexually;
they piefer sex partners several years older; they feel nervous with
the opposite sex.
Females have similar patterns. They too wish to do away with
marriage, opt for polygamy, and consider premarital sex all right.
They too believe that romantic love is an illusion, don’t object to
the use of four-letter swear words in mixed company and are not
disturbed by their sex partner having sex with someone else; they
don’t believe that the opposite sex will respect you more for being
not too familiar. In addition, they hold certain other views which
help to round off the picture. They do not consider virginity im¬
portant; they are not keen on babies; they had their first intercourse
early; they believe that sex is for personal pleasure, not only for
reproduction, and that females have equally strong sex drives as
males; they get excited sexually very easily. These items all refer
to impersonal, physical sexual satisfaction. In agreement with this
preference is their endorsement of so-called ‘perverted’ practices.
There is nothing they would not do with anyone; they could do any¬
thing with the person they love; there is nothing they do only to
please their sex partner; no form of love-making is disgusting to
them. Interestingly enough, such women have felt like humiliating

119
 SEX AND PERSONALITY

their sex partners, and are attracted by people of their own sex.
They have not suffered from frigidity, but have felt guilty about sex
experiences, and feel nervous with the opposite sex; here again we
find this curious contradiction between strong permissive sexuality
and personal feeling. All in all, high P scorers emerge as advocates
of impersonal, permissive sexual practices, the abandonment of
social rules and laws concerning marriage and the other aspects of
sexual behaviour, and an ‘all’s fair in love and war’ attitude; one
might be tempted to call this the Don Juan syndrome. It is interest¬
ing to note the absence of items signifying satisfaction; if anything
there is some indication of nervousness and even guilt. The total
picture is very similar to that found in our previous research, and
confirms results there reported on a quite different sample.
We next turn to the attitudes endorsed by male extraverts, as
opposed to introverts. First of all, they are clearly able to get on well
with the opposite sex. Extraverts have many dates; have many
friends of the opposite sex; sex contacts have never been a problem
for them; they feel at ease with the opposite sex and don’t find
it hard to talk to women; they are not nervous with the opposite
sex. Extraverts are highly sexed. They had intercourse early;
have strong sexual desires; get sexually excited very easily; think
about sex almost every day; are sometimes overpowered by sexual
feelings; are not afraid of sexual relationships; have no difficulty
in expressing strong sexual feelings. The sexual attitude of extra¬
verts is rather aggressive and overt. They usually feel aggressive
with their sex partner; they sometimes feel like scratching and
biting their sex partner; they have been involved with more than
one sex affair at the same time; they consider the dual standard of
morality natural; they consider absolute faithfulness in marriage
silly and would take part in an orgy if invited; hedonistic to the last,
they believe in taking their pleasures where they find them. Their
sexual development has clearly been rather more healthy than that
of the high P scorers. They have discussed sex with their parents;
they are not embarrassed to talk about sex; their religious beliefs
are not against sex; they don’t mind people touching them; they
don’t consider reading ‘girlie’ magazines to be a sign of failure
to achieve adult sex attitudes.
Female extraverts show a very similar pattern. They have strong
sexual desires; don’t feel sexually less competent than their friends;
consider themselves sexually attractive; have had many dates, many

120
 ADULT POPULATIONS

friends of the opposite sex, and for them sex contacts have never
been a problem. They love physical contact with people of the
opposite sex, feel at ease with them, and there is nothing they would
not do with anyone. They get excited sexually very easily, have not
been inhibited by their parents, are not nervous with the opposite
sex, don’t mind being touched by people, and don’t do things only
to please their sex partner. Extraverted women clearly have no
inhibitions; they always make love in the nude, consider the human
body a pleasing sight, would take part in an orgy, prefer to have
intercourse often, make lots of vocal noises during intercourse,
are not disturbed by seeing necking in public, and could do any¬
thing with a person they love. They are not embarrassed to talk
about sex, do not consider sex jokes disgusting, feel that sometimes
a woman should be sexually aggressive and prefer to make love with
the lights on and not under cover. Extraverts, both male and
female, are clearly of the ‘healthy animal’ type, unembarrassed and
uninhibited, but without the anti-social, somewhat abnormal, con¬
flictful admixture shown by the high P scorer.
When we turn to the high N males, there is clear evidence of
considerable abnormality and disturbance. They find thoughts about
sex disturbing, they worry a lot about sex, feel nervous with the
opposite sex, have felt guilty about sex experiences and cannot dis¬
cuss sexual matters with their wives. Yet they are highly sexed.
They confess to strong sexual desires, declare that sometimes
sexual feelings overpower them, have sometimes been afraid of
themselves for what they might do sexually, can think of nothing
but satisfaction when excited, and consider sex far and away their
greatest pleasure. Physical attraction is extremely important to them,
sex thoughts almost drive them crazy, yet thinking about sex makes
them very nervous. They get excited sexually very easily, believe
in taking pleasures where they find them and consider few things
more important than sex. However, there is also considerable
dissatisfaction with their love life. Conscience bothers them too
much; they are not satisfied with their sex life; something is lack¬
ing in their sex life; their love life has been disappointing; they
have had some bad sexual experiences; parents’ influence has
inhibited them sexually; they have been deprived sexually. There
is an aura of abnormality about their desires. They have some¬
times felt hostile to their sex partner; are excited by the thought
of an illicit relationship; have sometimes felt like humiliating their

121
 SEX AND PERSONALITY

sex partner; sometimes have problems in controlling their feelings;

find the thought of a coloured sex partner particularly exciting;
have been bothered by perverted thoughts; prefer a sex partner
several years older; usually feel aggressive about their sex partner;
consider that the need for birth control upsets love-making because
it makes everything so cold-blooded and planned. In addition, they
believe that women often use sex to gain all sorts of advantages;
wish that women would be more forthcoming sexually; would
protect children from contact with sex; consider virginity a girl’s
most important possession and do some things only to please sex
partner.
Women show a very similar pattern of conflict between strong
sexual desires and equally strong inhibitions. Conscience bothers
them too much; thinking about sex makes them nervous; they have
strong sexual feelings but can’t express them satisfactorily; sexual
feelings are sometimes unpleasant and they often wish that men
would be less demanding sexually. Yet when excited they think
of nothing but satisfaction; consider physical sex the most im¬
portant part of marriage; like to look at sexy pictures. They find
straight sex unsatisfactory; sometimes feel like humiliating their
sex partners; feel hostile to their sex partners; usually feel aggres¬
sive with their sex partner; consider most men sex-mad; prefer
a partner several years older; are bothered by perverted thoughts;
are embarrassed to talk about sex; find thoughts about sex dis¬
turbing; and consider the naked body not a pleasing sight. They are
afraid of sexual relationships; prefer their partner to dictate the
rules of the sexual game; had some bad sexual experiences when
young; have suffered from frigidity; have felt guilty about sex ex¬
periences; have been inhibited by parents’ influence; and didn’t
learn facts of life until late. They worry a lot about sex, feel
nervous with the opposite sex, feel sexually less competent than
their friends, are afraid of sexual relationships, find it hard to talk
to people of the opposite sex, and admit to strong inhibiting in¬
fluences. Something is lacking in their sex life; sex contacts have
usually been a problem; there are some things they would not do
with anyone, and some things which they do only to please their
sex partners. They did not learn the facts of life until quite old;
they are disgusted by seeing animals having sex in the street;
consider self-relief dangerous; consider tenderness the most im¬
portant quality in sex; and believe that in matters of sex women

122
 ADULT POPULATIONS

always seem to come off second best. The word that comes to mind
in characterising high N scorers is ‘crazy mixed up kids’ - although
of course these are not kids, but grown-up people demonstrating
their conflicts and complexes.
Little is known about the personality patterns associated with
high and low L scorers respectively, although one would expect high
L scorers to have very orthodox, socially approved views and atti¬
tudes, and to deny any thoughts of perverted practices and other
socially undesirable habits. Taking the male high L scorers first, we
find that this is precisely what they show. They are not overpowered
by sexual feelings; they have no problems in controlling sexual
feelings; they don’t believe in taking pleasures where they find
them; they would be bothered if the woman they married were
not a virgin; they had their first intercourse late; they have not been
involved in more than one sex affair at any one time; they never
had many dates and sex behaviour never caused them any trouble;
the strength of inhibiting influences is high. They never felt like
humiliating their sex partners; never felt hostile to them; never felt
like biting and scratching their partner during intercourse; have never
suffered from impotence; don’t get pleasant feelings from touch¬
ing their sexual parts. They would refuse to see a blue film, read
a pornographic book or take part in an orgy; they would also object
to the use of four-letter swear words in mixed company. They
oppose premarital sex, believe that pornographic writings should
be censored, that young people should not learn about sex through
their own experience, that women should not be sexually aggressive
and that young people should not be allowed out all night without
being closely checked. They would protect children from contact
with sex, consider absolute faithfulness in marriage not silly, do not
believe that females have such strong sexual desires as males and
consider it right that the man should be the dominant partner
in a sex relationship. They would need to love a woman in order to
have intercourse, consider the reading of ‘girlie’ magazines to sug¬
gest a failure to achieve adult attitudes to sex and believe that
we should not experiment with sex before marriage. They find the
thought of a sex orgy disgusting, do not consider it all right to
seduce a person old enough to know what she is doing and believe
that sexual permissiveness threatens to undermine the foundations
of society. They do not believe that most men are sex-mad, and do
not agree that the pill should be universally available.

123
 SEX AND PERSONALITY

The women share this highly conservative pattern. Female high

L scorers approve of the ‘double standard’ of sexual morality;
they believe that virginity is a girl’s most valuable possession; they
would prohibit pornographic writings; they object to the use of
four-letter swear words in mixed company. They deny that females
have as strong a sex drive as males, find the thought of a sex orgy
disgusting and believe there should be censorship of plays and films.
They feel that the opposite sex will not respect you if you are too
familiar, would be bothered if the person they married were not
a virgin and would oppose a law in favour of polygamy. Group
sex does not appeal to them, and they believe that sexual permis¬
siveness undermines the foundations of civilised society. They
would reject a highly pornographic book, would refuse to see a blue
film and believe that the pill should not be universally available.
Sex without love is considered highly unsatisfactory; they do not
understand homosexuals; the thought of an illicit sexual relation¬
ship does not excite them. They have never felt like humiliating
their sex partner; don’t feel like scratching and biting their sex
partners; have never suffered from frigidity; never feel hostile to
their sex partner; had first intercourse at a late age; do not think
about sex every day. They do not think that physical sex is the
most important part of marriage; would prefer to have intercourse
only rarely; consider sex not all that important to them; did not
have bad sex experiences when young; are not overpowered by
sex feelings and need conditions to be just right to get excited
sexually. Their sex partners satisfy all their physical needs com¬
pletely and there are some things they only do in order to please
their sex partners. Tenderness is the most important quality in a
sex relationship to them and they feel that it is better not to have sex
relations until you are married. These views are very conservative
and this is in good agreement with the fact that Wilson (1973) has
reported positive correlations between conservatism and L scores.
He also found a positive correlation between L scores and tender¬
mindedness and the attitudes summarised above agree with this
finding too. In the sexual realm, high L scorers are tender-minded
conservatives, which puts them into the religious quadrant (people
with positive religious beliefs tend to collect in this quadrant). In
agreement with this, Wilson found the highest correlation with L
in his religious-puritanical factor.
In looking at these brief and pictorial descriptions of the main

124
 ADULT POPULATIONS

results of this study, it should be borne in mind that the statements

made are relative, not absolute. When it is said that ‘extraverts
believe ..this simply means that more extraverts than introverts
hold this belief; it is quite possible that the belief is not in fact
held by a majority of extraverts. Thus if 25 per cent of extra¬
verts and 10 per cent of introverts believe that big-breasted women
are more attractive, then this belief would have been cited as
characteristic of extraverts, although, in fact, only one extravert
in four might hold it. This is therefore, strictly speaking, an in¬
accurate way of describing the results, but it obviates considerable
complexities of description that would otherwise be needed, and
this warning, together with the actual figures given in table 3.9,
will, it is hoped, make it unlikely that any misunderstanding will
arise.
In addition to correlating the four personality scales with each
of the items in the questionnaire, correlations were calculated be¬
tween these items and the age of the subjects, their married status,
their social class (working class v. middle class) and their occupation.
Subjects were asked whether their family was working-class or
middle-class and they were asked to give their occupation (house¬
wives giving that of their husbands); in addition, they were asked
to classify the occupation as higher professional/administrative,
lower professional/administrative, clerical, skilled manual, semi¬
skilled manual or unskilled. A positive correlation in the table indi¬
cates that the ‘Yes’ answer to an item is positively related to older
age, married status, middle class, low occupational status. As a
result, the signs of the correlations for social class and for occupa¬
tional status are opposite to each other; this should not prove
confusing as in the table the heading of each column indicates the
direction of the correlation.
The correlations are not large, rarely exceeding 0-5; this is of
course only to be expected by virtue of the low reliability of single
questions. Statistical significance is not very helpful in view of the
fact that the numbers of subjects are so large as to make even quite
small values of coefficients significant, and the additional fact that,
with so many items being correlated with so many social indicants,
the total number of correlations is so large that many would be
apparently ‘significant’ by chance. We will lay emphasis, as before,
on congruence of values from different items tapping the same
general attitude. In addition, we have correlated the various columns

125
 SEX AND PERSONALITY

in table 3.9; the results of this calculation are given in table 3.11,
with results for males being given above the leading diagonal and
those for females below the leading diagonal. The leading diagonal
itself is made up of the correlations for each of the indicants between
males and females; this set of figures shows to what extent high
correlations for a given item and a particular indicant are pro¬
duced by both men and women, and to what extent low correlations
are produced by both sexes. (These correlations have already been
abstracted in table 3.10.) The table has been extended to do the
same thing for our four personality scales.

Table 3.11 Correlations between columns in table 3.9 for four social
indicants and four personality traits: values above the leading diagonal are
for males, those below for females; values in the leading diagonal
(in italics) are correlations between corresponding columns for males
and females

1 2 3 4 5 6 7 8
1 Working class 0-52 -0-53 0-36 0-22 003 -0-35 -0-14 0-20
2 Middle-class -0-55 0-47 -0-43 -0-38 0-20 0-39 0-18 -024
occupation
3 Age 0-50 -0-35 0-71 0-58 -014 -0-44 0T0 0-59
4 Married status 0-35 -014 0-61 0-44 -0-28 -0-26 006 0-45
5 P -007 0T4 -0-38 -0-24 0-74 -0-02 -0-02 -0-35
6 E -0-37 004 -0-49 -0-28 - 003 0-57 0-03 -0-55
7 N 0-20 0-15 0-11 0-22 0-27 -0-52 0-55 -015
8 L 020 -0-04 0-60 0-30 -0-56 -0-28 -009 0-64

Let us first look at the leading diagonal. Clearly values for men
and women correlate reasonably highly together; in other words,
what is true of men is mostly also true of women. The highest values
are obtained by the P scale, among the personality measures, and
by age among the social indicants. Marital status, occupation and
social class give less clear-cut evidence for male-female agreement,
but the other three personality scales show reasonably high male-
female agreement. As regards the values in the body of the table,
we have divided this up into four quadrants. Two of these represent
the intercorrelations between the social indicants and the person¬
ality scales, respectively; the other two represent the cross-correla¬
tions between indicants and scales for men and women respectively.
Taking the personality scales first, we see that P and E show
opposite patterns for both sexes; so do E and N for the women,
but not for the men. L and E show opposite patterns for both

126
 ADULT POPULATIONS

sexes, but more so for the men. When we turn to the indicants we
find the expected negative relation between class and occupation,
owing to the reverse labelling of the two columns. Higher social
class goes with older age, and slightly with marital status. Age and
marital status are, of course, also quite highly correlated.
Looking at the cross-correlations, we find that E shows a pattern
of answering similar to that of the working class, and to that of
young people. P too shows this tendency to respond like young
people. N shows no noteworthy correlations, but L clearly pro¬
duces responses similar to those of older and married people. These
results are not unexpected. Scores on E and P decline with age,
while those on L increase (Eysenck and Eysenck 1975); so far,
the results in our table are predictable. However, N usually de¬
clines with age, too; yet the correlations with age in the table, while
very low, are positive instead of negative. This table demonstrates
the complexity of the interweaving of causal factors in the pro¬
duction of attitudes towards sex; in interpreting our data, this
complexity must always be kept in mind.
We will now turn to a discussion of the attitudes that characterise
old as opposed to young subjects, working-class as opposed to
middle-class subjects and married as opposed to single subjects.
In this discussion, it should always be borne in mind that we are
dealing with the results of correlational procedures, i.e. that com¬
parisons are strictly relative. When we say that old people prefer
to have intercourse less frequently, this statement is true (and mean¬
ingful) only in comparison with the younger groups in this study.
On an absolute scale, older subjects might still prefer to have inter¬
course quite frequently; it is only by comparing them with the
younger people who wish to have intercourse even more frequently
that we arrive at our conclusion. Similarly, the fact that older sub¬
jects find orgies more frequently ‘disgusting’ than do younger
subjects does not mean that a majority of older subjects holds this
view; it may merely mean that a larger minority can be found among
the older than among the younger subjects.
With this proviso in mind, we will first look at the attitudes of the
older subjects, as compared with the younger ones. We would expect
the libido of older subjects, both men and women, to be weaker than
that of younger subjects, and this is indeed the case. They prefer to
have intercourse less frequently; their habitual sex desire is weaker;
they do not think about sex every day; they do not get excited

127
 SEX AND PERSONALITY

sexually very easily; sexual feelings do not overpower them; they

do not feel like scratching and biting their partner during inter¬
course; it takes a lot to get them excited sexually; they do not enjoy
petting; they can take sex and they can leave it alone; they think
only rarely about sex. In addition, we would expect older people
to hold less ‘permissive’ and more conservative, ‘old fashioned’
views concerning morality and sexual behaviour; this too is what is
found. Older subjects disapprove of blue films and pornographic
books, and find orgies disgusting; they consider the reading of
‘girlie’ magazines to indicate a failure to achieve adult attitudes
towards sex. Similarly, people who attend ‘strip-tease’ shows are
considered sexually abnormal. They deplore premarital sex, con¬
sider it right that the man should be the dominant partner in the
sexual relationship, believe that there are too many immoral
plays on television, consider virginity a girl’s most valuable posses¬
sion, believe that sexual permissiveness undermines the entire
foundation of civilised society, consider impersonal sex highly
unsatisfactory, don’t feel it would be right to seduce a person old
enough to know what he or she is doing and believe that females
do not have such strong sexual desires as males. In addition, older
people seem to be more ‘inhibited’ than younger ones. They were
older at first intercourse, consider that their parents’ influence has
inhibited them, do not find it easy to tell sex partners what they
like and do not like about their love-making; they admit to strong
inhibitions, find some forms of love-making disgusting, prefer inter¬
course under bedclothes and with the lights off, don’t think they
could do anything with a person, even if they loved that person,
consciously try to keep sex thoughts out of their minds, don’t
like to look at sexy pictures, would particularly protect their chil¬
dren from contacts with sex, get sexually aroused at night, never in
the daytime, and do not make lots of vocal noises during inter¬
course. Sex is not all that important to them; they believe that
the opposite sex will respect you more if you are not too familiar
with them; there is a feeling that they have been deprived sexually.
The men do not prefer a sex partner several years older and they
do not believe that most men are sex-mad. Women find male
genitals aesthetically unpleasing; prefer the partner to dictate the
rules of the sexual game; wish that men would be less demanding
sexually, consider that men are more selfish in their love-making
than women, and believe that in matters of sex women always seem

128
 ADULT POPULATIONS

to come off second-best. Most of these attitudes fit in pretty well

with the stereotype of the ‘square’, inhibited, non-swinging type
of person the young ‘swingers’ imagine their parents to be; it is note¬
worthy, however, that there is little hint of any less satisfaction
with their sex lives among the older subjects. It would be interest¬
ing to discover whether the differences between the generations
are due to age by itself, i.e. whether attitudes of the older subjects
have actually changed as they grew older, or whether the attitudes
of the older subjects simply mirror a different generation’s patterns
of upbringing. If the former hypothesis is true, then the ‘young’
subjects of our experiment would be expected to grow into the
typical attitude patterns of our ‘old’ subjects in the course of time.
If the latter hypothesis is true, then no change would be expected.
Only a longitudinal study can answer questions of this type.
Married people (as contrasted with unmarried people not living
together with any particular person, but having had sexual ex¬
perience) do not differ on very many items from unmarried people
and the few items that do show significant correlations do not make
any very clear pattern, other than duplicating some of the features
of the comparison between old and young subjects. They do not
find it easy to tell sex partners what they like and do not like about
their love-making, and they cannot discuss sexual matters with their
husbands; the women are embarrassed to talk about sex. The need
for birth control upsets their love-making, but they do not consider
physical sex the most important part of marriage. The men are
not satisfied with their sex life, and the women feel that some¬
thing is lacking in their sex life - this would sound like a vote of
lack of confidence in the institution of marriage, were it not for
the fact that the correlations in question are not very large! Married
people are non-permissive, oppose orgies, wife-swapping, blue films,
voyeurism, sexy pictures, premarital intercourse, masturbation and
the young learning about sex through their own experience. Vir¬
ginity is important and sex should be for purposes of reproduction,
not for personal pleasure. Marriage being a traditional institution,
it is not unexpected that those who are married tend towards tradi¬
tional values and attitudes; it is perhaps somewhat unexpected
that the correlations found are not larger and more numerous.
The class and occupational comparisons show the working-class
groups as more ‘earthy’ and the middle-class groups as more ‘moral’
in their attitudes, which is perhaps not surprising in view of the

129
 SEX AND PERSONALITY

general finding of greater degrees of ‘tough-mindedness’ in working-

class groups and of ‘tender-mindedness’ in middle-class groups
(Eysenck 1954, Wilson 1973). Working-class groups are not averse
to group sex, wife-swapping, masturbation, perversions, orgies; they
consider physical sex the most important part of marriage. They
would not mind if sex partner had sexual relations with someone
else, consider sex more exciting with strangers, consider sex far
and away their greatest pleasure and believe that physical factors
in their sex partners are far more important than psychological
ones. They have strong sex drives, believe in taking their pleasures
where they find them, enjoy pornography, would watch people
making love, don’t object to the use of four-letter swear words
in mixed company, would find a coloured sex partner particularly
exciting, have sexual fantasies involving flogging, don’t mind see¬
ing necking in public and would enjoy watching their usual sex
partner having intercourse with someone else. Permissiveness clearly
is not a middle-class invention, although it should be borne in
mind that many of these items have very low incidence, so that our
caveat about the interpretation of correlations as purely relativistic
indicators applies doubly in connection with such items as watching
one’s usual sex partner having intercourse with someone else or
finding a coloured sex partner particularly exciting.
This concludes our discussion of the correlations between sex
attitudes and our various social indicants. There is very little in
the literature with which to compare our findings. There are data
in Kinsey’s work on the incidence of certain types of behaviour in
relation to age and class, but Kinsey was not concerned with
attitudes as such, and attitudes may be something very different
from actual behaviour. Women may take part in fellatio, but may
not enjoy it at all and have very negative attitudes towards such
practices (Eysenck 1971b); in such cases, knowledge of only the
attitudes or only the actual behaviour gives information that is
essentially one-sided, and such information should be comple¬
mented by knowledge of those aspects of the total situation that
tap a different class of components. Our data provide a beginning
of information in this complex and difficult field. The data tend
to form meaningful patterns, and we have tried to indicate the
sorts of patterns that are suggested by the figures in the table.
These patterns cut across those discussed in relation to personality
variables, and to obtain a general picture both sets of data should

130
 ADULT POPULATIONS

be borne in mind. The facts so disclosed may suggest testable hypo¬

theses to future workers in this field which may admit of a simple
design to test their adequacy or otherwise; while several hypotheses
were put forward for testing in this study, the whole field is so new
that these ‘hypotheses’ would perhaps better be called ‘hunches’
than anything deriving from a well-established theory permitting of
quantitative deductions.
Another scale was administered to the same population, namely
the Reiss Permissiveness Scale. This premarital sexual permissive¬
ness scale constructed by Reiss (1967) has the properties of a
Guttman-type scale, and has twelve items. In this study we have
used the shorter five-item ‘universal scale’ consisting of items 5, 7,
10, 11 and 12; it appears from the original book that little is lost
by thus shortening the scale. The items deal respectively with the
following questions to which the subject is required to answer on
a six-point scale, from strongly agree (1) through medium (2) and
slight agreement (3) to slight (4) and medium disagreement (5)
to strong disagreement (6). Item 1 states that petting is acceptable
before marriage when the subject is engaged to be married; item
2 states that petting is acceptable when subject has strong affection
for the partner; item 3 states that sexual relations are acceptable
before marriage when subject is in love; item 4 states that sexual
relations are acceptable before marriage when subject feels strong
affection for the partner; and item 5 states that sexual relations
are acceptable before marriage even if subject does not feel par¬
ticularly affectionate towards the partner. Each subject answers the
scale only for himself/herself, and not from both the male and
female points of view, as in Reiss’s original study.
Table 3.12 reports the means and standard deviations of male
and female subjects separately for the five permissiveness scales
and the four personality scales. From the nature of the scale, we
would expect that scores would increase monotonically (implying
greater disagreement with the permissive wording of the question
from the first to the fifth item), and this is indeed so. We would
expect males to be more permissive than females, and again
this is so on all items; all the comparison show differences that are
highly significant statistically. Women clearly are less permissive
than men on this inventory. (Women also seem to be less agreed
on their opinions; the standard deviations of the women are signi¬
ficantly greater on all items except the last than the standard

131
 SEX AND PERSONALITY

deviations of the men. This may be an artefact of the different scale

positions; values around the 3 and 4 positions are more likely
-to have larger standard deviations.)

Table 3.12 Means and standard deviations of five permissiveness scales

from the Reiss Courtship Inventory

Item Males Females

1 l-17±0-53 1-38 + 1-00
2 1-28 ±0-74 1-51 + 110
3 1-49 ±1-23 1-71 + 1-41
4 1-79 ±1-41 2-41 ±1-67
5 2-82+1-80 4-01 + 1-83

The correlations between the five items in the Reiss scale are
given in table 3.13 as are their correlations with P, E, N and L. As
expected, the intercorrelations of the five Reiss items follow the
superdiagonal form; i.e., correlations decrease the further they
are from the superdiagonal. This is true for both men and women,
and is a necessary property of a scale that has Guttman properties.
All the intercorrelations are of course fully significant statistically.
The correlations with personality scales are similar for males and
females and give a clear picture. P correlates positively with per¬
missiveness, with correlations much higher for the questions in¬
volving premarital sexual intercourse; the correlations quoted are
negative because high scores on the scales signify disagreements
with the permissive content of the item. L correlates negatively
with permissiveness, with correlations much higher for the ques¬
tions involving premarital sexual intercourse. N shows hardly any
correlations at all and correlations with E are slight but on the
whole indicate a permissive attitude. These correlations are all
in good agreement with predictions to be derived from our previous
work.
Of the subjects questioned, 241 were married to each other, or
in a number of cases were living with each other without being
married. The existence of these couples enables us to provide some
evidence for or against assortative mating. It is well known that
assortative mating with respect to intelligence is quite marked,
approximating a correlation of 0-6 between spouses; with respect
to personality, the evidence suggests little or no assortative mating,
except for spouses with some degree of psychiatric disability

132
 ADULT POPULATIONS

(Eysenck 1975a). There is little evidence regarding attitudes, but

it might be surmised that as attitudes are easier to change than
largely innate personality qualities, higher correlations between
spouses would be found with respect to sexual permissiveness,
although of course it would not be clear whether these correlations
were due to true assortative mating, or whether they were due
to a tendency towards agreement consequent upon marriage. The
correlations between spouses for the five permissiveness scales are
as follows: 0T1, 022, 044, 0-27; all except the first of these are
statistically significant at the <0 01 level. There clearly is evidence
for homogamy here with items 3 and 4 providing the highest figures.

Table 3.13 Intercorrelations of five courtship scales and correlations with

four personality scales. A single asterisk indicates statistical significance at
the 5 per cent level and double asterisks at the 1 per cent level

Females

1 2 3, 4 5 P E N L

Males
1 0-69 0-32 0-32 0-26 -007 0-03 -001 006
2 0-71 - 0-56 0-54 0-36 - 009* 001 -0-04 0T4**
3 0-48 0-58 - 0-85 0-51 -015** -0 09* -005 0-19**
4 0-39 0-59 0-78 - 0.66 -0-21** -Oil* -006 0-20**
5 0-26 0-38 0-47 0-65 - - 0-20** -Oil* -005 018**

P -0 03 -0 05 -0-22** -0-23** -0-25**

E -0 09 - 0 09*-0-01 -0 04 - 0 08
0T0* 002 -0-02 -002 -0'02
008 0-13** 0-22** 0-19** 017**

In addition to the Sex Attitudes Scale and the Permissiveness

Scale, a Sexual Behaviour Scale was also applied to the subjects
of the experiment. This scale is reproduced below; it differs from
an earlier version (Eysenck 1972), which contained more ‘low-level’
sexual behaviour items, because in this sample all subjects had
sexual experience with intercourse, and would therefore not have
been discriminated from each other by these items. The twelve
items included are not in ascending order; this arrangement was
intentional as an ascending order might produce artificial con¬
formity with the construction of the scale. The scale here printed is
that for females; that for males requires suitable alteration.

133
 SEX AND PERSONALITY

Sexual Behaviour Scale

Here are brief descriptions of sexual behaviour patterns in which people

indulge. There are three possible answers for each item; please put a
cross (x) in the appropriate category.

Column 1 means: ‘I have done this and enjoyed it’

Column 2 means: ‘I have done this, but didn’t like it’
Column 3 means: ‘I have never done this, but would like to do it’

If you do not endorse any of these three categories, we would take this
to mean that you have not done this, and don’t think you would enjoy
doing it.

Have done Have done Have not

and enjoyed and didn’t done but
like would like
to do

1 One-minute continuous lip kissing

2 Having genitals kissed by a male
3 Sexual intercourse, man above
woman
4 Having naked breasts fondled
5 Kissing male genitals
6 Sexual intercourse, man behind
woman
7 Having nipples kissed and/or
bitten by male
8 Mutual handling of genitals
9 Mutual kissing of genitals (‘69’)
10 Sexual intercourse in positions
other than 3 or 6
11 Intercourse with more than one
person present
12 Intercourse with several people
around (‘group sex’)

Table 3.14 gives the results of the study; it shows means and stan¬
dard deviations on the three parts of the scale for men and women
separately, as well as correlations of scores on the three parts with
the eleven sexual attitude factors, with the two sexual attitude
superfactors, with masculinity-femininity, and with the personality
scales (P, E, N, L). The mean value of the ‘Have done and enjoyed’
column is slightly higher for men than women.

134
 ADULT POPULATIONS

Table 3.14 Means and standard deviations of sexual behaviour scale for
males and females; also correlations with sexual attitudes scale

Have done Have done Would like

and enjoyed and disliked to do
(1) (2) (3)
M F M F M F
1 Permissiveness 0-18 0-36 -004 -0-20 010 008
2 Satisfaction 0-31 0-33 -0-06 -0-20 -0-33 -0-20
3 Neurotic sex -0-15 -003 -002 003 0-22 004
4 Impersonal sex 0-26 0-23 -0-09 -0-02 0-19 0-20
5 Pornography 0-33 0-40 -0-18 -0-17 0-08 0-10
6 Sexual shyness -0-26 - 0-20 0-09 010 0-25 018
7 Prudishness -0-39 -044 0-20 0-31 006 -0-02
9 Disgust -0-36 -0-51 0-24 0'44 0-02 004
10 Excitement 0-40 0-51 -0-28 -038 002 0-01
11 Physical sex 0-29 0-36 -Oil -0-23 000 001
12 Aggressive sex 0-21 0-32 -004 -015 -003 -004
SF1 Libido 0-38 0-45 -0-14 -0-24 0-16 013
SP2 Satisfaction 049 0-52 -017 -0-39 -0-32 -0-15
Masculinity-Femininity 0-32 0-38 -017 -0-17 0-20 0-17
P -004 009 006 -002 -0-12 0-08
E 0-18 0-09 -001 -005 010 -001
N -003 -0 06 001 008 002 0-03
L -0T4 -0-06 0-03 0-08 0-20 -007

Means
Males 8-92 ±2-16 0-21 ±0-60 1-37 ±1-71
Females 8-53 ±2-13 0'76± 1-21 0-56± 108

Figure 3.2 gives the actual percentage values for men and
women for the items in the ‘have done and enjoyed’ column. Items
1, 3, 4, 7 and 8 are all around the 95 per cent level, and there
seemed little point in separating them out; none of the differences
are statistically significant. Consequently, they have been designated
as ‘Base level: Intercourse’, and grouped together at the 95 per cent
level for both men and women. (There are no meaningful sex
differences on these items.) Of interest in the curves shown in figure
3.2 are the sex differences for later items, which, although not large
in absolute terms, are quite meaningful and mostly significant statis¬
tically. Thus it seems that women have done and enjoyed cunnilingus
more than fellatio, while for men the position is reversed. It is not
perhaps surprising that passive oral activity is more pleasurable
than active, although other explanations of the finding may be
possible. For all the later items (6, 9, 11, 12) men have a higher
percentage than women; this again is not perhaps surprising. These

135
 SEX AND PERSONALITY

figures are of interest mainly on two accounts. They extended

earlier work which has mostly concentrated on students and other
adolescent and very young groups, to adult men and women. And
secondly, they characterise our sample by sexual activity; this
descriptive function takes its place beside the description by person¬
ality already given.

Figure 3.2 Frequency of different sexual behaviours for male and female
participants in experimental study

There are no English norms with which this figure could be com¬
pared. If the random sample of French adults by Simon (1972)
can be used as a rough guide for present-day European adults, and
if we can accept the veridical nature of their responses, then our
sample gives rather similar responses to those of the French group.
For mutual kissing of genitals, 60 and 55 per cent respectively of
French males and females answer ‘Yes’, as compared with our
65 and 61 per cent. (These are only for those respondents in the
French sample who actually answered the questions; if it is reason¬
able to assume that most of those refusing to answer had had no
experience with this type of conduct, the ‘true’ figures would be
somewhat lower.) For the question regarding intercourse with
others present, 10 per cent of males and 1 per cent of females
answered in the affirmative, again of those who answered the
question at all. Granted that there are many difficulties in making
such a comparison, it nevertheless becomes clear that the figures
are not too dissimilar to those of our own inquiry.

136
 ADULT POPULATIONS

As regards activities which our subjects ‘have done and disliked’,

the distributions are very abnormal and indeed J-shaped; this is
clearly brought out by the fact that the standard deviations are
much larger than the means. However, clearly there are important
sex differences; women have done almost four times as many things
they disliked as men; and they are much more variable in this res¬
pect. This result is not surprising in view of the fact that it is
men who mainly dictate what is to be done in sexual intercourse.
As regards things our subjects ‘would like to do’, again we note
a J-shaped distribution, with the men listing almost three times
as many items as the women. This result does not seem un¬
expected, in view of the traditionally greater sexual libido of men,
which we have also found in our own work. In view of the abnormal
distributions involved statistical tests of significance may be mis¬
leading; for the sake of interest, however, we may note that
the differences between men and women on all three scales are
statistically significant. (5)
When we turn to the correlations of the three scales (which we
shall refer to as B.S.ls B.S.2 and B.S.3 (B.S. for behaviour scale)
with personality, we find only low correlations (table 3.14). B.S.X
correlates positively with E, negatively with N and L; these corre¬
lations are consistent for the two sexes and may therefore, in spite
of the low values of the coefficients, represent some factual rela¬
tionship. B.S.2 does not show any meaningful correlations with the
personality scales. B.S.3 has a positive correlation with L, but only
for the men; for the women the correlation is very small and
negative. This may represent a genuine sex difference, or else a
statistical artefact. For P, too, the coefficients for the two sexes
have opposite signs; they are probably too low to require inter¬
pretation. Altogether, therefore, there is little evidence for any
involvement of personality with sexual behaviour as here indexed.
Correlations of the three scales with the attitude scales are
congruent for the two sexes and reach quite reasonable values.
Taking B.Sn first, we find that this correlates (positively) with
permissiveness, satisfaction, impersonal sex, pornography, excite¬
ment, physical sex, aggressive sex, libido, satisfaction and mascu¬
linity (in both sexes!). There are negative correlations with neurotic
sex, sexual shyness, prudishness and disgust. These correlations all
make perfectly good sense if we postulate some general form of sex
drive which, if strong, leads to varied sexual performance, as well as

137
 SEX AND PERSONALITY

attitudes favouring diverse forms of sexual expression. The results

contradict an often expressed hypothesis that liking for pornography
represents an outlet for sexual desire which is chosen instead of
more direct sexual activity; our results suggest that pornography
represents such an outlet in addition to more direct sexual activity.
B.S.jj produces much lower correlations; this may in part be due
to the small number of positive endorsements in this scale. It does
appear, however, that dislike of some sexual activities indulged
in is positively correlated with sexual shyness, prudishness and
disgust; negative correlations appear with excitement, physical sex,
libido, satisfaction, masculinity, pornography, impersonal sex and
permissiveness. Again these correlations are easy to understand.
B.S.3 shows few even medium-sized correlations. The most marked
are perhaps a negative correlation with sexual satisfaction and a
positive one with sexual shyness; there is also a positive one with
impersonal sex and another with masculinity. On the whole, people
who are satisfied with their sex lives have no ‘hang-ups’ about what
they would like to do, whereas shy people do. Here too, the results
do not seem to contradict common sense.
It should be noted that the correlations between sexual attitudes
and sexual behaviour are not likely to have been caused by per¬
sonality traits; the correlations between personality and sexual
behaviour, as we have noted, are uniformly low and certainly much
lower than those between attitudes and behaviour. Even if person¬
ality were partialled out from the correlations between attitudes and
behaviour, the resulting partial correlations would only be marginally
different from those actually found.
It is interesting to note the relationship between sexual be¬
haviour and the two independent superfactors we found in our
analysis of the attitude scales. Libido and satisfaction are un¬
correlated, but subjects who had done and enjoyed more than their
fair share of the activities mentioned in our behaviour scale have
high scores on both libido and satisfaction. On the contrary, those
who have done some of the things described and disliked them
have low scores on both the libido and the satisfaction scales - the
women more markedly so than the men. Only with respect to things
subjects had not done but would like to do do libido and satisfaction
part company; subjects high on libido would, those high on satis¬
faction would not, want to indulge in these activities.
A special analysis was made of the ‘perverted’ sex attitudes (11

138
 ADULT POPULATIONS

and/or 12), relating to group sex (Eysenck 1976). Participation of

subjects in such perverted practices was significantly correlated
with P in females and with E in males; there was also a significant
correlation with L (negative) in males. (All the observed differences
mentioned above were significant at p <(0 01. All other comparisons
gave non-significant differences between perverts and non-perverts.)
Participation in perverse practices also correlated with the libido
factor (p <0 001) and with the masculinity factor (p <0 001) for
both males and females, but not with satisfaction. There was also
a significant correlation with tough-mindedness (0 05 and 0-01 res¬
pectively for men and women). (The meaning of this term is ex¬
plained in some detail in the next chapter.)
The major finding of interest is that relating to the personality
variables. It is not unexpected that perversity relates to P in women,
but to E in men; such practices are more usually regarded as ‘typic¬
ally masculine’; a woman would have to be rather ‘abnormal’ (high
P) to behave in a manner regarded as unfeminine. For men, such
behaviour being regarded as more nearly normal, indulgence would
only require a certain amount of extraversion. There is evidence
from other types of conduct (criminality, V.D., etc.) that women
who behave in a ‘masculine’ fashion tend to be characterised by
particularly high P scores (Eysenck 1976).
1. Note that while the factors of libido and satisfaction correlated
significantly for the women, the scales did not. We shall take the scale
intercorrelations more seriously because of statistical doubts about the
accuracy of inter-factor correlation estimates (Guilford 1975).
2. This is possibly the wrong term; scores at the time of ‘mating’
may have been different from those after years of association.
3. Byrne (1971) has reviewed the evidence on husband-wife simi¬
larity, concluding that spouses indicate congruence on a wide variety of
ideals and attitudes. Byrne and others (1973) also report a study that
is directly relevant to our discussion. Using forty-two married couples,
they exposed their subjects to a variety of erotic stimuli (pictorial, ver¬
bal, imaginal) and recorded their reactions. Total arousal scores corre¬
lated 0-47 between spouses; pornographic judgements (i.e. degree of
pornography judged for each stimulus) correlated 0-32. Restrictiveness
scores were obtained by summing the pro-censorship responses to
questions that advocated forbidding the sexual materials in question
to various groups; the correlation for this index was 0-39. Authori¬
tarianism scores were also obtained, and showed a strong correlation
(r = 0-64). In addition, authoritarians reported greater arousal, and
advocated greater restrictiveness. These results are relevant to our own
work, as reported in a later chapter, dealing with social attitudes;

139
 SEX AND PERSONALITY

authoritarianism correlates positively with both tough-mindedness and

conservatism, and it will be seen that the relations reported by Byrne
and others are similar to those found by us.
4. This item also figures in the P scale, and its high correlation with
that scale is, therefore, partly spurious. However, it is only one item
in twenty, and hence only a small part of the correlation is due to this
overlap. The same caution applies to one or two of the E items later.
5. The statistical assumption underlying t tests are of course well
known, and so are the effects of deviating from these assumptions.
Boneau (1959), Havlicek and Peterson (1974) and Welch (1937) have
investigated the results of such violations, and on the whole the depar¬
tures from the conditions for which t tests are appropriate are not
such as to make the use of the test contra-indicated here. In particular,
it appears that when sampling from two non-normal distributions, it is
important for the use of t that both sets of scores should be skewed
in the same direction and that variances should not be too dissimilar.
As t values obtained in the present comparisons have p values of
<0-01 and <0-001 attached to them, there seems little doubt that the
differences are, in fact, replicable.

140
 APPENDIX A
ITEMS AND KEYS FOR PRIMARY
FACTOR SCALES

1 Permissiveness 2 Satisfaction
5 - 4 +
17 - 11 -
25 + 16 +
38 - 19 -
41 - 20 +
57 + 21 -
64 - 56 -
78 + 108 +
79 + 113 +
81 - 114 -
85 + 118 -
87 - 139 —
93 +
134 -

3 Neurotic sex 4 Impersonal sex

7 + 2 -
18 + 40 -
20 - 65 +
23 + 83 +
24 + 89 +
26 + 92 +
32 + 95 +
44 + . 97 +
46 + 102 +
56 + 119 +
59 + 120 +
60 + 135 -
84 + 144 +
153 -

141
 SEX AND PERSONALITY

5 Pornography 6 Sexual shyness

10 + 47 +
43 + 50 —

58 - 52 +
76 + 59 +
77 + 61 +
141 + 124 +
151 -
152 -

7 Prudishness
51 +
55 -
58 +
64 +
71 -
112 —

122 -

126 +
141 -

9 Sexual disgust 10 Sexual excitement

9 - 3 -
34 + 6 -

104 + 9 +
112 - 30 +
133 + 34 -
146 + 37 +
39 +
71 +
146

11 Physical sex 12 Aggressive sex

31 68 +
48 + 75 +
49 + 101 +
71 + 116 +
86 + 121 +
106 + 132 -
109 -
111 +
127 +
131 +

142
 APPENDIX B
ITEMS AND KEYS FOR SUPERFACTOR
SCALES

Sexual satisfaction Sexual libido

Item Key Item Key Item Key

4 + 1 _ 76 +
11 — 2 — 77 +
15 — 5 — 78 +
18 — 6 — 79 +
19 — 10 + 81 —

20 + 25 + 85 +
21 — 37 + 87 —

31 — 38 — 89 +
32 — 39 + 92 +
44 — 40 — 95 +
56 — 41 — 96 +
108 + 42 + 119 +
117 — 43 + 120 +
118 — 46 + 134 —

124 — 65 + 135 —

133 — 72 + 151 +a
74 + 152 +a
153 +a
154 + efg

143
 APPENDIX C
MASCULINITY-FEMININITY SCALE KEY

Item Key Item Key Item Key

2 _ 64 _ 96 4-
3 — 65 + 97 4-
7 + 67 + 101 —

10 + 69 — 102 4-
13 + 76 + 106 4-
16 — 77 + 109 —

18 ' — 78 + 113 4-
22 + 79 + 114 —

30 + 80 — 116 —

37 + 84 — 119 4-
39 + 85 + 120 +
40 — 86 + 128 —

42 + 89 + 135 —

43 + 91 — 145 4-
44 — 92 + 146 —

55 4- 95 + 147 —

58 —

63 +

144
 APPENDIX D
PERSONALITY QUESTIONNAIRE

INSTRUCTIONS
Please answer each question by putting a circle around the ‘Yes’ or the
‘No following the question. There are no right or wrong answers, and
no trick questions. Work quickly and do not think too long about the
exact meaning of the question.

REMEMBER TO ANSWER EACH QUESTION

1 Do you often long for excitement? ’ YES NO

2 Does your mood often go up and down? ..YES NO
3 Would parachute jumping appeal to you? (Aes NO
4 Have you ever taken the credit for something you
knew someone else had really done? Cyssp NO
5 Are you a talkative person? YES ^ NO ^
6 Do you ever feel ‘just miserable’ for no good reason? (YESy NO
7 Do most things taste the same to you? YES f NO '
8 Were you ever greedy by helping yourself to more
than your share of anything? YES NO
9 Are you usually carefree? YES ( NtV
10 Do you often worry about things you should not
have done or said? (yes) NO
11 Would it upset you a lot to see a child or an animal
suffer? £es ' NO
12 If you say you will do something, do you always keep
your promise no matter how inconvenient it might
be? YES t'' NO‘
13 Can you usually let yourself go and enjoy yourself
a lot at a gay party? YES (jicP
14 Are your feelings rather easily hurt? YES ’) NO

145
 SEX AND PERSONALITY

15 Do you think that marriage is old-fashioned and

should be done away with? YES
16 Have you ever blamed anyone for doing something
you knew was really your fault? YES
17 Are you an irritable person? YES
18 Do you have many different hobbies? YES
19 Do you love your mother? (yes""'
20 Are all your habits good and desirable ones? YES
21 Do you often do things on the spur of the moment? YES
22 Are you often troubled about feelings of guilt? (y§> NO
23 Do fast roundabouts and swings at funfairs give you
a thrill? NO
24 Have you ever taken anything (even a pin or button)
that belonged to someone else? YES NO
25 Do you like going out a lot? NO
26 Would you call yourself tense or ‘highly-strung’? (yes5
27 Do you read the Bible every day? YES
28 Do you sometimes talk about things you know
nothing about? 'sYES' NO
29 Do you have many friends? NO
30 Do you worry about awful things that might happen? Qyes
31 Do you enjoy hurting people you love? YES
32 Do you always say you are sorry when you have been
rude? NO
33 Do you hate being with a crowd who play harmless
jokes on one another? NO
34 Would you call yourself a nervous person? NO
35 Can you easily understand the way people feel when
they tell you their troubles? NO
36 As a child, did you do as you were told immediately
and without grumbling?
37 Do you like talking about sex?
38 Do you worry about your health?
39 Would you like to learn to fly an aeroplane?
40 Have you ever broken or lost something belonging to
someone else? YES
41 If there is something you want to know about, would
you rather look it up in a book than talk to someone
about it?
42 Do you suffer from sleeplessness?
43 Would you like to think that other people are afraid
of you?
44 Are you always quiet when other people are talking?
45 Are you rather lively?
46 Do you very easily feel bored?
47 Would you like to see blood sports like fox hunting
forbidden because they are cruel?

146
 ADULT POPULATIONS

48 Do you throw waste paper on the floor when there is

no waste paper basket handy?
49 Can you easily get some life into a rather dull party?
Pi
(YES''
50 Do you sometimes sulk? YES
51 Do you find it hard to trust anyone?
52 Do you sometimes boast a little?
53 Do you like doing things in which you have to act
quickly?
54 Do you often feel life is very dull?
55 Would you take drugs that may have strange or
dangerous effects?
56 Have you ever written your name in a school or
library book? NO
57 Do you prefer reading to meeting people? NO
58 Have you often felt listless and tired for no good
reason? NO
59 Do you usually note the number of a car you think is
being driven too fast and tell the police?
60 Have you ever said anything bad or nasty about
anyone? NO
61 Are you mostly quiet when you are with other people? NO
62 Do you often feel ‘fed up’? NO
63 Do you enjoy practical jokes that can sometimes
really hurt people?
64 As a child were you ever cheeky to your parents?
65 Do you like having long chats on the telephone?
66 Are you touchy about some things?
67 Are you slow and unhurried in the way you move?
68 Do you always wash before a meal?
69 Would you rather plan things than do things?
70 Are you sometimes bubbling over with energy and
sometimes very sluggish?
71 Do you give something to every beggar you see?
72 Have you ever cheated at a game?
73 Do you often take on more activities than you have
time for? YES /no':
74 Do you worry too long after an embarrassing ex¬
perience?
75 Is your mother a good person?
76 Have you ever ‘played sick’ to get out of something?
77 Would you say that you were fairly self-confident?
78 Have you always been known as a loner?
79 Do you get nervous in places like lifts, trains or
tunnels?
80 Have you ever taken advantage of someone?
81 Do you like cracking jokes and telling funny stories
to your friends?

147
 SEX AND PERSONALITY

82 Do your friendships break up easily without it being

your fault?
83 Do you like mixing with people?
84 Do you suffer from ‘nerves’?
85 Are you always polite even to unpleasant people?
86 Would you feel very sorry for an animal caught in
a trap?
87 Do you nearly always have a ‘ready answer’ when
people talk to you?
88 Do you daydream a lot?
89 Are you always specially careful with other people’s
things?
90 Have you ever insisted on having your own way?
91 Would you call yourself happy-go-lucky?
92 Do you have many nightmares?
93 When you are in a crowd, do you worry about
catching germs?
94 Would you dodge paying taxes if you were sure you
would never be found out?
95 Do you mind selling things or asking people for
money for some good cause?
96 Do you try not to be rude to people?
97 Have you ever deliberately said something to hurt
someone’s feelings?
98 Are you easily hurt when people find fault with you
or the work you do?
99 Would you rather be on a desert island than in the
middle of a bustling city crowd?
100 Do you sometimes get cross?
101 Do you sometimes feel life is just not worth living?
102 Before taking decisions do you generally ask some¬
one’s advice?
103 Do you always practise what you preach?
104 Do you always keep smiling even when things go
wrong?
105 Are you troubled by aches and pains?
106 Have you ever told a lie?
107 Do you prefer to have few but special friends?
108 Do good manners and cleanliness matter much to
you?
109 Do you go for things you want rather than wait for
them to come to you?
110 Did you mind filling in this questionnaire?

148
 ADULT POPULATIONS

Key to personality scale

p E N L
7 5 2 - 4
- 11 13 6 - 8
15 18 10 12
- 19 21 14 -16
31 25 17 20
- 35 29 22 -24
43 - 33 26 -28
- 75 45 30 32
55 49 34 -40
63 - 57 38 -52
78 - 61 42 -60
82 65 50 -64
- 86 - 69 54 68
- 89 73 58 -72
93 81 62 -80
- 96 83 66 85
-100 87 70 -90
-106 91 74 -94
-108 - 95 84 -97
110 -107 98 103

149
 4

Social attitudes and sexual behaviour

In addition to the questionnaires mentioned already, and discussed

in some detail (sexual attitudes, sexual behaviour, permissiveness,
personality), the subjects of our main inquiry, as described in
chapter 3, were also administered a social attitudes inventory. The
purpose of this additional inquiry was of course to see to what
extent social attitudes might be related to sexual attitudes and
behaviour. The inventory was a revision and extension of the
questionnaire used in The Psychology of Politics (Eysenck 1954);
it contains eighty-eight items, as well as questions regarding the
social class and voting intentions of the subject. It was expected
that the main factors to emerge from this inventory would be con¬
servatism and toughmindedness; these two major factors originally
appeared in the analyses reported in The Psychology of Politics, and
have been duplicated several times since (cf. Wilson 1973). More
recently we have found that a third factor, capitalism v. socialism,
could be isolated; this factor was sufficiently independent of con¬
servatism (with a small c!) to deserve separate measurement, and
consequently the questionnaire here used contained several more
questions on socialism and capitalism than did the original one.
Our expectation to find relations between social attitudes and
sexual ones was heightened by the fact that
1 there is a strong genetic factor in the causation of social attitudes
2 social attitudes are related to personality factors
3 these relations themselves have a genetic rather than an
environmental basis (Eaves and Eysenck 1974).
The heritability of the conservatism-radicalism factor was 0-65, and
that of toughmindedness 0-54; toughmindedness was significantly
associated with personality factors E and P on a genetic basis. In

150
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

this study no separate factor of capitalism-socialism was looked for

or found, the original Social Attitude Inventory being used (Eysenck
1954). A later study (Eysenck 1975b) was designed to investigate
the possibility that this might be an additional, third, major factor
in the social attitudes field. The choice of items in this study was
similar to that used in the present study, and may therefore be
worthy of discussion.
First of all a thorough survey was made of published articles
and books dealing with social attitudes, and each item listed that
seemed in any way related to the three putative factors of conser¬
vatism, capitalism or toughmindedness. Duplicates and similar-
to-each-other items were eliminated, and the remaining 176 items
randomly assigned to one of two sets, A and B. One of these
sets constitutes the present groups of items, while the other was
used in the published study. It cannot of course be claimed that
in this way we have succeeded in procuring a truly random sample
of all relevant social attitude statements or questions, but this
procedure seems to cut subjectivity down to a certain minimum. The
questionnaire was administered to a reasonably random sample of
subjects (n—368), and factor-analysed; three effectively indepen¬
dent factors corresponding to conservatism, capitalism and tough¬
mindedness were extracted (Eysenck 1975b). This finding suggested
that similar factors would emerge from the present study, using
the second set of items.
The Public Opinion Inventory is given below (table 4.1), together
with instructions. Given after each question is the percentage of
males and females giving a positive answer to each question. It
will be seen that, although men and women do differ on many
items, their differences are on the whole much smaller than were
those on the sexual attitudes inventory. It would seem that it is
particularly with respect to sexual matters that men and women
disagree, whereas with respect to general social attitudes their dis¬
agreements remain rather muted.

Table 4.1 Public Opinion Inventory, together with percentage ‘yes’

answers of men and women
Sex- Age- Name- (optional)
It is hoped you will be interested in this survey of public opinion. Below are
given 88 statements which represent widely held opinions on various social
questions, selected from speeches, books, newspapers and other sources. They
were chosen in such a way that most people are likely to agree with some,
and to disagree with others.

151
 SEX AND PERSONALITY

After each statement, you are requested to record your personal opinion
regarding it. You should use the following system of marking:
+ + if you strongly agree with the statement
+ if you agree on the whole
0 if you can’t decide for or against, or if you think the question
is worded in such a way that you can’t give an answer
- if you disagree on the whole
— if you strongly disagree
Please answer frankly. Remember this is not a test; there are no ‘right’ or
‘wrong’ answers. The answer required is your own personal opinion. Be sure
not to omit any questions.
Do not consult any other person while you are giving your answers.
Opinion statements Your Social attitudes
opinion % yes answers
M F

1 Our treatment of criminals is too harsh; we

should try to cure them, not punish them 58 61
2 It will always be necessary to have a few strong,
able people actually running everything 72 78
3 The United Nations Organisation is useless and
does not deserve our support 9 9
4 Pacifism is simply not a practical philosophy in
the world today 39 28
5 The idea of God is an invention of the human
mind 63 44
6 We should not restrict immigration into this
country as we have done in the past 30 26
7 Private profit is the main motive for hard work 61 60
8 There is very little discipline in today’s youths 36 39
9 It is the moral responsibility of strong nations
to protect and develop weaker and poorer nations 75 75
10 Sex crimes, such as rape and attacks on children,
deserve more than mere imprisonment; such
criminals ought to be flogged or worse 25 31
11 Communists should not be allowed to hold jobs
in government service 33 31
12 We spend too little money on foreign aid 39 33
13 An occupation by a foreign power is better than
war 28 32
14 The average man can live a good enough life
without religion 82 77
15 In capitalist countries there is an inevitable
conflict between workers and employers 72 68
16 Great wealth should be shared much more than
at present 71 67
17 A white lie is often a good thing 79 85
18 The great increase in drug-taking by the young is
another example of how far our society has
deteriorated 31 37
19 The death penalty is barbaric, and its abolition
right and proper 64 61

152
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

M F
20 The police should have the right to listen in on
private telephone conversations when investigat¬
ing crime 32 32
21 The so-called underdog deserves little sympathy
or help from successful people 13 12
22 We have never, as a nation, fought an unjust war —-- 11 14
23 We should believe without question all that we
are taught by the Church 2 3
24 In this country, the most able rise to the top - 36 37
25 A firm should produce what is most profitable,
not what the government believes to be in the
national interest 32 23
26 The practical man is of more use to society than
the thinker 18 20
27 Even though the masses behave pretty stupidly at
times, I have a lot of faith in the common
sense of the ordinary man - 59 68
28 The maintenance of internal order within the
nation is more important than ensuring that
there is complete freedom for all 53 43
29 Poverty, mental illness and other problems are
a responsibility for the whole community 95 95
30 The dropping of the first atom bomb on a
Japanese city, killing thousands of innocent
women and children, was morally wrong and
incompatible with our kind of civilisation 56 72
31 Life is not perfect nowadays but it is much
better than it used to be 82 77
32 Sunday observance is old-fashioned and should
cease to govern our behaviour 77 71
33 Capitalism is immoral because it exploits the
worker by failing to give him full value for his
productive labour 34 35
34 Nowadays more and more people are prying into
matters which do not concern them 51 49
35 There are many advantages to having a king
or queen) to govern the country, provided they
do not have too much power 56 56
36 Most modem art is pretentious nonsense - 32 26
37 The ‘welfare state’ tends to destroy individual
initiative 39 34
38 A person should be free to take his own life,
if he wishes to do so without any interference
from society —--- 62 65
39 On the whole workers in this country are fairly
treated by their employers 62 69
40 If you start trying to change things very much,
you usually make them worse 19 20
41 Christ was divine, wholly or partly, in a sense
different from other men 29 40

153
 SEX AND PERSONALITY

M F

42 The nationalisation of the great industries is

likely to lead to inefficiency, bureaucracy and
stagnation 56 51
43 People should realise that their greatest
obligation is to their family 49 49
44 Business competition is necessary for national
welfare
45 Most strikes are caused by bad management 36 36
46 We ought to have a world government to
guarantee the welfare of all nations irrespective
of the rights of any nation 50 46
47 Sex relations except in marriage are always
wrong 5 7
48 The way wealth is distributed at the moment is
unsound and unjust 57 56
49 It is always a good idea to look for new ways of
doing things 96 93
50 There are no such things as ‘supernatural
powers’ 37 22
51 Economic security for all is impossible under
capitalism 37 36
52 The less government the better 33 22
53 There are many responsible positions for which
women are unsuited, such as judgeships, mini¬
sterial office and high positions in banking and
industry 18 12
54 Trade unions do more harm than good to
industrial progress 30 37
55 I would support my country even against my
convictions 17 15
56 Free love between men and women should be
encouraged as a means towards mental and
physical health 46 39
57 Scientific inventions have carried us too far too
fast; we should be given a resting pause now 26 33
58 Government nowadays is too centralised 55 52
59 Negroes are often denied opportunities for good
jobs and promotions that are given to white
people 82 86
60 Our nation is more democratic than any other
nation 37 37
61 Housing will never be adequate until the govern¬
ment acquires ownership of all land 19 12
62 It is wrong that men should be permitted greater
sexual freedom than women by society 77 81
63 The government should do a lot more to regul¬
ate the activities of labour unions 52 49
64 In taking part in any form of world organisation,
this country should make certain that none of its
independence and power is lost 36 50

154
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

M F
65 The practice of birth control should be dis¬
couraged 2 2
66 Modern adolescents are no more immoral than
were their parents or grandparents at their age - 77 73
67 There may be a few exceptions, but in general
Jews are pretty much alike - 20 21
68 Workers should take part in the running of
businesses in which they are employed '- 71 69
69 Women are not really the equals of men, and
never will be 24 28
70 Teachers have no business to take an active part
in politics 17 14
71 Censorship of books and films should be com¬
pletely abolished 74 57
72 A national health service does not give doctors
an opportunity to do their best for their patients - 25 33
73 Most politicians can be trusted to do what they
think is best for the country —- 37 34
74 Conscientious objectors are traitors to their
country, and should be treated accordingly - 5 6
75 The laws restricting abortion should be abolished - 63 66
76 The permissive modern ways of bringing up
children are an improvement on older methods - 69 68
77 All kinds of discrimination against the coloured
races, the Jews etc. should be made illegal,
and subject to heavy penalties --— 65 70
78 Democracy depends fundamentally on the
existence of free business enterprise 36 42
79 Slumps and unemployment are the inevitable
consequences of capitalism 33 28
80 The government must ensure above everything
else that unemployment is kept very low - 28 36
81 The school leaving age should be raised as much
as possible, whether young people want to stay
on or not 27 21
82 It seems to me that, whoever you vote for, things
go on pretty much the same 64 66
83 This country is just as selfish as any other
nation 77 77
84 In the interest of peace the private manufacture
of arms and ammunition must be abolished - 50 60
85 The nation exists for the benefit of the individual,
not the individual for the benefit of the nation - 62 61
86 When it comes to the things that count, all races
are certainly not equal 43 42
87 Stable peace will be possible only in a socialist
world 20 23
88 Morals in this country are pretty bad, and getting
worse 14 18
If you had to choose, would you vote for:
89 A Conservative candidate -

155
 SEX AND PERSONALITY

90 A Labour candidate -
91 A Liberal candidate -
92 Other (specify)-

Would you say that your family was:

93 Working-class -
94 Middle-class -
95 Other (specify) -

What is your occupation? (housewives please give

occupation of husband)
Would you clarify this as:
Higher professional / administrative -
Lower professional/administrative -
Clerical -
Skilled manual -
Semi-skilled manual -
Unskilled manual -

Previous work (Eysenck 1954) had shown women to be less tough-

minded and more conservative; the few differences in our table
that exceed ten points bear this out. Women are more in favour
of pacifism, opposed to the private manufacture of arms, and
consider the dropping of the atom bomb immoral; they are more
religious, emphasising belief in God, the divinity of Christ and
the existence of supernatural powers. They favour government,
censorship and world organisations as opposed to nationalism; in
other words, they are on the side of law and order, even in the
international sphere. None of these differences is large, and they
are all in the expected direction. It would not be very useful to
construct a masculinity-femininity scale from these items, as was
done with our sexual attitudes inventory; there simply are too few
differences between men and women in the field of general attitudes
to make this possible.
Table 4.2 gives the main results from a factor analysis of the inter¬
correlations (product-moment) of the eighty-eight items of the Public
Opinion Inventory. Three factors were extracted from the matrix
of intercorrelations and rotated by Promax to oblique simple
structure, the table gives the loadings resulting from this analysis,
which was carried out separately for men and women. The result¬
ing factors are very similar for the two sexes, and are identified as
1 conservatism
2 capitalism
3 toughmindedness.

156
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

Table 4.3 shows the intercorrelations between the three factors;

capitalism and conservatism show a moderate to high correlation,
and both correlate negatively and not at all highly with tough¬
mindedness. For all practical purposes toughmindedness is in¬
dependent of both capitalism and conservatism.

Table 4.2 Factor loadings of items on conservatism, capitalism and

toughminded factors

Conservatism Capitalism Toughmindedness

M F M F M F
1 -0-57 -0-58 - 0-25 -0-20 001 0-00
2 0-33 0-42 0-30 Oil 000 0-01
3 0-53 0-37 -001 -0-12 0-28 0-16
4 0-31 0-37 0-12 0-16 007 004
5 -0-07 0-04 -0-05 -Oil 0-63 0-72
6 -0-42 -0-36 -0-25 -0-35 -0-04 007
7 0-61 0-45 0-04 008 0-15 0-23
8 0-57 0-53 -0-09 -006 -0-07 -008
9 -0-15 -0-28 -0-34 -0-21 -0-30 -0-25
10 0-67 0-74 -005 -0-12 -0-04 001
11 0-46 0-48 0-18 0T7 -0-12 -0-15
12 -0-57 -0-34 -0-24 -0-31 -0-26 -0-14
13 -0-22 -0-02 -019 -0-24 -001 0-10
14 016 015 -Oil -009 0-65 0-71
15 0-25 0-16 -0-57 -0-47 0-14 016
16 -0-17 -0-09 -0-60 -0-69 -003 -009
17 0-15 0-24 0-10 006 0-28 0-41
18 0-54 0-44 -0-14 -013 -0-29 -0-34
19 -0-56 -0-61 -0-22 -0-19 -0-04 -0-17
20 018 0-30 0-14 009 -0-12 -0-02
21 0-40 0-45 0-09 008 0-14 016
22 0-52 0-57 -002 -0-15 -0-14 0-08
23 0-34 0-36 -0-23 -0-16 -0-38 -0-38
24 0-20 0-35 0-20 0-18 -009 0-05
25 0-37 0-43 0-25 0-21 008 003
26 0-50 0-54 -0-26 -019 0-04 0-03
27 0-30 0-17 -0-35 -0-19 -0-32 -0-14
28 019 0-42 0-30 006 -0-20 -0-02
29 -0-29 -0-30 -0-24 -0-21 -0-07 -008
30 -0-24 -0-28 -0-35 -0-32 -012 -0-17
31 -0-03 009 0-09 004 0-08 0-18
32 -0-02 006 0-07 -008 0-67 0-67
33 007 0-25 -0-82 -0-84 006 007
34 0-44 0-40 -0-28 -0-17 009 0-05
35 0-03 018 0-40 0-22 -0-21 -0-23
36 0-56 0-51 -0-02 -0-11 001 -0-07
37 0-42 0-40 019 0-22 0-02 003
38 013 0-12 -0-08 -009 0-45 0-44
39 0-15 0T0 0-57 0-55 -006 -010

157
 SEX AND PERSONALITY

Conservatism Capitalism Toughmindedness

M P M F M F
40 0-45 0-55 -0-14 -0-09 -005 -001
41 0-01 -0-02 0-05 006 -0-68 -0-70
42 0-34 0-25 0-45 0-46 0-06 -0-04
43 0-46 0-45 -009 -0-08 -0-18 -0 09
44 0-38 0-32 0-46 0-42 -006 -0-07
45 0-06 010 -0-49 -0-51 -009 012
46 -0-22 -0-12 -0-22 -0-41 -0-02 003
47 -0-04 0-24 -0-07 -0-20 -0-62 -0-56
48 -023 -0T3 -0-58 -0-59 -001 -006
49 -0-02 -0-11 0-02 001 0-29 0T3
50 -002 002 -0-17 -015 0-38 0-39
51 013 0-17 -0-68 -0-67 0-12 0-11
52 Oil 0-20 -0-05 -Oil Oil 0-24
53 0-54 0-43 -004 -Oil -003 -0-12
54 0-47 0-50 0-18 0-21 001 0-04
55 0-46 0-51 006 0-02 -013 -0-08
56 0-25 0-14 -0-31 -016 0-52 0-63
57 0-18 0-18 -0-32 -0-27 -0-17 -0-20
58 016 Oil -015 -0-24 -0-05 -0-02
59 -0-24 -0-25 -0-18 -0-19 0-10 0-02
60 0-27 0-28 0-09 0-03 -0-12 0 00
61 0-07 0-04 -0-71 -0-61 003 0-02
62 -0-34 -0-14 0-04 -016 0-12 0-17
63 0-35 0-35 0-44 0-43 - 006 -004
64 0-73 0-59 -006 0-08 -002 -0-05
65 015 0-25 -0-15 -0-21 -0-32 -0-25
66 -015 -0-25 0-14 0-11 0-30 0-23
67 0-57 0-62 -0-21 -0-21 -003 0-04
68 -0-21 -0-26 - 0-46 -0-45 -015 -0-04
69 0-47 0-51 -0-07 -0-05 -010 0-00
70 0-50 0-58 0-02 -0-03 001 0-02
71 005 -0-07 -0-13 -007 0-58 0-50
72 0-41 0-40 Oil 0-07 006 -002
73 -004 019 0-20 0-20 -0-25 -0-09
74 0-52 0'70 -002 -0-14 -0-03 001
75 0-02 -005 -0-06 -0-08 0-60 0-50
76 -0-18 -0-18 -008 -0-22 0-45 0-29
77 -0-20 -0-28 -0-17 -0-23 005 -005
78 0-51 0-40 0-21 0-28 -0-07 -010
79 010 019 -0-75 -0-72 0-06 004
80 0-49 0-39 -065 -0-58 -007 -006
81 0-08 0-26 -0-34 -0-35 -0-02 -009
82 0-42 0-32 -0-20 -0-18 0-14 0-11
83 -0-08 -016 -003 -0-03 018 013
84 -0-12 -013 -0-45 -0-43 -002 -Oil
85 -0-26 -0-21 Oil 0-06 0-11 0-04
86 0-49 0-47 0-15 012 0-06 0-04
87 005 0-00 -0-74 -0-74 009 0-01
88 0-32 0-35 -0-10 -0-16 -0-45 -0-43

158
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

Table 4.3 Intercorrelations of three superfactors for males (above leading

diagonal) and females (below leading diagonal): in brackets, intercorrelations
between corresponding scales

Conservatism Capitalism Tough¬

mindedness
Conservatism - 048 ( 0-49) -0T9 (-0T7)
Capitalism 0-43 ( 0 40) - -0T6 (-0T9)
Toughmindedness -0-27 (-0-30) -0T6 (-0-22) ——

Items with high loadings (above 0-35) clearly identify these

three factors. The most clear-cut factor of all is perhaps capitalism
v. socialism; the main items with high loadings on this factor
are the following: capitalism does not mean conflict between
classes (item 15); great wealth should not be shared (16); little
faith in common man (27); capitalism not immoral (33); kings are
advantageous to country (35); workers are treated fairly (39);
nationalisation is inefficient (42); business competition is necessary
(44); strikes are not caused by bad management (45); distribution of
wealth is not unjust (48); economic security is not impossible under
capitalism (51); housing will be adequate without nationalisation
of land (61); the government should regulate the unions (63); workers
should not have a part in the running of business (68); slums and
unemployment are not the inevitable consequences of capitalism
(79); low unemployment not absolute priority of government (80);
the private manufacture of arms need not be abolished (84); and
stable peace is not only possible under socialism (87). These are
the major items for males; to these should be added the following
from the females: immigration should remain restricted (6); we
ought not to have a world government (46); and the school leaving
age should not be raised (81). The nature of these items, and in
particular the highest-loading items, leaves little doubt about the
nature of this factor.
The conservatism factor is characterised by the following items
having loadings above 0-35: Treatment of criminals is not too harsh
(1); United Nations is useless (3); immigration should be restricted
(6); private profit is the main motive (7); there is little discipline
in modern youth (8); sex crimes deserve flogging (10); communists
should not be allowed to hold office (11); foreign aid is not too little
(12); drug-taking is an index of social deterioration (18); death
penalty is not barbaric (19); the underdog deserves little sympathy

159
 SEX AND PERSONALITY

(21); our nation has never fought an unjust war (22); firms should
be profitable (25); the practical man is worth more than the
thinker (26); there are more people prying (34); modern art is non¬
sense (36); the welfare state destroys initiative (37); change is for the
worse (40); one’s greatest obligation is to the family (43); business
competition is necessary for the nation’s welfare (44); women are
not suited for many positions (53); trade unions do harm (54);
always support one’s country (55); the government should regulate
unions (63); preserve national independence (64); Jews are all
pretty much alike (67); women are not the equals of men (69);
teachers should keep out of politics (70); national service not good
(72); conscientious objectors are traitors (74); democracy depends
on free business enterprise (78); low unemployment not primary
consideration, vote is useless (82); and all races are not equal (86).
To these the women added the following items: few strong, able
people to run everything (2); pacifism not practicable (4); believe
all we are told by church (23); the most able rise to the top (24);
order more important than complete freedom (28); morals in this
country are pretty bad, and getting worse (88).
The third factor, toughmindedness, emerges less clearly than
the other two, probably because the choice of items was directed
more at getting capitalism and conservatism more closely identified.
Nevertheless, the items overlap considerably with the usual ones
that have defined this factor in the past. They are: God is an
invention of the human mind (5); the average man can live a
good life without religion (14); we should not believe what we are
told by the church (23); Sunday observance is old-fashioned (32);
euthanasia is a good thing (38); Christ was not divine (41); disagree
that sex except in marriage is wrong (47); there are no superior
powers (50); free love should be encouraged (56); censorship should
be abolished (71); abortion laws should be abolished (75); per¬
missive ways of bringing up children are good (76); morals nowa¬
days are not bad (88). To which the females add the following: a
white lie is sometimes a good thing (17); and the increase in drug¬
taking is not an index of social deterioration (18). It would clearly
be possible to regard this factor as one opposing religion to per¬
missiveness, or ethical to hedonistic ideals, or idealism to realism;
however, there is little point in discussing at length the name of
the factor when its nature is fairly clear. We shall continue to call
it toughmindedness here, but bear in mind that the items making

160
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

up this factor provide little more than a skeleton of the larger

number previously used to define this factor.
Three scales were drawn up to measure these three factors; the
items contained in these scales, and their keying, are given in
Appendix E. The reliabilities of these scales (alpha coefficients) were
found to be as follows (males first, females second); capitalism:
0-84, 0-82; conservatism: 0-89, 0-89; toughmindedness; 0-76,
0-76. These are adequate for our purposes.
Sex differences on these scales are small or non-existent for con¬
servatism and capitalism, but marked on toughmindedness. For
conservatism, the scores of males and females are 14-09 + 7 08
and 14-55 + 7 00; for capitalism they are 1T32 + 4-50 and 10-85
+ 4-12. Neither of these differences is significant. For toughminded¬
ness the scores are 12-40 + 2-89 and 11-60 + 2-95; this difference is
significant at the 0 001 level. The result, showing males to be more
toughminded, is fully in line with earlier work. The difference for
conservatism, while in the right direction according to earlier work,
is much too small to be of any practical significance.
In addition to the three-factor solution just described, which
essentially gives us the major or higher-order factors, it was also
thought interesting to make a primary factor analysis, using twelve
factors and rotating these by Promax into simple oblique structure.
Seven of these factors were easily and clearly interpretable; load¬
ings on the items above 0-35 are given below, in connection with
a brief discussion of each factor. Factor 1 is clearly ‘Capitalism’,
very similar in nature to the three-factor solution.

Factor 1: Capitalism

Item Loading Item

Males
2 0-51 Few strong, able people to run government
15 -0-52 Capitalist countries conflict between workers and
employers
16 -0-63 Share wealth
24 0-35 ■ Most able rise to top
33 -0-80 Capitalism immoral
35 -0-52 Royalty an advantage
39 0-63 Workers treated fairly
42 0-55 Nationalisation inefficient
44 0-63 Business competition needed
48 -0-56 Distribution of wealth not fair
51 -0-72 Economic security impossible under capitalism

161
 SEX AND PERSONALITY

Item Loading Item

Males
61 -0-69 Housing inadequate until land nationalised
63 047 Regulate labour unions
68 -0-35 Workers share running of business
78 0-39 Democracy depends on free business
79 -0-77 Slump result of capitalism
80 -0-59 Government must keep unemployment low
87 -0-79 Stable peace only in socialist world

Females
6 -0-36 Restrict immigration
15 -0-60 Conflict workers and employers in capitalism
16 -0-65 Share wealth
33 -0-82 Capitalism immoral
37 0-35 Welfare state destroys initiative
39 0-64 Workers fairly treated
42 0-41 Nationalisation inefficient
44 0-40 Business necessary
48 -0-58 Wealth distribution unsound
51 -0-71 Economic security impossible under capitalism
61 -0-56 Ownership of land
63 -0-49 Government regulates unions
79 -0-80 Slumps consequence of capitalism
80 -0-44 Unemployment must be kept low
87 -0-69 Peace only in socialist world

Factor 2, at least for the males, is similar to the toughmindedness

one already discussed; for the women, the religious part of this
factor is overpoweringly strong, so that for them it might be better
renamed ‘Anti-religionism’ or something of that kind.

Factor 2: Toughmindedness

Item Loading Item

Males
5 0-70 God invention of human mind
14 0-70 Good life without religion
32 0-57 Sunday observance old-fashioned
38 0-36 Euthanasia
41 -0-72 Christ divine
47 -0-56 Sex outside marriage wrong
50 0-56 No supernatural powers
56 0-47 Free love good
71 0-53 Abolish censorship
75 0-55 Abortion laws abolished
76 0-45 Permissiveness in education good
88 -0-43 Morals bad in this country

162
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

Females
5 0-79 God invention of human mind
14 0-72 Life without religion
23 - 0 41 Believe all we are taught by Church
32 0-54 Sunday observance old-fashioned
41 -0-82 Christ divine
50 0-68 No supernatural powers

Factor 3 seems to be concerned with problems of discrimination,

particularly discrimination against women; it is accordingly named
‘Discrimination’.

Factor 3: Discrimination
Item Loading Item
Males
53 0-65 Women unsuited for responsible positions
62 -0-61 Men should not be permitted greater sexual
freedom
67 0-36 Jews pretty much alike
69 0-70 Women not equals of men
70 0-35 Teachers not in politics
77 -0-38 Discrimination illegal
Females
53 0-71 Women unsuited for responsible positions
62 -0-49 Wrong that men have greater sexual freedom
65 0-53 Discourage birth control
69 0-72 Women not equals of men
70 0-40 Teachers not in politics
74 0-42 Conscientious objectors traitors

Factor 4 is concerned with pacifism (strictly, anti-pacifism); all

the items are in accord with this interpretation.
Factor 4: Anti-pacifism
Item Loading Item

Males
4 0-62 Pacifism impracticable
13 -0-66 Occupation by foreign power better than war
46 - 0-41 World government
84 -0-36 Abolish private manufacture of arms
Females
4 0-51 Pacifism no good
13 -0-68 Occupation by foreign power better than war
55 0-37 Would support country against convictions
60 0-47 Our nation more democratic than others
82 -0-42 Whoever you vote for, things much the same
83 -061 This country just as selfish as others

163
 SEX AND PERSONALITY

Factor 5 is very well defined; it is the conservatism factor already

encountered.

Factor 5: Conservatism

Item Loading Item

Males
1 -0-42 Treatment of criminals too harsh
8 0-48 Little discipline in today’s youth
10 0-46 Flogging for sex crimes
18 0-69 Drug-taking bad
19 -0-50 Death penalty barbaric
20 0-49 Police right to listen
36 0-67 Modern art pretentious nonsense
37 0-48 Welfare state destroys initiative
54 0-42 Trade unions do harm
63 0-45 Government should control unions
74 0-39 Conscientious objectors traitors
88 0-50 Morals bad in this country

Females
7 0-65 Private profit motive
8 0-43 Little discipline in today’s youth
24 0-42 Most able rise to top
34 0-62 Prying
35 0-52 Royalty advantage
40 0-69 Changing things makes them worse
42 0-43 Nationalisation inefficient
54 0-42 Trade unions do harm
60 0-43 Our nation more democratic than others
64 0-39 In world organisation, not lose independence
72 . 0-50 National health service not good
78 0-46 Democracy depends on free enterprise
82 0-48 Whoever you vote for, things much the same

Factor 6 might be labelled ‘Idealism’; clearly the items suggest

some such underlying attitude.

Factor 6: Idealism

Item Loading Item

Males
9 -0-56 Moral responsibility of strong nations
21 0-45 Underdog deserves little sympathy
29 -0-60 Poverty responsibility of whole country
46 -0-47 World government
49 -0-44 Look for new ways of doing things
68 -0-51 Workers should take part in running business

164
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

Females
1 0-37 Treatment of criminals harsh
3 -0-38 United nations useless
4 -0-41 Pacifism no good
9 0-62 Strong nations protect weaker
10 -0-38 Flogging
12 0-36 Spend too little on foreign aid
19 0-39 Death penalty barbaric
29 0-63 Poverty responsibility of community
30 0-45 Atom bomb immoral
49 0-41 New ways of doing things
77 0-41 Discrimination illegal
84 0-38 Nationalise arms industry

Factor 7 may be called ‘Pro-government’, in the sense that govern¬

ment is not distrusted and disliked.

Factor 7: Pro-government

Item Loading Item

Males
28 0-36 Internal order more important than freedom
34 -0-47 People prying
52 -0-76 The less government the better
58 -0-65 Government too centralised

Females
3 -0-45 United Nations useless
52 -0-56 The less government the better
57 -0-58 Too much scientific invention
58 -0-61 Government too centralised
73 0-41 Politicians can be trusted

The remaining factors are difficult or impossible to identify,

and will not be listed.
Table 4.4 shows the correlations between the primary factors and
superfactors in the social attitudes field on the one hand and the per¬
sonality factors, social class and age on the other. Significant cor¬
relations have been italicised; these should be taken seriously
only when there is agreement between males and females, at least
on the direction of the correlation found. It may be useful to discuss
the major findings by a consideration of the variables in each
column. Beginning with P, we find that high P scorers are predomi¬
nantly toughminded, and generally anti-minded; they are against
conservatism, against the government, against capitalism, against
idealism and against pacifism; most of these correlations are fairly

165
 as OO J— (
o o <N O © © Os
Uh O o o
© 6 © 6 © © © 6 o <6
o 1 1
oo 1 i
<
o Tf 00 OS so SO 00
o *“h O o o o O o
© © © © <S © © 6 o <6
1 1 1 l

oo sn (N m

CO
u- O
6
o
©
p
©
o
©
p
©
3
6
3
6
00
o
o
as
o
6
s6
CO 11 1 i
-a *2 1 1
o
field with personality, class and age. Italicised correlations significant
Table 4.4 Correlations of primary and superfactors in the social attitude

sn Tf Tf 00 r*
s o o o o o o o o
<6 © © © 6 © © 6 6
1 1 i

u. s© oo
o rsj
<N
<N
nh
<6
8
co
p
*o
o
<o>
fNj
© CO © © <s o <s
*■4 1 1

<o> ra Cb <N so <N| Os **^l

s ’■"I p o <N O O <N
o <6> © 6 <s 6 6 o
1 1 * i 1
at p < 0 05 level

r- SO <N so <N Oo so CO
Pi o o o O <N o p o
© 6 6 © <6 6 <6> <6 6 6
1 1 I. 1 1 1
oo r*» (N so
s
p p o ’"H O o O p
© © 6 <6 6 © o <6 o
1 l 1 I

cn ■nr cs o> r- nf <N

Uh o o o o o o o 8 o
© © 6 © © <o> o o 6 o
1 l 1 1 1 1

tj- so so m 00 (N
S
p s o p p O o
Tf-
o o o
© 6 © © 6 6 6 6
6 6
l 1 1 1 1 i

^h so <N nt so 00 Os
<N »—H p 00
Pi <N o **N
<6> 6 <6 6 © 6 <6
1 l l l l 1
a.

»n 00 m as
(N O 00 OnJ 00
§ o O o o o
<6 © 6 © cS o 6 o <6
1 1 1
1

CO CO
co
<L> *-> CO
a Q
a CS
o §
OJ +3 g g g c g
M
T3
d>
CO CO
s
CO
no
.5
g
d
.s
CCS
o c3
00
13
§
E
<D
CO
13 s
CO
n3
.s
c p ccj r*
Ph
<D > >
u
43 40 a <D JU O
'a
oo *c
o 1 CO *i 00
1
<D
CO 3 00
3 CO CS 'a
cd o c O o C p
U H 0 <
C Ui O rt o
U < cu O O
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

low, with the exception of those linking P with toughmindedness.

This is not an unexpected finding; in the Manual of the E.P.Q.
(Eysenck Personality Questionnaire) (Eysenck and Eysenck 1975)
we have already suggested as an alternative name for P the term
‘toughminded’.
The personality factor E has hardly any significant correlations
with the social attitudes, and does not seem to be related to any
of the variables here considered. N shows some slight correlation
with conservatism (particularly among the women), but this corre¬
lation is too small and uncertain to be readily interpreted. L is
clearly opposed somewhat to toughmindedness, and positively cor¬
related with conservatism. These correlations are not large, and
should not be overinterpreted.
Class and age correlations are small throughout. Middle-class
respondents are in favour of capitalism and slightly against con¬
servatism (with a small c!); this is not unexpected (Eysenck 1975b).
It is possible that the correlations are not larger because of the
difficulty in determining social class from the answers given in a
questionnaire; there is considerable inaccuracy involved, and it
is possible that this unreliability has caused the observed correla¬
tions to be smaller than they would otherwise have been (par¬
ticularly for women).
Age correlates slightly with conservatism, but perhaps much
less so than one might have expected. It also appears that the
young are more toughminded; this agrees with the observation
that P scores decrease with age, and that P is correlated with tough¬
mindedness.
We must now turn to a consideration of the interrelations (if any)
between the social attitude variables and the sexual attitude vari¬
ables. We would not expect much in the way of relationships for the
capitalism factor; conservatism might be expected to show some cor¬
relations with such sex factors as permissiveness, but on the whole
conservatism too is not likely to show any very close relationship.
It is with respect to toughmindedness that our main expectations
must lie; toughminded persons, as we have seen, are high P scorers,
and the close relation between P and several of the sex scales has
already been shown.
Let us consider first of all the primary social attitude factors.
Conservatism does indeed, as expected, correlate negatively with
permissiveness {r— —0-22), and also with physical sex (r= —0-20).

167
 SEX AND PERSONALITY

Toughmindedness correlates 0-70 with permissiveness, 0-28 with

pornography, 0-40 with impersonal sex, 0-20 with aggressive sex
and 055 with the Libido superfactor. Capitalism shows a correlation
only with permissiveness (r= —0-22). Pacifism correlates with physi¬
cal sex (r= —0-22) and with dominance (r= —0 37). For the social
attitude superfactors the relationships are similar. Capitalism corre¬
lates only with permissiveness (r= —0-22); conservatism correlates
with permissiveness (r=— 0-22), physical sex (r= — 0-23) and
dominance (r= — 0-28). Toughmindedness correlates with permis¬
siveness (r=0-74), pornography (r=0-24), impersonal sex (r=0-41)
and the Libido superfactor (r=0*56). These are the values for the
males; those for the females are similar.
For the women, the primary factor of conservatism correlates
with permissiveness (r=— 0-23), physical sex (r=— 0-22), and
dominance (/*=— 0-33). Toughmindedness correlates with permis¬
siveness (r=0-38), impersonal set (r=0-20) and the Libido super¬
factor (r=0-31). Discrimination correlates with permissiveness
(r=0-32), physical sex (r=0-23) and dominance (r=0*30). Pacifism
correlates with permissiveness (r=— 0-24) and dominance (r=
—0-24). The anti-government factor correlates with aggressive sex
(r=0-20). Turning now to the superfactors, we begin with conser¬
vatism; this correlates with permissiveness (r=—0-34), sexual
excitement (r=-0-22), shyness (r=0-35), pornography (r=-0-25),
neurotic sex (r=0-42) and the Libido superfactor (r= —0-28).(l)
Capitalism correlates with physical sex (r= -0-25) and dominance
(r=0-42). Toughmindedness correlates with permissiveness (r—
0-67), and with excitement (r=0-25); with the Libido superfactor the
correlation is 0-58. Most if not all of these correlations make sense
as they stand, and do not require much discussion.
It will be remembered that almost half the sample were constituted
of matched pairs, i.e. men and women either married or living
together on a semi-permanent basis. It is interesting to know to what
extent the social attitude of such pairs are similar, i.e. to what extent
there is assortative mating in this field. Table 4.5 shows the results;
it will be noted that assortative mating is quite strong for the three
superfactors, and rather less strong for the primary factors. It must
of course be remembered that the values given in this table are not,
strictly speaking, evidence for assortative mating; the attitudes of
the couple may have changed towards greater conformity after they
started living together. It is a moot point whether this fact would

168
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

disqualify the values in the table from being considered evidence

of assortative mating; in one sense of the term the time of greatest
interest would be not that of beginning an association, but rather
that of having offspring. We have no data on which to base such a
comparison.(2)

Table 4.5 Assortative mating: correlations between males and females

married or living together

1 Capitalism 0-59
2 Toughmindedness 0-50
3 Conservatism 0-42
4 Anti-idealism 0-36
5 Anti-pacifism 0-28
6 Pro-government 0-21
7 Discrimination 0-18
SF1 Conservatism 0-67
SF2 Capitalism 0-65
SF3 Toughmindedness 0-60
Class 0-40
Age 0-88

Also given in table 4.5 are values dealing with social class and
age. It will be seen that for age homogamy is very clear; this is only
to be expected. For social class the correlation is much lower; this
may in part be due to the difficulties of ascertaining social class at
all, and in particular of ascertaining it in women. The figure quoted
is probably a lower limit; the true value is likely to be in excess of
this figure. The table as a whole enables us to say that, as in the
case of intelligence, and unlike the case of personality, like mates
with like as far as social attitudes are concerned; this was our con¬
clusion too in the case of sexual attitudes.
One last comparison requires to be made, namely that between
people who state that they voted for one of the major parties
(Conservative, Labour, Liberal) or for some other party (which in
the context of the sample in question means for all practical pur¬
poses the Communist Party, or some other left-wing extremist
group). (In Scotland or Wales, or in Northern Ireland, the term
‘Other’ would of course in the majority of cases refer to nationalist
candidates, but as our sample was exclusively English this point
does not arise. There may have been one or two National Front
sympathisers in the ‘Other’ group, but remarks and comments
written into the inventory suggest that this is not a likely con-

169
 N.S.
N.S.
QMMWOO

0-05
&DC/3!/3(>9t>r!&Oiy3(Z)C/5<y}

0-01

0-01
5 Z z Z 6 o 6 o z Z Z Z Z Z Z Z Z z

GO fi'Xicooi'i'fOf'NKiix)'Cnoifioo®oom

084
1-80

084
3-32
5-58
1f)|ncoaq\M^^oai\oovoifnonHa\

0-70±
9-42+

12-34 ±
+1 +1 +1 +1 +1 +1 4-1 4-1 4-1 4-1 4-1 4-1 4-1 44 4-1 4-1 44 44

0-73 ±

26-00±
<D
Mean social attitude scale scores for conservative, labour, liberal

xi cnT-Hvoooc*->»oo\Tfc4Tfo\ooTfrr^ooOi-Hoo
^.Tlr!HooHa\h'HOhcoiorlONOfSM
ff)Ocbv*>^iTrr'Nc»oio^bfSvb*H\brj(^’H
I CO «0 o T—<

oo
On
m Tf SO 00 00 fN Sas-not^r-r^r-sor^

2-77

3-26
0-78

6-94
VO Tf »0

2-21
• oo ocp»pTfrnooTtr4r^
^'’t4nr;'d\6\NrnN-H^rn^^HrH6
O +i +i +i +i +i +i +i +i +i +i +i +i +i +i +i +i +i +i

8-20±
0-32+
1-70+
11-80+
23-58±
>
Ohfnh^^4^cohNvONhNin\ooo
and ‘other voters: males

<D On o 'O -H
X>

C\
209

1-74

6-69
3-11
oo<^T7<‘pc\cpioTf(^oc^,^*oo\»o(STj’r'>* 0-51
«HTj-10njHOd\NfS(S'-lNr»6*H»H»H6
+1 +1 +1 +1 +1 +1 +| +1 +1 +1 +| +| +| +| -f| +| +| +|
12-23 ±
8-93±
0T7±

29-76+
1-48+
o
> &f^^l2SC,rlNN0ooo-Ht'pn««mM
cS^oo*^-cpt^-t^'0\CT\(sr7'i—looooootno
3
NOO\lOC004’,Hh,<0'0(^l0^h>HNrt
^ OO >D MD *-i
O
X)
hJ
2-94
046
T86

7 16
1-38

o trfr>^op^'r'^-t7-t7'q\vp(svocnT.Ttinoo
> «^tTtrr1oooOO\(NtS-C'C(SC^^^Tl(^,o
a> vo +1 +1 +1 +1 +1 +1 +1 +| +| +| +| +| .j.| +| +| +| +! +|
9T5±

12-91+
24-13±
0T6±
1-20+

> (N
?fSvp^H(S'O»-ifSTi-'<*-0\ir)t-~O(S00mO
ts—■00'0(SVD(S—
Table 4.6

a
o
u
factors
Super-

CO
CO
2

<d
1

X
t£ a>60 <L> X
Sex Behaviour
Sex Behaviour
Sex Behaviour

T3 D
j
Satisfaction 1

6 <D <U X ««
, *o a§ 03 <U
ill s '3G o co (D
£5 «5 ,ZL *3 o G oj >
r X3 q w 5> •■* O — *3 ’«>
s 3 4-s 03 O oo 00 CO -e ti
3 a o s
Libido

u co X5 Oj) G *8 |
"rt G o< p g & G >rd 00-ft CO J-H
►. O c3 O ni ■§ D £5 ^
•2 * xi 5P ££
£ CO £ on Oh Q w 0h <
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

tingency as far as the majority of respondents are concerned.) It

is interesting, and reassuring as far as the representativeness of
the sample is concerned, that the numbers ‘voting’ for each party
are roughly in the same proportion as was found in the two general
elections held in 1974. The actual numbers were (numbers for males
precede numbers for females): Conservative, 126 and 134; Labour,
149 and 157; Liberal, 98 and 110; Others, 48 and 28. A few subjects
who did not fill in the information on voting behaviour have been
omitted.
Mean scores and standard deviations of voters for these four
parties were calculated on the following scores: P, E, N and L;
conservatism, capitalism and toughmindedness; the eleven primary
sex factors (using questionnaire scores, not factor scores); the three
sex behaviour scores; and the two sex superfactors, libido and
satisfaction. The actual scores obtained are given in tables 4.6 and
4.7; also given are the p values resulting from analyses of variance
for each score. The figure must speak for themselves, but a brief
discussion may be of value. In this discussion we have concentrated
on issues that are of interest, and where the significance values and
the agreement between males and females suggest that the results
are replicable.
As far as the personality scales are concerned, the following
points seem worthy of mention. On P, for both males and females,
‘other’ voters are clearly much higher than voters for the orthodox
majority parties; this agrees well with the findings of Eaves and
Eysenck (1974) that there are genetic links between toughminded¬
ness and P, and Eysenck’s (1954) finding that extremist parties
(fascists and communists) are high on toughmindedness. In agree¬
ment with this, we find that in our present study also ‘other’ voters
have higher toughmindedness scores in both the male and female
samples. Eaves and Eysenck (1974) also found E associated
genetically with toughmindedness; in the present study ‘other’ voters
have high E scores in both samples, but conservative voters also
have high E scores, and are less toughminded than either ‘other’ or
Labour voters. It is possible that these correlations are predicated
on different aspects of extraversion; perhaps the correlation with
conservative voting is related to sociability, that with ‘other’ voting
with impulsiveness. Only further research can decide on this point.
For N the two sexes give contradictory results, but for L Conserva¬
tives clearly have the highest, and ‘other’ voters the lowest scores.

171
 0001
N.S.
COW

005
0'05
0-01
^-h vn v~> C/D o o
O O ojop oo^-^oooop^:^:
000^00 000 X^OOOOO^^

CN m T-H

6-39
<N SO

0-69
T03
00

3-11
,,cf m

T61
SO 0 0 00 00 m
OO ON p Tf SO so 00 O cp p O p p1 *7* p tJ- 00
(N T-Hcb cb *b> rb so cb cb iH cb cb cb T-H (N T-H T-H 6
<N T-H

12-11+
19-70±
0-61 ±
8-82 +
046 +
<D +1 +1 +1 +1 +1 +1 +1 +1 44 +1 +1 +1 +1 +1 +1 +1 +1 +1
(S ON m CN CS 0 ON t-H O m 0 ,d- so T-H 0
X
liberal

p r- Tf © OO p Cp ON p cp sp 00 cp VI 00 ON cp
rb cb T-H »o cb 4f tH 00 tT 4f cb T-H VT T-H cb T-H
t-H OO v-> so
labour,

o\

T70
T07
2-55

3-25
7-35
On ON SO (N so so <N _ ON cN O
p rp tp Cp rp T-H ON ON O cp tJ- O h 10 O ^ h
T-H cb Tf’ cb 00 cb On cb cb cb cb cb cb
Mean social attitude scores for conservative,

T09+
0'53 +
11-81+
18-75+
8*01 ±
o +1 +1 +1 +1 +1 44 4-1 4-1 4-1 4-1 4-1 4-1 4-1
> t-H <N O vn T-H r^ T-H CN O <N
rp rp rp t- so rp SO p cp cp O Cp p OO'O't
T-H T-H cb so rb cb O 0 cb b- VI cb rb
ON so SO T—'
and ‘other’ voters: female

T-H
CD
X

•o Q m 00 On T-H r^ CJ ON so m ON Tf
1T0
T09
r^ ON VN
T83

2 87
6-23
p T-H p cp p 4? P p tp 00 Tp p p <N rp p rp 00
b-t 4r 4f rb ON ON ON rb rb ** rb rb rb ^H rb ^H T-H 6
8-86 +
0'57±

20 11+
0-62 +
13-08+
4-1 4-1 4-1 -H 4-1 4-1 4-1 -H 4-1 4-1 4-1 4-1 4-1 4-1 4-1 4-1 +1 4-1
,-H so ,-H 00 m 00 VN ON 00 ON O 00 m (N ON T—1
o p p 00 so cp Cp p cp p rp cp p p cp OO 99 rp T-H
> rb rb 6 so cb rb rb bn On 10 rb cb T-H sb TH rb rb
5-1 00 VN so T-H
3
o
X)
cd
hJ
T-H 00 t-H r- VN 00 00 fN r- IO On rN c-"
2-37
1T9

3-14

VN O CO Tf
7-18
I’ll

cp p p 99 p T-H p O rp T-H rp rp 99 cp cp p p so
T-H cb b- rb so cb On rb rb rb rb rb rb rb vb 6
O
8-32 +
090 +

17-93±
0-51 +
12-89+

+1 -H 4-1 4-1 4-1 +1 4-1 4-1 4-1 4-1 44 -H 4-1 4-1 4-1 44 4-1 4-1
<D . t-H O r^ O T-H 0 VN ro so 00 O VN 00 ON T-H
> cp p P ON tj* 99 so rp p p q\ qp ON °9 rp cp p p
X rb so rb rb tb 0 ON b- rb rb rb vb rb rb T-H
ctf ; T-H rb
Table 4.7

T-H T-H T-H r- VN T-H

>

c
o
U
Sex Behaviour 2
Sex Behaviour 3
Sex Behaviour 1

CD
G x
<D
■3
Satisfaction

T-H X
cd Cl <D
G
O
cd G
co H
CD
CO . 12 g> S “ 3
Libido

13 x t«o
6CL) ft G & E o-g S>-~
^ 2 •“ x
Ph
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

The first of these two findings is in good agreement with the

findings of Wilson and Brazendale (1973), who found a correlation
of 0-26 between conservatism and L\ they also found a correlation
of -0-26 between L and realism, which is their name for tough¬
mindedness (Wilson 1973). Wilson and Brazendale found other
correlations which do not agree with our findings, e.g. a significant
correlation between P and conservatism and a negative correlation
between E and conservatism. They also found a positive correlation
between E and realism—toughmindedness, which is in agreement
with our general findings. Their results were based on ninety-seven
female student teachers, and it is possible that in such small and
rather unrepresentative groups population trends may be reversed;
thus student groups tend to be exceptionally ‘progressive’, and
it may need a high P to confess to conservative leanings! Similarly,
the high popularity of ‘progressive’ ideas may make conservatives
less popular (low E) and high ‘realism’ scorers more popular (high
E). Clearly the relationship between personality and social attitudes
requires much further study. It should perhaps be added that the
correlation between L and conservative voting is more likely to
be explained in terms of the conventionality-conformity aspect of
this scale than its dissimulation-measuring properties; the corre¬
lation between L and N is no higher in the conservatives than
the other groups, and as Michaelis and Eysenck (1971) have shown,
it is this correlation that indicates dissimulation.
The relationship between voting behaviour and the social attitude
variables is very much as one might have expected, with conserva¬
tives, liberals, labour and ‘other’ voters having scores on conserva¬
tism and capitalism which decline in that order. We have already
commented on the high toughmindedness scores of ‘other’ voters;
at the tenderminded end we find liberals (which agrees with
Eysenck’s 1954 finding) and (for the women) conservatives; it is
not clear why women should be different from men in this respect.
With this one exception, however, results support the original
hypothesis suggested by Eysenck (1954), to the effect that political
parties can be arranged in the shape of a horseshoe in the two-
dimensional space generated by the conservatism-radicalism and
the toughmindedness-tendermindedness factors, with liberals inter¬
mediate on the former, and most tenderminded on the latter factor,
and conservatives and fascists diverging upward towards tough¬
mindedness and right towards conservatism, and labour voters and

173
 SEX AND PERSONALITY

communists diverging upward towards toughmindedness and left

towards radicalism.
When we turn next to the sexual attitudes variables, we find
smaller differences between voters for the different parties; this is of
course not unexpected in view of the fact that political party pro¬
grammes seldom if ever contain proposals relevant to this field.
On permissiveness the conservatives have the lowest scores, the
‘other’ voters the highest; this is not unexpected. On satisfaction
conservatives come highest; this may be simply a display of con¬
ventional sentiments, or it may reveal the more satisfactory nature
of old-fashioned family-based mores. Our data do not enable us
to answer this question. On neurotic sex it is difficult to see any
pattern in the responses of the two sexes; the same is true of imper¬
sonal sex. Pornography is regarded least unfavourably by liberal
voters, on the whole. Sexual shyness is least among conservatives,
greatest among ‘other’ voters; this may be linked with the high extra¬
version of the conservatives (sociability?), but does not go well
with the high extraversion of the ‘other’ voters (unless we agree
that this is produced by some other sub-trait of E than sociability).
Prudishness, disgust and excitement do not convey a very dissimilar
picture for voters for the different parties; neither does physical
sex. Aggressive sex shows a slight trend for ‘other’ voters to have
high scores; this agrees with Eysenck and Coulter’s (1972) finding
of greater aggressiveness among communists and fascists.(3) On the
superfactor scores of satisfaction and libido, we note again that
conservatives tend to have the higher satisfaction scores, although
this finding is not strong enough to deserve much credence until
replicated. For libido the results for the two sexes are too contra¬
dictory to make it possible to draw any conclusions. None of these
sexual attitude differences is very large or compelling, and the find¬
ings must of course be accepted with particular caution in view
of the small size and the not completely representative nature of
the sample.
The sex behaviour data are also difficult to interpret. On sex
behaviour 1 (Have done ...) the only consistent finding is the low
position of the liberals; there is a tendency in this direction also in
connection with the libido superfactor scores. This behaviour syn¬
drome may in turn be related to the tenderminded attitudes of
the liberals, i.e. their idealistic view of the world. Seen in this
context the findings (which of course require confirmation) possibly

174
 SOCIAL ATTITUDES AND SEXUAL BEHAVIOUR

make some sense. They are partly confirmed by the high scores of
the liberals on sex behaviour 2 (Have done and didn’t like ...);
conscience thus doth make cowards of us all. On sex behaviour 3
(Have not done and would like to do ...) the results are not at all
clear, and do not agree for the two sexes; it is probably best not to
try to interpret the results.
On the whole, we may perhaps say that such results as have come
up in connection with political party preference do not show any
strong relation with sexual attitudes and behaviour, and that such
rather weak associations as could be observed are in line with com¬
mon sense and with theoretical predictions based on previous work.
Clearly, much further work will be necessary to establish the degree
to which we can put any trust in these generalisations; at best they
serve to formulate more detailed hypotheses for testing. The same
caution should be applied to the interpretation of the apparent ten¬
dencies for females to show a much closer and more significant
relationship between voting behaviour and sexual attitudes. Perhaps
it is true, as is sometimes said, that women look upon everything
from the point of view of sex, but these data are hardly sufficient
to establish the truth of this ancient adage!

1. These results are very similar to those reported by Thomas (1975)

in a study of 114 male and 223 female students. Using the Wilson-
Patterson scale of conservatism, he found a correlation between C
and sexual experience of -0-41. C correlated -0-58 with attitudes
towards premarital sex; correlations of C with church attendance were
positive (r=049 for males, 0-58 for females); and church attendance
correlated negatively with sexual experience (r=—0-29, -0-48 res¬
pectively). These correlations were obtained in New Zealand.
2. Homogamy for attitudes has been extensively documented in such
papers as those by Kirkpatrick and Stone (1935), Morgan and Remmers
(1935), Newcomb and Svehla (1937), Schiller (1932), Schooley (1936)
and Schuster and Elderton (1906). All found sizeable correlations, com¬
parable with those here reported.
3. See also the work of Jaffe and others (1974) demonstrating a link
between sexual arousal and aggressiveness.

175
 APPENDIX E
KEY FOR CONSERVATISM, CAPITALISM
AND TOUGHMINDEDNESS SCALES

Conservatism Capitalism T oughminded

- 1 53 -13 5
2 54 -15 9
3 55 -16 14
- 6 64 -27 17
7 67 -30 32
8 69 -33 38
10 70 35 -41
11 72 39 -47
-12 74 42 49
18 78 44 50
-19 82 -45 56
20 86 -48 -65
21 -51 66
22 -57 71
23 -61 75
24 -68 76
25 -79 -88
26 -80
28 -81
34 -84
36 -87
37
40
43

176
 5

Sexual attitudes and behaviour in

psychiatric patients

It seemed of interest to apply the attitude and behaviour scales used

in our study of normal men and women to a particularly abnormal
sample, in order to see to what extent the relationships between
factors and personality, and even the very nature of the factors,
could be replicated. The very abnormality of the sample of crimin¬
ally insane males tested at Broadmoor makes it likely that, if our
main findings could be replicated with them, then any lack of ran¬
domness in the selection of the original normal sample could not
have been responsible for producing the results observed. Accord¬
ingly a sample of 186 Broadmoor patients was administered the
personality questionnaires, the sexual attitudes inventory and the
sexual behaviour scale described in chapter 4. The patients
come from a population described in some detail by Tennent and
others (1974); they were all male, with an average age slightly above
that of our normal group, and were all judged intelligent enough
to understand the questions, and not so disturbed as to fail to
respond properly to the situation. Each test was administered per¬
sonally by Miss Gloria Braganga (1972), who gives a detailed
description of the sample and the method of testing. The sample
does not differ significantly from the total group of patients with
respect to offender category: 16 per cent are sex offenders; 11 per
cent have been committed for arson; 26 per cent for murder and
manslaughter; 28 per cent for attempted murder and wounding;
16 per cent for other violence, assault and property offences; and
3 per cent were admitted under Section 26 of the Mental Health
Act. Patients volunteered to undergo the test, but of those ap¬
proached over thirty failed to complete the inventory.
Table 5.1 lists the percentage of ‘Yes’ answers of the patients, and

177
 SEX AND PERSONALITY

also gives for the sake of comparison the percentage of ‘Yes’ answers
of our normal sample, quoted from the preceding chapter. Questions
155 and 159 were omitted by too many patients to make it possible
to include them in any of our analyses.

Table 5.1 Percentage of ‘yes' answers of Broadmoor patients and normal

males to sex inventory items. Asterisked items omitted from factor analysis

Question Broadmoor Normal Question Broadmoor Normal

1 60 49 78 55 74
2 58 43 79 69 82
3 52 15 80 57 56
4 49 64 81 23 9
5 51 8 82 51 19
6 37 5 83 12 8
7 48 50 84 25 13
*8 18 4 85 53 73
9 60 69 86 27 35
10 60 81 87 37 9
11 44 22 88 13 0
*12 20 5 89 35 41
13 37 43 *90 36 7
14 18 10 91 58 45
15 18 4 92 12 18
16 50 32 93 57 78
17 19 12 94 18 17
18 13 6 95 18 31
19 51 35 96 49 55
20 54 54 97 16 7
21 46 24 *98 12 1
22 47 45 99 18 6
23 29 3 100 10 9
24 31 34 101 12 26
25 79 78 102 32 21
26 25 16 103 56 64
*27 34 20 104 60 45
*28 46 56 105 11 3
29 39 17 106 33 44
30 59 75 *107 66 29
31 24 29 108 56 44
32 16 9 109 40 11
*33 22 5 110 32 12
34 70 50 111 50 57
*35 85 96 112 65 69
36 60 57 113 56 65
37 46 87 114 52 39
38 17 7 *115 40 55
39 48 68 116 26 53
40 38 15 117 18 5
41 23 10 118 23 11
42 30 32 119 23 33

178
 PSYCHIATRIC PATIENTS

Question Broadmoor Normal Question Broadmoor Normal

43 72 80 120 32 52
44 21 13 121 18 24
*45 9 1 122 73 66
46 26 25 123 11 8
47 35 20 124 12 6
48 9 7 125 59 69
49 32 24 126 17 6
50 66 76 127 34 21
51 10 3 128 20 9
52 25 8 129 24 19
53 34 31 *130 30 38
*54 34 14 131 63 63
55 82 95 132 68 53
56 18 13 133 18 13
57 70 85 134 63 41
58 25 6 135 64 37
59 22 11 *136 24 16
60 37 24 *137 24 16
61 19 6 *138 74 83
*62 48 39 139 18 3
63 59 95 *140 13 5
64 17 7 141 79 90
65 47 34 *142 68 41
*66 52 58 *143 53 26
67 58 69 144 32 36
68 19 22 145 55 55
69 34 6 146 47 15
*70 81 84 147 49 29
71 62 74 *148 65 59
72 40 42 149 30 22
73 73 87 150 33 20
74 70 67 151 77 85
75 26 54 152 74 86
76 74 84 153 38 48
77 35 67

It will be clear from looking at the figures that the Broadmoor

patients are on the whole much more inhibited sexually than are
the ‘normal’ group. Thus they are less easily excited sexually; condi¬
tions have to be just right; they think only rarely about sex; they
consciously try to keep sex thoughts out of their mind; when they
have strong sex feelings they cannot express them; they don’t think
about sex almost every day; they say that they do not get excited
,very easily; they look upon sex as being only for reproduction and
not for pleasure; sex is not all that important to them; they are not
excited by the thought of an illicit relationship; they can take sex or
leave it alone. They draw sharp lines between what is right and

179
 SEX AND PERSONALITY

what is not in sexual conduct: they find it disgusting to see animals

having sexual intercourse; there are some things they would not do
with anyone; they find the thought of a sex orgy disgusting; they
don’t think that sometimes a woman should be sexually aggressive;
they prefer intercourse under bedcovers and in the dark; they do
not feel like scratching and biting their sex partners; they object to
four-letter swear words in mixed company; they find wife-swapping
distasteful; and they find some forms of love-making disgusting.
They hold rather conservative views on sexual matters: virginity is
a girl’s most valuable possession; seeing a person nude does not in¬
terest them; they would protect their children from contact with sex;
they would not take a chance to watch people making love; they are
against pornographic writing being freely published; they believe in a
sexual censorship; they do not uphold the dual standard of morality;
they think that sexual permissiveness undermines society; they do
not consider sex play among young children harmless; and they think
it right that the man should be dominant.
They consider that their lives have not been successful; they
feel they have been deprived sexually; their love life has been
disappointing; they find it hard to talk to people of the opposite
sex; and they feel more comfortable with people of their own sex.
As we have seen, satisfaction with one’s love life correlates nega¬
tively with personality factors P (psychoticism) and N (neuroticism);
in this psychiatrically abnormal group it is therefore not surprising
to find such dissatisfaction with their sex life.
The personality scale scores of the patients are given in table
5.2, together with the normative scores of a sample of 1,000 normal
males of similar age and social class. The patients are clearly
more introverted, more neurotic and very slightly higher on P.
This slight difference in P may at first seem odd in a sample of
insane persons, but it finds an explanation in the extremely high
L score of this group, which is well over twice as high as that of
the normal standardisation group. It is often found that psychotics
give high L scores, thus showing a tendency to dissimulate which
effectively lowers their P and N scores, sometimes very markedly
(Eysenck and Eysenck 1976). Had answers been honest there is
little doubt that the P and N scores would have been considerably
higher. These high L scores must of course also throw some doubt
on the accuracy and honesty of the answers given to the sex
questionnaires.

180
 PSYCHIATRIC PATIENTS

Table 5.2 Personality scale scores of Broadmoor patients and

standardisation sample

Broadmoor Standardisation
P 2-69 + 2-36 2- 50 + 2-71
E 11-45 + 3-89 12-75 + 4-11
N 9-52 + 5-33 7-33 + 4-37
L 8-03+4-68 3- 64 + 2-34

A factor analysis was undertaken of the sexual attitudes question¬

naire, but as in the case of the normal sample it was impossible to
use all the questions as the computer could not handle more than
135 items. It was decided to select 131 items, mostly but not entirely
identical with those used previously with our normal sample, and to
include in the analysis the P, E, N and L scales. The items omitted
from the sex attitude questionnaire are indicated in table 5.1 by
asterisks, so as to make identification possible.
Factor analysis of the correlations between the 131 items and the
four personality scales gave rise to ten interpretable factors. These
are given below; in each case the factor is named, and the items hav¬
ing highest saturations are given, together with the loadings and the
item number in the actual questionnaire. This is done to enable
readers to check item wording; the wording in the tables below
has been paraphrased from the original to avoid too-lengthy state¬
ments. The actual factors are very similar to those obtained on our
normal group, although the items included in the analysis were not
the same in every instance; this may account for the occasional
differences. Thus in this analysis items referring to homosexuality
were included, giving rise to a factor of homosexuality, while in
the analysis of our normal sample these items were for the most
part excluded, thus producing no such factor. The naming of the
factors is of course somewhat subjective; we have followed the
naming procedure of the corresponding factors in the normal group.
Below are set out details concerning factor 1, permissiveness.
Statements are worded in the same direction as in the questionnaire;
if the loading is preceded by a minus sign, the statement should
of course be reversed. Thus people high on the permissiveness
factor do not think about sex only rarely. It will be seen that high
L scorers are not permissive, but that high E scorers are, as are high
P scorers.

181
 SEX AND PERSONALITY

Factor 1: Permissiveness
Question Loading

6 -0-44 Think about sex rarely

17 -0-51 Disturbing to see necking in public
23 -0-43 Try to keep sex thoughts out of mind
30 -0-37 Doesn’t take much to get sexually excited
37 0-31 Think about sex almost every day
38 -0-46 Should not experiment with sex before
marriage
39 0-36 Get sexually excited very easily
41 -0-34 Better not to have sex until married
51 -0-63 Don’t like to be kissed
55 0-57 Enjoy petting
58 -0-59 Seeing a person nude doesn’t interest me
63 0-37 Sometimes a woman should be sexually
aggressive
82 -0-30 Dual standards of morality, allowing men
greater freedom, natural
88 - 0-61 Sex should be for reproduction, not pleasure
109 -0-53 Sex not all that important
110 -0-38 Most men sex-mad
112 0-42 Enjoy lengthy pre-coital play
116 0-32 Sometimes like scratching and biting partner
during intercourse
141 0-41 Naked body is pleasing
154 0-38 Prefer intercourse frequently
P 0-24
E 0-31
N -0-06
L -0-39

Factor 2 is connected with satisfaction. This factor correlates

positively with E, negatively with P and N. There is no correlation
with L.
Factor 2: Satisfaction (1)
Question Loading

4 0-61 Am satisfied with sex life

11 -0-68 Have been deprived sexually
14 -0-33 Sexually rather unattractive
19 -0-52 Something lacking in sex life
21 -0-33 Love life has been disappointing
75 -0-35 Sometimes have felt hostile to partner
111 -0-34 Good in bed is important in marriage partner
112 0-30 Believe sexual activities are average
126 -0-36 Can’t stand people touching me
P -0-14
E 0-29
N -0-22
L -001

182
 PSYCHIATRIC PATIENTS

Factor 3 deals with neurotic sex. Not surprisingly, N has a high

correlation with this factor (r=0-51). P, too, is positively correlated
with this factor, while E is negatively correlated. The L scale again
fails to show any notable relationship with the factor.

Factor 3 : Neurotic sex

Question Loading

14 0-35 Am sexually unattractive

18 0-47 Sexual feelings sometimes unpleasant
23 0-32 Consciously try to keep sex thoughts out of
mind
24 0-42 Felt guilty about sex experiences
26 0-51 Afraid of what I may do sexually
31 0-35 Parents’ influences have inhibited me sexually
32 0-70 Sex thoughts disturb more than should
44 0-70 Conscience bothers me too much
46 0-43 Sexual feelings overpower me
48 0-74 Sex thoughts drive me almost crazy
56 0-66 Worry a lot about sex
59 0-78 Sometimes thinking about sex makes me
nervous
60 0-48 Perverted thoughts have sometimes bothered me
68 0-41 Sometimes felt like humiliating sex partner
84 0-42 Had some bad sex experiences when young
111 -0-48 Being good in bed is important in marriage
partner
117 0-33 No one has satisfied me sexually
124 0-51 Am afraid of sexual relationship
139 0-42 Cannot discuss sexual matters with wife

P 0-25
E -0-31
N 0-51
L -0-10

Factor 4 deals with impersonal sex. Flere L has a high negative

loading, while P and E have small positive ones.

Factor 4: Impersonal sex

Question Loading

13 0-39 Do not need respect or love to enjoy petting/

intercourse with partner
34 -0-39 Some things I wouldn’t do with anyone
72 0-41 Have been involved with more than one sex
affair at same time
77 0-35 If had chance to see people making love,
would take it
97 0-47 Prefer to have new partner every night
102 0-38 Sex more exciting with strangers

183
 SEX AND PERSONALITY

Question Loading

108 -0-34 Sex partner satisfies all physical needs

119 0-70 Group sex appeals
120 0-49 Thought of illicit relationship excites me
146 -0-34 ' Some forms of love-making disgusting
152 0-31 Offered highly pornographic book, would
accept
153 0-77 Invitation to an orgy accepted

P 0-18
E 018
N 0-07
L -0-31

Factor 5 is clearly concerned with pornography. It is interesting

to note that liking for pornography is positively correlated with
extraversion and negatively with the L score.

Factor 5: Pornography
Question Loading

25 0-39 Wouldn’t bother if person married not virgin

36 0-35 Understand homosexuals
43 0-79 Like to look at sexy pictures
61 -0-32 Embarrassed to talk about sex
64 -0-40 Sex jokes disgust
73 0-50 Homosexuality normal for some people
76 0'76 Like to look at pictures of nudes
78 0-37 Pornographic writing freely allowed to be
published
81 -0-67 Too many immoral plays on T.V.
82 0-38 Dual standard of morality natural and should
be continued
87 -0-52 Sexual permissiveness undermines civilised
society
91 -0-40 Preoccupation with sex created by media
123 -0-62 Disturbs me to look at sexy photographs
149 -0-45 Reading girlie magazines failed adult attitude
to sex
151 0-56 Would see blue films
152 0-46 Would accept pornographic book

P -001
E 0-31
N -Oil
L -0-34

Factor 6 deals with sexual nervousness. This factor correlates

negatively with extraversion and positively with neuroticism.

184
 PSYCHIATRIC PATIENTS

Factor 6: Sexual nervousness

Question Loading
4 -0-44 Am satisfied with sex life
7 0-39 Has been a problem to control sex feelings
16 -0-41 Sex contacts never a problem
19 0-40 Something lacking in sex life
20 -0-39 My sex behaviour has never caused me any
trouble
21 0-46 Love life been disappointing
22 0-43 Never had many dates
29 0-57 Have strong sex feelings but given a chance
can’t express myself
47 0-76 Feel nervous with opposite sex
50 -0-75 Feel at ease with opposite sex
52 0-81 Hard to talk with opposite sex
53 0-41 Didn’t learn facts of life until quite old
61 0-38 Am embarrassed to talk about sex
72 -0-40 Have been involved with more than one sex
affair at a time
113 -0-31 Find it easy to tell partner what I like and
dislike about love-making
118 0-33 Feel sexually less competent than friends
P 010
E -0-28
N 0-31
L -0T4

Factor 7 deals with sexual excitement. This factor has only a

slight loading for extraversion.

Factor 7: Excitement
Question Loading

30 0-38 Doesn’t take much to get me sexually excited

39 0-32 Get excited sexually very easily
65 0-31 Believe in taking pleasures where find them
80 0-34 Decisions about abortion should be concern
of woman only
82 0'37 Dual standards of morality allowing men
greater freedom
91 0-32 Present preoccupation with sex created by film,
T.V., adverts
99 -0-38 Prefer partners several years older
132 -0-43 Tenderness is most important quality
147 0-67 It’s right that man should be dominant in sex
relationship
P -0-07
E 0-15
N 0-05
L 0-06

185
 SEX AND PERSONALITY

Factor 8 is concerned with physical sex. It has positive loadings

for psychoticism and neuroticism.

Factor 8: Physical sex

Question Loading

39 0-40 Get excited sexually very easily

42 0-36 Coloured sex partner particularly exciting
79 0-56 When excited can think of nothing but
satisfaction
86 0-63 Sex is far and away my greatest pleasure
106 0-56 Few things are more important than sex
110 0-57 Most men are sex-mad
127 0-46 Physical sex is most important part of marriage

P 0-25
E 0-02
N 0-20
L 001

Factor 9 deals with sexual aggressiveness. It has positive loadings

on psychoticism and negative loadings on L.

Factor 9: Aggressiveness

Question Loading

100 0-55 Sexual fancies often involve flogging

101 0-61 Make vocal noises during intercourse
113 0-30 Find it easy to tell partner like/dislike love-
making
116 0-52 Feel like biting and scratching partner during
intercourse
121 0-55 Usually feel aggressive with partner
123 0-31 Disturbing to look at sexy photographs
132 -0-38 Tenderness is most important quality
135 -0-35 ‘Wife-swapping’ is extremely distasteful

P 0-29
E 002
N 0-08
L -0-24

Factor 10 is one of homosexuality. Neuroticism loads this factor

positively, L loads it negatively.

186
 PSYCHIATRIC PATIENTS

Factor 10: Homosexuality

Question Loading
36 0-41 Understand homosexuals
60 0-31 Perverted thoughts sometimes bother me
73 0-34 Homosexuality is normal for some people
84 0-44 Had bad sex experiences when young
104 0-64 Some things do only to please partner
145 0-33 Men more selfish in love-making than women

P -0-12
E 006
N 0-20
L -0-21

These factors are not independent of each other, and their corre¬
lations can in turn be factored to produce higher-order factors
(superfactors). Two of these emerge from the analysis, and are given
below in tables 5.3 and 5.4. Table 5.3 shows the loadings on super¬
factor 1 (sexual libido), while table 5.4 shows the loadings on
superfactor 2 (sexual satisfaction). These two factors are very
similar to the two superfactors extracted from the analysis of data
in our previous work (Eysenck 1971a), and also from the analysis
of data collected from normal samples filling in the questionnaire
here used. As in our previous work, the correlation between these
two superfactors is quite insignificant (r=— 0-03); in other words,
a person’s satisfaction with his sexual life is in no way dependent
on the degree or strength of his libido.

Table 5.3 Factor loadings on superfactor 1 (sexual libido)

Question Loading

1 — 0-31 Opposite sex will respect you more if not too

familiar
6 -0-42 Think rarely of sex
9 0-42 Love a person, could do anything
10 0-40 Pleasant feeling from touching sexual parts
17 -0-44 Disturbing to see necking in public
22 -0-30 Never had many dates
23 -0-37 Consciously try to keep sex thoughts out of
mind
30 0-39 Doesn’t take much to get sexually excited
34 -0-35 Some things wouldn’t do with anyone
36 046 Understand homosexuals

187
 SEX AND PERSONALITY

Question Loading

37 0-32 Think about sex every day

38 -0-47 Should not experiment with sex before
marriage
39 0-42 Get sexually excited very easily
40 -0-58 Thought of sex orgy disgusting
41 -0-54 Better not to have sex until married
42 0-43 Thought of a coloured partner particularly
exciting
43 0-59 Like to look at sexy pictures
53 -0-35 Didn’t learn facts of life until quite old
55 0-49 Enjoy petting
57 0-33 Pill should be universally available
58 -0-41 Seeing a person nude doesn’t interest me
61 -0-39 Am embarrassed to talk about sex
63 0-30 Sometimes women should be sexually
aggressive
64 -0-48 Sex jokes disgust me
65 0-45 Believe in taking pleasure where I find it
69 -0-40 Would protect children from contacts with sex
72 0-48 Have been involved with more than one affair
at time
73 0-41 Homosexuality is normal for some people
74 0-47 All right to seduce a person who is old enough
to know what he/she is doing
76 0-58 Like to look at pictures of nudes
77 0-48 If had a chance to see people making love
unobserved would take it
78 0-69 Pornography should be freely published
79 0-40 Prostitution should be legally permitted
80 0-38 Decisions about abortion should be concern
of no one but woman
81 -0-42 Too many immoral plays on T.V.
85 0-47 No censorship on sexual grounds of plays and
films
87 -0-47 Sexual permissiveness threatens to undermine
civilised society
88 -0-30 Sex for reproduction, not personal pleasure
89 0-41 Absolute faithfulness to partner as silly as
celibacy
93 0-36 Sex play among children harmless
95 0-38 Would vote for a law permitting polygamy
96 0-51 Though having regular intercourse,
masturbation good for a change
97 0-36 Prefer new sex partner every night
102 0-39 Sex more exciting with a stranger
103 -0-32 Never discussed sex with parents
109 -0-40 Sex not all that important
112 0-42 Enjoy lengthy pre-coital play
116 0-46 Feel like biting and scratching partner during
intercourse
119 0-53 Group sex appeals to me

188
 PSYCHIATRIC PATIENTS

Question Loading

120 0-42 Thought of illicit relationship excites me

123 -0-39 Disturbing to look at sexy photographs
124 -0-42 Am afraid of sexual relationships
125 0-30 Wish women would be more forthcoming
sexually
126 -0-32 Can’t stand people touching me
134 -0-35 Object to four-letter words in mixed company
135 -0-47 Idea of ‘wife-swapping’ extremely distasteful
139 -0-36 Cannot discuss sexual matters with wife
141 0-37 Naked body a pleasing sight
144 0-37 Would not disturb if partner has sex with
someone else, as long as she returned
146 -0-46 Some forms of love-making disgust
151 0-56 Invitation to blue film accepted
152 0-58 Offer of a pornographic book accepted
153 0-56 Invitation to an orgy accepted
154 0-43 Preference for having intercourse frequently

Table 5.4 Factor loadings on superfactor 2 (sexual satisfaction)

Question Loading

4 0-41 Am satisfied with sex life

7 -0-53 Sometimes a problem to control sex feelings
11 -0-42 Have been deprived sexually
18 -0-40 Sexual feelings sometimes unpleasant
19 -0-44 Something lacking in sex life
20 0-36 My sex behaviour has never caused me trouble
21 -0-49 Love life has been disappointing
22 -0-42 Never had many dates
23 -0-45 Consciously try to keep sex thoughts out of
mind
24 -0-47 Have felt guilty about sex experiences
26 -0-39 Have been afraid for what I might do sexually
29 -0-54 Have strong sex feelings but given a chance
can’t express myself
31 -0-33 Parents’ influence has inhibited sexually
32 -0-56 Sex thoughts have disturbed more than should
37 -0-31 Think about sex almost every day
39 -0-35 Get sexually excited easily
42 -0-35 Thought of coloured partner particularly
exciting
44 -0-44 Conscience bothers me too much
46 -0-60 Sexual feelings overpower me
47 -0-36 Feel nervous with opposite sex
48 -0-51 Sex thoughts drive me crazy
49 -0-45 When excited can think of nothing but
satisfaction

189
 SEX AND PERSONALITY

Question Loading

50 0-38 Feel at ease with the opposite sex

52 -0-36 Hard to talk with opposite sex
56 -0-52 Worry a lot about sex
59 -0-53 Thinking about sex makes me nervous
60 -0-42 Perverted thoughts sometimes bother
61 -0-49 Am embarrassed to talk about sex
65 -0-30 Believe in taking pleasures where I find them
68 -0-48 Sometimes felt like humiliating partner
75 -0-43 Sometimes felt hostile to partner
84 -0-34 Have had some bad sex experiences when
young
86 0-30 Sex is greatest pleasure
92 -0-31 Enjoy partner having intercourse with someone
else
97 -0-30 Prefer to have a new partner every night
100 -0-37 Sexual fancies often involve flogging
102 -0-36 Sex more exciting with a stranger
103 -0-32 Never discussed sex with parents
105 -0-44 Don’t always know when have had an orgasm
110 -0-35 Most men are sex-mad
114 -0-33 Like partner to be more expert and experienced
117 -0-45 No one has been able to satisfy me sexually
118 -0-47 Feel sexually less competent than my friends
121 -0-38 Usually feel aggressive with partner
122 0-31 My sexual activities are average
124 -0-44 Am afraid of sexual relationships
125 -0-36 Often wish women were more forthcoming
126 -0-35 Can’t stand people touching me
128 -0-41 Prefer partner to dictate rules of the game
129 -0-34 Find ‘straight sex’ unsatisfactory
133 -0-31 Female genitals are aesthetically pleasing
139 -0-43 Cannot discuss sexual matters with wife
150 -0-32 Matters of sex, women always come off
second-best

These superfactors are also correlated with personality. Super¬

factor 1 (libido) correlates positively with E (r=0-38) and negatively
with L (r=—0-39). Superfactor 2 (satisfaction) correlates negatively
with P (r=—043) and N (r= — 0-52) and positively with E (r=
0-36). Clearly both the psychiatric factors (P and N) are connected
with pronounced dissatisfaction with a person’s sex life, while
extraversion is connected both with a happy sex life and also
with strong libidinal tendencies. High L scorers, while having (or
pretending to have) a weak libido, show a mild degree of satis¬
faction with their sex life (r=0T2). These figures on the whole
agree well with our previous work, which also tended to show that

190
 PSYCHIATRIC PATIENTS

P and N interfere with a normal enjoyment of sex while E en¬

hances it. The whole set of correlations is given in table 5.5.

Table 5.5 Correlations between libido, satisfaction and the four

personality variables

P E N L

S.F.l. 0-04 0-38 -0-10 -0-39 Libido

S.F.2. -0-43 0-36 -0-52 0T0 Satisfaction

Patients were also administered the sexual behaviour question¬

naire; their mean scores are shown in table 5.6. This table also
gives the correlations with the personality variables. In brackets
are given comparison scores with our normal sample of males.
It will be seen that the prisoners (if they are to be believed) have on
the average been less advanced in their love-making techniques
than the normal males by over two steps in the scale; that is quite
a large difference. There are more things they would have liked
to do, and more things they had done and did not like. These
data are in good agreement with the differences already given for
the attitudes questionnaire; they emerge as more inhibited and less
successful and satisfied. This also agrees with their greater intro¬
version, and their (probably) greater neuroticism and psycho-
ticism (granted that the observed scores should be corrected for
dissimulation).

Table 5.6 Scores on sexual behaviour scales of prisoners and correlation

of scales with personality variables

Prisoners Normal P E N L
Controls

Have done and

enjoyed 6-75 + 3-34 8-92 + 2-16 -0-07 0-39 - 010 -019
Have done and
didn’t like 0-70 + 1-33 0-21+0-60 000 -0-09 - 0-01 -0-01
Would have liked
to have done 2-03 + 2-63 1-37 + 0-71 016 -0-12 0-15 -0-05

191
 6

Genetic factors in sexual behaviour (1)

N. G. Martin and H. J. Eysenck

There seems to be little doubt that genetic factors are likely to be

responsible for many of the individual differences in sexual attitudes
and behaviour that we have documented on previous pages; never¬
theless, there is very little scientific evidence even for genetic
causes of male-female differences other than those directly associ¬
ated with physical characteristics. Even in animals individual
differences have not been widely studied in a genetic context; such
a wide-ranging text as that of Scott and Fuller (1965), for instance,
makes no mention of such factors, and neither does the textbook
on Behaviour Genetics by Fuller and Thompson (1960). The present
chapter may thus be regarded as a preliminary exploration of un¬
charted waters, unlikely to provide definitive results but perhaps
capable of suggesting likely conclusions. Our main intention here
is to explore the possibilities of fitting a relatively simple genetic
model to our data; the results may be useful as guidelines to future,
more detailed, explorations.
Before turning to our data, it may be necessary to stress cer¬
tain points that are not always understood by psychologists and
psychiatrists who approach the genetic field on the assumption that
the major, or even the only, interest of geneticists is the discovery of
a single heritability index which could be used to divide the total
variance into an ‘inherited’ portion and another portion due to ‘en¬
vironment’. Such an index might be of interest to stockbreeders,
but would have little scientific importance. The well-known debates
concerning the inheritance of intelligence have stressed the concept
of heritability to the exclusion of other, equally relevant, considera-

192
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

tions. It is important to realise the complexity of the task, and the

number of variables that enters into any reasonable model of
genetic and environmental action and interaction.
If all the genetic and environmental influences acting on a trait
are purely additive and independent, then the total variance of the
trait in the population can be partitioned into the following four
components:

Ex within-families environmental component

G1 within-families genetic component
E2 between-families environmental component
G2 between-families genetic component.

Notice that we distinguish between Ev which includes variance due

to environmental influences specific to an individual and errors
of measurement, and E2, which includes cultural and social in¬
fluences common to members of the same family but differing
between families. The components of Gx and G3 will differ according
to the way in which the genes act. For this reason it is usually more
instructive to talk directly in terms of the sources of genetic variation.
If the genes affecting a trait act in a purely additive way then the only
source of variation is DE, the additive genetic component. How¬
ever, if some genes are dominant to others then this will create a
dominance genetic component HE, although in practice this is notori¬
ously difficult to detect.
A further complication arises if there is a correlation between
spouses for a trait that has some degree of genetic determination.
This phenomenon of like marrying like is known as assortative mat¬
ing and tends to inflate the additive genetic variance between
families. For this reason in twin studies it will be automatically
confounded with an estimate of E2 and when we detect such a com¬
ponent, as we do for radicalism below, we cannot tell to what extent
it is due to cultural differences between families or assortative mat¬
ing. If we have independent evidence of a correlation between hus¬
band and wife, as we do for radicalism, then this would suggest that
at least some part of our apparent E2 is actually extra additive
genetic variance due to assortative mating.
The extent to which Elt Et, Dn, Hn and assortative mating contri¬
bute to the mean squares from an analysis of variance of M.Z. and
D.Z. twin scores is shown in the basic model below.

193
 SEX AND PERSONALITY

Basic model for mean squares of twins reared together

E* E** Dn f HJ

A §
M.Z. between 1 2 1+ —7 i
1 -A
within 1 - -

D.Z. between 1 2 A+ fa
4+ 1 -A
within 1 1 3
16
4

* E,=within-family environmental variance

** E3=between-families environmental variance
t Dr=additive genetic variance
tH:R= dominance variance
§ /t = Fisher’s assortative mating parameter - correlation between the
additive deviations of spouses

We have described the situation in which the different environ¬

mental and genetic influences on a trait act additively and independ¬
ently. However, there are two types of effect that can complicate
the simple breakdown of the total phenotypic variance that we have
described above, viz. genotype-environment interaction (Gx£) and
genotype environmental covariation (COVGE).
If a given environmental stimulus has a bigger effect on the
behaviour of one genotype than on another then this is known
as genotype-environment interaction. For example, glutamic acid
is believed to raise the I.Q. of dull and mentally defective child¬
ren but not of average or bright ones (Eysenck 1973b). Our twin
study provides a simple test of one such type of interaction, des¬
cribed below. However, interaction can be generated as an artefact
of the scale of measurement, and in this case it can often be removed
by rescaling the data before carrying out a normal additive analysis.
This has been done successfully with several factors in this study.
Genotype-environment covariation occurs when genetic and en¬
vironmental deviations are correlated. Eaves (1976) has distin¬
guished three types of COV'
GE

1 A genotype seeks out an environment that is most compatible

with the advantages and limitations bestowed by the genes. Thus a
person inclined to read will seek out libraries and a child predisposed
towards music but not to sport will sit and listen to a record rather
than go out and play games. It is difficult to think of an experiment

194
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

with humans in which this type of COVGE could be distinguished

from the genetic effects themselves.
2 Intelligent parents often have bright children who inherit not
only their parents’ genes but also their environment, well stocked
with good books and journals and an atmosphere conducive to learn¬
ing. These cultural effects (which themselves have a genetic com¬
ponent) will covary with the direct influences of the child’s genes.
However this type of COVGE produced in the presence of cultural
transmission can be detected only by comparing relatives reared
together and apart and in the simple twin study we have here
will be confounded with E2 and assortative mating.
3 Brothers and sisters feared together will have different in¬
fluences on each other according to their genotypes. These in¬
fluences will covary with the direct influences of the genes on the
child himself, and the greater the degree of genetic similarity
between two siblings, the greater will be the covariance. Thus the
covariance between M.Z. twins will be greater than that between
D.Z. twins and this will affect the pattern of mean squares in a way
that allows us to detect this source of COVGE as we shall see below
in the case of sexual satisfaction in females.

The methods by means of which we can obtain data to fit such a

complex model are discussed in detail by Mather and Jinks (1971);
worked examples to illustrate the application of such models to
human personality data are given by Jinks and Fulker (1970) and
by Eaves and Eysenck (1975). The results of applying genetic
methods to human personality are summarised in some detail by
Eysenck (1975a). Quite briefly, it appears that the major personality
factors P, E and N show little evidence of non-additive genetic
factors, which in this context means primarily that dominance
and assortative mating are both absent or present only to a quite
minor extent. A simple model involving additive genetic variance
and within-families environmental variance seems sufficient to fit
the data. There is no evidence for any between-families environ¬
mental variance, which suggests strongly that psychoanalytic and
other theories regarding the determination of personality differences
by maternal or paternal behaviour are in error. For all three person¬
ality traits the value of h2, when correction is made for the un¬
reliability of the scales, lies between 50 and 80 per cent; no great
precision attends these values, of course, but they suggest the region
within which the true values are likely to lie. It should also be

195
 SEX AND PERSONALITY

noted, in view of some of the results reported below, that social

attitudes, i.e. radicalism-conservatism, toughmindedness-tender-
mindedness and emphasis all showed intermediate heritability,
on the basis of a similar simple model; in addition it was found
that toughmindedness was genetically associated with P and E\
emphasis was similarly associated with P (Eaves and Eysenck 1974).
Only one study has been found in the literature that has attempted
to investigate the question of genetic determination of sexual be¬
haviour, and this study was self-confessedly a pilot one only, using
a very small number of subjects (ten pairs of M.Z. and six pairs of
D.Z. same-sexed twins). Equal numbers of twins in each group were
male and female, with mean ages of twenty-six and twenty-seven
years for M.Z. and D.Z. pairs respectively. Zygosity was established
by means of a blood sample test using eight independent blood
groups; data were ascertained by means of interviews, personality
inventory and body measurement. (The personality inventory used
was the E.P.I.) Information was sought on maximum autosexual
outlets (maximum number of times masturbation to orgasm had
occurred in one day); age at menarche; age at first ejaculation; age
when regular masturbation began; age at first orgasm; age at first
passive genital stimulation; age at first active genital stimulation;
age at first intercourse. Heritabilities were estimated by means
of Holzinger’s formula:

H=(Vdz-Vmz)/Vdz
and significance by means of the formula F=FDZ/FMZ (Chilton
1972). This formula is of doubtful value for the determination of heri¬
tability, but may be used as indicating the significance of the greater
similarity of M.Z. as compared with D.Z. twins, for the particular
trait investigated. The methodology of this study is clearly faulty
from the point of view of modern biometrical genetics, and may with
advantage be contrasted with the more adequate methods illustrated
in our own research. Jinks and Fulker (1970) have demonstrated
the inadequacy of Holzinger’s formula and the estimate of F used.
The findings may gain interest by comparing the heritabilities
found against certain landmarks. Thus the heritability of height was
0-90; that for weight 0-97; that for body build (the ponderal index)
0-88. Table 6.1 shows the major findings for the various sex be¬
haviour variables. In view of the small number of twins, it is sur¬
prising to find that any of the values are statistically significant; it

196
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

Table 6.1 Holzinger heritabilities for sexual behaviours: asterisks indicate

statistical significance of p ^ 005 level

Male Female Total

Maximum autosexual outlet 0-89* 0-65 0-80*

Age of menarche - 0-61 -
Age at first
ejaculation 0-69 - -
masturbation 0-80 076 0-78*
orgasm 0-70 085 081*
passive genital stimulation 0-87 046 0-80*
active genital stimulation 0-90* 000 0-69
sexual intercourse 0-81 000 061

is less surprising that practically all the values are well above
zero. The two exceptions relate to female age at first active genital
stimulation and at first intercourse; it is possible that these depend
to a large extent on the more active male, whose behaviour is
probably more self-determined in sexual matters in our society.
The author summarises these results as follows.

The sexual drive is not a uniform one - just as the need and appetite
for food varies from one individual to another so does sexual drive
and capacity ... The results from this pilot study suggest that
genetic factors play a part in determining this variation, that the
age at which an individual begins developing and exploring his
sexuality is not just a function of environmental experiences but is
also, in part, an expression of his genotype.

It is noteworthy that age at menarche, with an H of 0-61, was

insignificant in relation to genetic determination; others had pre¬
viously found this determination fully significant (e.g. Tisserand-
Perrier 1953). This suggests that the other almost significant H
values are also likely to lack significance simply because of the
very small number of cases involved. For the same reason, it seems
likely that the personality scale scores also failed of significance.
It is interesting, though, that there was a clear association between
introversion and neuroticism, on the one hand, and ectomorphic
body build on the other;(2) this is very much in line with expectation
(Eysenck 1970a, Rees 1973). A high ectomorphic rating was associ¬
ated with later sexual development; and, in particular, ectomorphs
tended to experience first coitus later than either mesomorphs or
endomorphs. This finding is supported by McNeill and Livson

197
 SEX AND PERSONALITY

(1963), who showed that ectomorphs reached menarche later than

individuals with a lower ponderal index. These results agree well
with our own, as far as personality and its relationship with sexual
behaviour are concerned.
The present study was carried out on 153 male twins and 339
female twins from the Institute of Psychiatry Twin Register.(3)
Many more pairs of twins were approached, but the sexual nature
of the questionnaire put off many twins who had taken part in
previous investigations. Roughly one in three of those approached
filled in the inventories anonymously. However, the mean person¬
ality scale scores of those who did fill in the sex inventories were
not very different from those of the total twin sample; it appears
that of our total twin sample, the group that constitutes our present
sub-sample was not entirely unrepresentative. The questionnaires
administered to this sub-sample were the sexual attitudes question¬
naire discussed in chapter 3, the social attitudes questionnaire
discussed in chapter 4 and the P, E, N and L scales of the E.P.Q.
(Eysenck and Eysenck 1975). Only the major sex attitudes scales
(satisfaction, libido, masculinity-femininity) were analysed from
the sex inventory; R, T and Emphasis were analysed from the social
attitudes inventory. Thus we used ten scales in all, namely the
above six scales and the four personality scales. Table 6.2 gives
the means and standard deviations of the twin groups on these
scales.

Table 6.2 Means and standard deviations for twin sample

Female (N = 339) Male (AT =153)

Satisfaction 12-40± 3-22 12-45+ 3-43
Libido 12-56+ 5-95 19-99+ 6-27
Masculinity 19-15 ± 5-02 28-23+ 6-16
Radicalism 145-61+21-01 147-08 + 24-60
T oughmindedness 45-94+ 7-85 49-67+ 8-20
Emphasis 95-50+18-16 100-53 + 18-90
Psychoticism 1-05 ± 1-32 1-77+ 1-62
Extraversion 11-59 ± 4-07 12-61+ 4-06
Neuroticism 11-47+ 4-61 8-99+ 5-00
Lie 8-11 ± 3-89 6-74+ 3-81

Table 6.3 shows the intercorrelations, for males and females

separately, of the ten scales in question. It will be seen that the
general features of our previous work with different samples are
reproduced with a considerable degree of accuracy.

198
 -K * *K
* ■K -K * -K ■K
On O «o VO VO CO
i CO (N CO O O
>•4
© © 6 © 6 o 6 6 o
1 1 1 1 1 1

•K
•K ★
-K * * * *
(N 1—( r- >o

-0 09
r- VO
»0 i 1—< o T-* 04
© 6 6 6 © 6 O O
L 1

* *
* * M
co Tf

-0-10
Tj- VO o r*» r-
W co <N p o
© 6 © © © 6 6 6
1 I i
personality scales: males above leading diagonal, females below. Asterisks
Twin correlations between sex attitudes, social attitudes and

★ * •K
■K * * * * + *
oo 00 C"- 00

-0-21
a. r- VO O 'Jr
r4 »—1 <N xj- i—i
6 © 6 6 6 6 6 6
-i i

C/3
indicate significance levels {0 05, 0 01 and 0-001)

’£/)
C3 -K •K ■K -K
00 ON i—i CS

0 01
X3 Os o O
a o o 04 o
E 6 6 6 6 © 6 6 6

C/i *K
tJi * * *
* *

-0-20*
<D * * -K
lM <N ON <N ON »o *o
a 04 CO
DO *o CO CO O 04 o o
*■1
—> T3
o
o T3 6 6 © © 6 O 6 6
H C i i. l

6 e
c/i ■*. ■K -0-25*
* -K
*o o 04 on r- o
o <N cp 04 <N s 8
*6 6 © © © 6 6 O 6
ca i 1 i
P4
*
* *
*
-0 14*

-K -K * -K
Ph 04 r- CO CO <N r^- o
1 cs op 04 Tf O 8 7"1
s
© © © 6 o 6 O O
1 1

o *
■K
*
- 0-23*

* * * -K
-o *
O VO ’“H r- CO
'£ 00 CO VO s r4 o t-h
6 © © 6 o 6 6 6
Table 6.3

c
o *K * ■¥■ ■K
0-10

CO Tf lO O
o © o o (N CO
6 6 © 6 6 6 6 O
09 ii i I l |
’+2 i
a
C/0

g
a ’S
o C/3
Cfl
‘£3 ’§ <D
O <u a
a o <+H T3
C/3 JO 1 <D
*-♦3 TD
JD 6
c3 ^5 C/3 .5 C/3
C/3
‘3 'c/3
•“"H "3 H o3
3
O o xs
DO
3 cfl P3 a
X ca *3
03 O 3
£ a>
10 L

<Z3 CO 2 H w Oh
CN co VO VO r- 00 ON
 SEX AND PERSONALITY

1 Sexual satisfaction and libido are uncorrelated.

2 Masculinity-femininity correlates with libido, but not with
satisfaction.
3 Libido and masculinity correlate with toughmindedness, but
satisfaction does not.
4 Sexual satisfaction correlates negatively with the pathological
scales, P and N.
5 Sexual satisfaction correlates positively with E.
6 Libido correlates positively with P, E and N, the correlation
with E being the highest for the males and the lowest for the
females.
7 L correlates positively with satisfaction and negatively with
libido and masculinity.
8 P correlates with toughmindedness and emphasis.
9 P,E,N and L show only slight correlations, and are essentially
independent.
10 The relationships for males are very similar to those for
females, although the different number of subjects in the male and
female samples makes the pattern of significant correlations differ
somewhat.

We may conclude that our sample of twins is essentially similar

in all important relationships to the various samples previously
studied, so that results on heritability obtained from them may
with some confidence be transferred to non-twin samples. (There are
differences in means between samples, but this is probably less
important in the case of genetic analysis where we are mainly
concerned with variances and covariances.)
Analyses of genetic models for our ten scales were made both
with the raw data, and also with age-corrected, rescaled data. Raw
scales in psychological measurement often produce artefacts in
genetic analyses that may take the form of apparent genotype-
environment interaction; such artefacts may be detected by means
of the sum-difference regression test for M.Z. twins (Jinks and
Fulker 1970). It is entirely appropriate to rescale data to minimise
this interaction between pair sums which may represent the geno¬
typic value of a pair (Gj) and pair differences which are due to
environmental influences specific to the individual (Ef). Bartlett
(1947) has given a list of useful transformations to apply when
various relationships of mean and variance are found. In many

200
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

instances it is obvious that these relationships are generated by

the distributions of the traits. Thus if individuals are better dis¬
criminated at one end of the scale than at the other this will pro¬
duce a highly skewed distribution of scores which results in a
monotonic regression curve with positive slope if the skewness is
positive, and negative slope if the skewness is negative. If skew¬
ness is positive it can often be removed by taking square roots or
logarithms; female radicalism scores show such a distribution.
Where skewness is negative, some variant of the antilog trans¬
formation may be used. Other transformations that have been used
on these data are angles and normal weights; it would be in¬
appropriate here to go into details, particularly as the uncorrected
results are strikingly similar to those achieved laboriously with age
and scale correction; while recognising the rather low discrimin¬
ating power we have with this sample, this similarity speaks well for
the robust nature of the data and findings.
In generating and fitting genetic models, we usually wish to
provide the most parsimonious description of the data, and add
parameters only if simpler models fail. We routinely fit the follow¬
ing five models to a set of data:

1 E1 This model tests whether there is residual variance

after removing error and individual environmental
(within-family) variance
2 ExE2 Simple environmental model
3 E Dr Simple genetical model (DE denotes additive genetic
variance)
4 E&Dj,,
5 (Hv refers to dominance variance).
Eaves and Eysenck (1975) have discussed in detail the assump¬
tions and mechanics of fitting genetic models to win twin data along
these lines. We usually resort to interpretation in terms of three
parameters only when the simpler models fail. However, if we have
external evidence for the importance of a third parameter, if we
particularly want to estimate a parameter regardless of its signi¬
ficance or if estimation of a third parameter causes a significant
reduction in the residual chi square, then fitting three-parameter
models may be worthwhile. In particular, the E1EiDR model can
sometimes help elucidate a situation in which both two-parameter
models fit.

201
 SEX AND PERSONALITY

These five models have been fitted to the following combinations

of data from the five twin groups (same sex males and females, M.Z.
and D.Z. and opposite sex twins). Note, for example, that for
females we have between and within-mean squares for both M.Z.
and D.Z. twins, giving four statistics in all:

1 Females 4 statistics
2 Males 4 statistics
3 Same-sex pairs 8 statistics
4 Same-sex and opposite-sex pairs 10 statistics

Before fitting a model to all the same-sex pairs jointly, we must

decide whether the same set of parameters is appropriate for both
males and females. If we sum the residual chi squares obtained
after fitting a model to male and female data separately and sub¬
tract this from the residual chi square after fitting the model to all
eight statistics together we obtain a chi square which tests the hetero¬
geneity of the fit of the model over sexes. If the heterogeneity chi
square is suspiciously large, then we must fit different parameters,
perhaps even different models, to the sexes separately. If there
is some sort of sex limitation or sex linkage, the heterogeneity chi
square will not necessarily be significant, but the opposite-sex mean
square will be different from the D.Z. same-sex mean squares, and
obviously different parameters should be fitted to the two sexes. If
neither of these complications is present then we may safely fit the
same model to all ten statistics.
Estimates of parameters and heritability, where appropriate, are
given for the most illuminating cases in table 6.4. The residual chi
squares and the significance of the parameters (one-tail normal
deviate) and chi square are also indicated. For the £jDE model the
heritability is calculated as:

h2 =

and for the model as:

h2= -
Wr+E1+E2
and their standard deviations are calculated as the variance of a
ratio as shown by Eaves (1970).

202
 On m oo O vrj

0-33 + 0 19
00 cn

0-72 + 005
O o T-4 Tf p
6 6 6 6 6 6 6
+l +l +*1 +1 +1 +l +1
r- o CO Tf o
fN p vp vn Tf p
6 6 6 6 6 6 6

^ M ^ ^ MMrt CN M ^ <N fN fN fN ^ (S N rn

2
2

1
Table 6.4 Estimates of parameters from selected cases of model-fitting to
different sets of normally scaled, age-corrected data. Heritability estimates

* + +

4-74+
r- o© oo fN 'O o xnn *-< fN o

0-62
4-03
o 00 »0 Tt 00
v2

p p ^ vo r- poo vo Tf m
fN 6 6 6 4 6 466
given where appropriate (+=0-05(p 010)

+
m
r-
H

-
-
-
l 1 1 1 1 1 1 1 1- 1 1 1 1 1 1 1 1
c©
fN
00
*k

299-07***
* *
* * -k * * -K

141-21*
* * * *k * * ■K
r- fN fN OO r- oo VO ^H fN OO VO Tf
p fp fN P fN On r- 00 ^ op m fN »p
D

-
r-H 6 4 1 4 1 fN 6 1 fN 6 I wh 6 I 6 4
fN VO VO VO ^ OO t"- co oo m vo
rn fN ^ fN fN fN *-h fN

136 00***
*k *
-k * * *
* + ■k

8002*
* -k * * -k *
On On ?H CN O VO vo On fN On fN
P OO fN oo
9 p 9 p
E„

-
*0 1
4 m 6 fN —H fN vo 4 no */T
cn o vo fN On fN fN 00 O vo
»-h

■k ■k ■k ■k ■k■k
75-76***

•k
59-23***
62-26***

-k * -k * * -k -k -k -k -k -k -k
* ■k * * -k -k * ■k * •k -k •k -k •k ■k •k -k •k-k
■k •k ■k * -k -k -k •k ■k ■k -k •k -k -k -k ■k ■k -k-k
00 On m m o fN r- On 00 m m CO OnCO
cn m
E,

fN c© °p 9 p p op P fN p 9 rn fp vo
6 4 6 do Os fN WO rn 4 fN fN 6 6 6 fN N- 4
00 VO N" m fN fN i-H 1-H fN On On fN fN N* fN fN
1—< 1—1 1“1 1"H i-H 1-H i—1 i"H 1-H i—i i—< TH l-H i—i i—<
Data Model

rs m >n fN c*i tJ- n n Tf M m t M n Tt N m rf

3

4
F

Ph

C ;>>
Radicalism

o 4-*

o ’3
(J
a O
o3
Factor

03( -o
•O £
C/3 4

203
 h- fH Tf VO Tf 00 Tf r- OO O
O co o r4 o t—< o O co
© o © 6 6 6 6 6 6 6
+1 +1 +1 +1 +l +i +1 +1 +1 +l
wo r- Tf Tf CO VO ON Tf ON ^
r- co Tf vo wo co r4 CO CO CO
6 6 6 6 6 © 6 © 6 6

CN C4 r-- r- vo n n
"S* (N (N I VO CN CN t—i N H (N CS

-k -k + Omt'
o Tf co vo c4 Tf r* wo C4 r~ vo <s
Tf m
■—i r—
<0 Tt ^
H <N ON T—H cs »n Tf <N rj- vp Tf t-' NmN
cb 6 6 cb vb T-H *b vb o CTs <0 Tf ~ 6 ©

1 1 1 1. I- 1 1 I 1 1 1 1 1 1 1 L 1- 1 1 1 1
<ES

* * “k -k •k
* * *k -k -K *
* * -k -k -k
Table 6.4—cont.

*k * * -k * *
Tf WO VO Tf wo 00 co p O ON OO VO r- r4
r- oo vo oo
ec <N on wo vo cp O p *0
I Tf
<N
r- 1 T-H -b* I bn Tf
<Q cb cb ON I © ^ 1 cb cb 1 Tf
00 Tf 00 r- <N co co WO 00 c4 Tf WO (N Tf t-H
<N *-h *—< VCS
T <N T-H t-h CN hh r4 T-H T-H
1 1

* * * -k
* * * *k •k -k ■k
■k * •k * ■k -k ■k ■k -k
o -h ro T-H *"H T-H T-H (N Tf CO tj- T-H
j « wo ■ cn *0 1
T-H (N cp
1 9 OO 1 ^
O i ^ o I 9
l 9
<t^ VO 3- O On
T-H
cb vb T-H »b CO
•n
cb t2-< C4
T-H
CO t vo CN CO wo vo vo 00 CO vo
T-H T-H <N

* -k ■k ■k -k * -k •k -k -k -k -k -k ■k -k * ■k ■k -k -k -k *
* -k •k -k *k -k ■k -k -k -k -k -k -k -k -k * * * -k -k -k -k
-k -k ■k •k -k -k * •k ■k -k ■k -k -k -k -k * -k -k -k -k -k *
oo vo 00 CO vo O r4 T-HON ro T-H T-H On VO n CO r4 vo 00
(N ro op p r^ T-H op r- O o (N CN op op cp 9 9 p O
9
ON r- oo do vb vb rb Tj- WO Tf *b wb o VO Tf cb do tb cb cb cb cb
»o ^t wo ON r- i> 00 00 ON wo ro ro wo WO wo vo Tj- Tf <S T-H T-H
T-H T-H T-H T-H T-H T-H T—H T-H H
1

<L)
*u
o Tf <N ro Tj* (N ro Tf
2

(N ro cs ro Tf
3

r4 ro
3

Tf
s
ctf
"c3 <u Uh S 2 u.
C/3
Q
C/3
C/3
0>
a
X) C/3

• noa> ‘tocd
*H
o *^H C
fcO.E -d
o
ft
ctf § S 6
Um H W Bn

204
 1-4 SO m

0-50 + 0-12
<n ON © © OO 00 l-H O oo
O <N © cp © <N i—< ri- ©
© © 6 6 © © © © © © © © ©
+1 +1 +! +1 -H +1 +! +1 +1 +1 +1 +1 +1
r- <N C\ oo oo •A) 44 <N Tf
•an cn 4- <N •an cn up ©
Tf-
cp
© © 6 6 © © © © © © © © © -

2
2
1
1
r- r- vo N N « N N ^ cn I-* eN (N i-H M N ^ h

OO CN r- Tf fN

2-92
2-92
00 m »an 0\ f~ Tj-

3-08
cn <N 44 •an ON co cn

2-81
JN
up p © On T* ^ © 0O Os cp NO 5 © (S O' © rp ON On On
1-4 i—■ Tf cb ri n fS ^4 © © m tn N 1^4 <N © © ©
© ©

cn

180-76
NO

-
-
1 1 1 1 1- 1 1 1 1 1 1- 1 ! 1 1- 1 1 1 1 1 *r>
cn
1-4

195-40**
* * * * *fc
* t + * * ■K *
Tf rf »an Tt cs

239 74
ON <s r-* r-

104-18
© NO ’'t © ©
rp Up rp on m »an cp © SO NO © p* rp rp ©

-
Tf ON 1 c4 1 ^ Os 1 «an c-i i © r* 1 © NO I 6 NO IN
fS T—1 SO 1-4 SO Tfr r- oo r- nan Tt* r- r^
fN 44 fS ^ 1-4 rS ^ 1-4 44

* * *
* * *
92-16**
* * * *
* * * * * ■¥■ *
m
- 22-59
ON NO NO •an NO NO On cn r- 00 •an
Cp
8 3 rp cs cp NO
-

-
NO
Os
,
On NO
NO
1 ^
CN
(N
r-
NO
1 *0
oo r^
<N
1 ^
NO
Tj-
©
00
<S •an ©
l T1
NO
ON
cb
•AN
1. ^
NO
1-4

■K ■¥■ ■K * * ■K ■K
99-21***
98-66***
98-66***

* * * *
108-85***

* * * * * * * •K * •K
* * * * * * •K * * * ** *K * * ■K * *K * *K * ■K
* * * * * * *K * * * ■K
* •K * * *K * •K * *K * *
''t 00 00 Tf <N 00 <N•an © 44
Tf © cn NO r- r^ NO NO
p- 1-4 ON fp On cn (N ip ©Ip © On 'H* wp •P cn 44 oo wp up
© •AN ON TJ- On <N NO 1-4 Th o <N On 44 •AN 14 •AN oo •AN Tf
<N © © <N © NO •AN NO o © Tf ’'t © Os 8 m cn cn
44 1-4 1-4 i-4 »-4 i-4 1-4 1-4 1-4 1-4 1-4 1-4 1-4 1-4 44 1™^ i-' i-4 i^ 1-4 44

Nn't M ml- N n -t N n't m cn n- N C<N Tj- CN cn T}- »AN

O
*—i 2 s Uh

205
 SEX AND PERSONALITY

Table 6.4 contains in essence the sum of our results. It may

repay detailed study, and it may also be worthwhile to discuss
some of the major results as they pertain to the various scales
analysed. Note first of all that there is only one case of heterogeneity
of fit over sexes (toughmindedness); this will be discussed later. Note
next that the E1 model either fails completely or is a very poor fit in
all cases. This indicates that there is significant variation to explain
in addition to that caused by errors of measurement and individual
(within-family) environmental influences. Our trouble comes in try¬
ing to decide whether this residual variation can best be accounted
for by between-families environmental variance (E2), additive genetic
variation (D^ or both.
In most cases, both two-parameter models yield non-significant
residual chi squares. Formally, we must accept that either E2 or DR
is responsible for the residual variation over and above Ev but in
practice we can compare the two chi squares and also fit the EXEJ)B
model and see which of the extra parameters is significant or takes
a nonsense value. Although we have found little significant hetero¬
geneity for any of the traits, it is still instructive to compare the
model fitting to male and female data. We shall examine the results
of the model fitting for each factor.

1 Sexual satisfaction

Either two-parameter model fits the male data, but the simple
genetic model fits slightly but not significantly better and gives
a heritability for males of 0 40. For females, the environmental
model fails and the genetic model almost does so. The three-
parameter models fit, but one gives a large negative En and the other
a significant negative Dn. Inspection of the mean squares reveals
a pattern strongly suggestive of competition between siblings for
limited resources. If a particular genotype requires a particular
resource in short supply, then competition for that resource will
be more intense between M.Z. twins, having the same genotype,
than between D.Z. twins, having only half their genes in common.
Eaves (1976) has provided a mathematical model for sibling
effects in man and shown how competition or cooperation between
individuals of varying degrees of genetic similarity can produce
differences in total variance of the groups. If £)E represents the

206
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

additive genetic variance of a phenotype produced by a set of genes,

Dh" the variance produced by the effect of those genes on a sibling
and De' the covariance of the additive effect on the individual and
the environmental effect on his sibling, then we may write our model
for the mean square of twins reared together as follows.

Ex Dn Db'

M.Z.T. between 1 1 1 2
within 1 — — —

D.Z.T. between 3 3
1 4 * lj

within 1
1 1
4

We see that Da and Du" have the same coefficients and have
to be estimated as one parameter. If there is cooperation between
twins, then Dn' will be positive, and we expect the total variance of
the M.Z. twins to be greater than that of D.Z. twins: cr2 >cr2 .
If there is competition between twins, then Dv' will be negative, and
we expect cr2TM7<cr2XDz, and furthermore it is possible that the
between mean square will actually be less than the within mean
square, especially for D.Z. twins. This is what Eaves found for
psychoticism in males, and in fact is the pattern we observe in the
mean squares for sexual satisfaction in females. Fitting the competi¬
tion model as shown in table 6.5 gives a remarkably good fit to the
data (perhaps reflecting our inspection of the data before fitting
the model!). Whether we regard the competition parameter as signi¬
ficant depends on whether we do a one-tail or a two-tail test. In
any case, it is clear that there is a large competition effect. Its con¬
sequence is that competition is more intense, and so additive
genetic variation is less important in the expression of sexual
satisfaction in M.Z. twins than in D.Z., and this is reflected in the
lower heritability for M.Z. females (0-31) than for D.Z. females
60-45). We can interpret this as meaning that M.Z. females attract
the same types of male, and have to compete for his attention to a
greater extent than D.Z. females, who will tend to draw com¬
panions from populations that do not overlap to the same extent.
Males, of course, tend to go out and meet companions outside the
home in a way not so characteristic of females; hence the same
factors leading to competition are not present to the same extent.

207
 SEX AND PERSONALITY

Table 6.5 Fitting the competition model to the data for sexual satisfaction
in females

E D D ' df Observed Expected

mean mean
squares squares

M.Z. b l 1 2 93 269 270

w l — 95 141 141
D.Z. b l 3
4 H 52 241 238
w l 1 -T 54 275 275
T

E1 141-1 ± 20-4 c=6-92***

A
331-8 ± 118-9 c=2-79**
A
101-2 ± 55-5 c = 1-82
Dn

r2
X i
_
— 0-005
A A.

i(DR+ 2D *')
h2 = = 0-3145 + 00882
k V 2£e')+e;
^ Kdu+ zy)
h2 = --- = 0-4497+00983
DZ HD^+DJ+E,

2 Libido and masculinity

In the highly correlated traits of libido and masculinity, the simple

environmental model seems more appropriate to the female data
than the simple genetic model, while the reverse seems true for
the males. It would tally with popular folklore if sexual attitudes
on these two scales were determined largely by cultural pressures
in females, while men were less inhibited by such pressures (i.e., if
there were less in the way of pressures on men!), and variation
were an expression of their genotypes.
It may also be noted that there is still significant between-
families environmental variance in females (Ea), and this suggests
that daughters tend to conform to a set of family standards and
attitudes more so than sons. If there were cultural or biological
pressures on females to adopt a certain set of sexual attitudes, then
we should expect to find between-families variance reduced, relative
to that in males. This would be akin to the sex-limited genetic

208
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

variation found for a trait like baldness, where variation of the

major gene effects is largely restricted to males. We should thus ex¬
pect to find total variance smaller in females than in males,
although Ej should be of the same magnitude in both sexes. The
first of these anticipations is fulfilled but the second is not, so that
the evidence for coercion to a standard set of attitudes in females
is not strong. Nevertheless, it is an hypothesis worthy of more
specific formulation and testing in future studies. Of course, if such
sex-limited genetic variation for libido were found, it could just
as readily be genetic factors that are suppressing the expression of
genetic variation in women, as occurs in baldness.

3 Radicalism

For both sexes the simple genetic model gives a better fit than
the simple environmental model (which almost fails in females),
yielding high heritability in both sexes. However, it is interesting
that this is one of the few traits in which the E,E2Dn model yields
positive estimates of both E2 and DR that are consistent over sexes
indicating the importance of cultural variation or assortative mating
(or both) in addition to additive genetic variation and Ev The
heritabilities, taking account of E2, are similar for the two sexes
and indeed the parameter estimates appear consistent over all ten
statistics. This is striking confirmation of a similar need for an E2
or an assortative mating parameter found by Hewitt (1974) in
analysing another body of social attitude data in twins, and by
Eaves and Eysenck on yet another set of data using an adaptation
of the Wilson-Paterson Conservatism Scale (unpublished). (We
have already noted that assortative mating is indeed strongly pre¬
sent in our data; cf. chapter 3.)
If cultural effects are important in influencing radicalism through¬
out life, rather than at a particular formative period, then we should
expect these to act increasingly on individuals rather than pairs
as twins get older and are separated for longer. This should be
reflected in larger within-pair differences in older twins and we
should detect these unequivocally as E1 effects in M.Z. twins. We
do indeed find a significant positive regression of pair differences
on age in M.Z. females. Although this regression accounts for only
about 2 per cent of the radicalism within-pairs variance, or about

209
 SEX AND PERSONALITY

1 per cent of the total variance, we may have here further evidence
for the action of cultural effects in radicalism scores.

4 Toughmindedness

Here again our results echo Hewitt’s (1974) in finding different

parameters required for males and females. He found that the
simple environmental model failed for both sexes, but that a simple
genetic model, while fitting both sexes, yielded a considerably
higher heritability for females than males. Perhaps because of
poor sampling, our D.Z. within mean square in males is actually
smaller than the corresponding M.Z. mean square, and this causes
the simple genetic model to fail in males, while the simple environ¬
mental model fails in females. We may have here evidence of sex
limitation of genetic expression in which sons are more likely
to be conditioned into more or less toughminded attitudes that
are held by the family than are daughters, of whom no such ex¬
pectations are held. In contemporary jargon, the large E2 compon¬
ent for females in libido and masculinity, and for males in tough¬
mindedness, could be evidence of conditioning into ‘sexual roles’.

5 Emphasis

Both two-parameter models fit for males, but the simple environ¬
mental model almost fails for females, and the simple genetic
hypothesis appears to give an adequate fit for all ten statistics,
additive genetic variation accounting for about a third of the total,
a finding in almost perfect agreement with Hewitt (1974).

6 Personality

We do not propose to say much here about personality variables

as these have been studied on much larger samples by Eaves and
Eysenck, and will be discussed separately in a forthcoming publica¬
tion. It is noteworthy, however, that the four personality variables
are remarkable in their consistency. No two- or three-parameter
model fails with either sex. However, where we estimate that
heritability is in the range 0-34—0-39 for females, the male values

210
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

are all in the range 0-50-0-58. This finding is the more striking
since, given equal heritabilities in the male and female popula¬
tions, we should expect the greater restriction in sampling of male
twins to produce lower estimates of heritability. Although the dif¬
ference in heritability is not significant for any trait taken singly,
the directional evidence is very strong here. Nichols (1969) too has
found higher heritabilities for males than females in almost every
personality trait that he measured.
Our heritability estimates from the simple genetic model, fitted
to all ten statistics, are 0-43, 0-40, 0-40 and 0-42 for P, E, N and
L\ they are of the same order as the estimates for P, E and N
obtained by Eaves and Eysenck (1974) for much larger samples
of same-sex twins (0-35, 0-48 and 0-49). These estimates are of
course underestimates of heritability because we have made no
attempt to correct for inconsistency of item responses; this is
traditionally included with the environmental variance, but strictly
speaking should not be so included. The same is true, of course,
for all the other estimates of heritability here made; these are
lower thresholds rather than firm determinations, even within their
fiducial limits.
As an example of the more detailed sort of calculation that
might be carried out in this connection, we have selected the libido
variable, as being perhaps the most interesting of our factors, for
further analysis. By putting both pairs by items interaction and
main effects mean squares as a single data set, and estimating
genetic and environmental parameters of each, we can estimate the
main effects parameters corrected for internal unreliability. This
has the effect of altering the heritability upwards or downwards by
an amount depending upon the relative size of the genetic and
environmental components of the interaction mean square (Eaves
and Eysenck 1976, Hewitt 1974). Table 6.6 shows simple genetic
models fitted simultaneously to the main effects and interaction
mean squares for libido in males. The compound model fits well,
and all the parameter estimates are highly significant. The estimate
of the heritability of the items interaction is 0 18+0 02, identical
with a value obtained from a separate analysis of the interaction
mean squares. The interaction mean square thus contributes some
information to both the genetic and environmental main effects
parameters, and making allowance for these contributions we find
that the heritability of the libido factor is raised from 0-61+0-10

211
 ■g *
£O pcr< vo vo as oo CO oo (N vo —x
(N CO as CO vo Os co as CO vo
vo vo r-H O o vo CO vo
x a 00 Tf ▼—' Vo oo 1 cb vb cb cb
o 6 © © 6 © <N
W g II II II II
6 o o o o

(N SO 00 co
C/5 (N o
<3 O o o
Fitting a compound model to the main effects and items

© o o ©
s 5 oo n as y—i <N oo © © 6 ©
£ a* (N © as CO CO
0) CO
CO vo <N as vo vo CO +1 +1 +1 +1
CO oo 1-H vo as (N co
X> 6 6 6 6 o © as co
o vo
o 8 co
O
© ©
interaction mean squares for libido in males

W .r « Ph
<Q <Q
Ol O vo (N vo vo as
vo O o O i—1 o
X5 r- CO vo Tf vo Tj- CO oo
op vo op Tf vo 00 CM
*55 vo CO CO CO CO CO CO
Os 1—1 as T“H r** r-
£ <N Os
C4 o vo
fN tj-

(N
o
©
oo as o vo VO r- o vo
CO CO CO vo *-■ VO as +1
CO CO vo vo co 00 t^
co VO

vo | i I r- as i |.
<o cs

<N
vo vo | | VO VO I I
CO CO CO CO

<D ►-
Table 6.6

I I
I Q
bJ

N N si si
S S a ci

212
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

when our correction is made for items inconsistency to a value

of 067 + 011 when we make this correction. This would be our
best estimate of heritability obtainable from the data in our pos¬
session.
It may be useful to recapitulate our main findings, at least in
as much as our sex factors are concerned. The discussion so far has
been tangled up with technical issues which may have distracted
attention from the major conclusions that are suggested by the
data and the analysis. The term ‘suggested’ is used on purpose
because the number of twin pairs, although large in comparison
with the sort of sample usually employed in studies of this kind,
was too small to allow any definitive conclusions, even if prob¬
lems of selection and other difficulties discussed already could be
neglected. We note then that, with respect to libido, additive genetic
factors play a very strong part for men, but with women cultural in¬
fluences seem to be far more important. When corrections are made
for unreliability in the male sample, the heritability of libido
reaches the figure of 67 per cent. Masculinity behaves rather like
libido - not surprisingly in view of the high correlation between
the two scores.
Sexual satisfaction presents a rather more complex picture, with
heritability somewhat lower, and competition likely for M.Z.
females. These findings are not perhaps surprising. A person’s
libido is to some extent independent of others; it can be expressed
in different ways depending on circumstances, and genetic factors
therefore have a more or less free field. Sexual satisfaction, on the
other hand, depends much more on other people - lovers, spouses
and the like. Consequently there is more room here'for environ¬
mental factors to exert an influence, although these in turn are of
course in part a reflection of genotypic aspects, such as good
looks, intelligence and so forth. Competition, on a priori grounds,
is not likely to affect libido, but may quite easily arise in connection
with satisfaction; there are several treatments of this theme in the
world’s literature!
We believe that this discussion of our results, brief though
it has inevitably been, has succeeded in demonstrating the far-
reaching nature of modern biometrical genetic investigations. The
stress that has been put on the statistical assessment of ‘heritability’
has distorted the aims of such research, which are concerned with
the construction of a proper model of the influences, genetic and

213
 SEX AND PERSONALITY

environmental, that go to determine the phenotype. Critics who

complained that heritability could not be determined in any mean¬
ingful fashion from available data were both right and wrong; they
were right when dealing with the usual type of investigation using
the simple Holzinger formula, but wrong when dealing with the
much more sophisticated model-fitting methods of Mather, Jinks
and Eaves. Again, little interest attaches to studies using small
numbers of twin pairs with traits of intermediate heritability such
as those in the attitudes and personality domain; unfortunately data
from such small samples are compatible with almost any model,
however unlikely, and cannot be used to differentiate between
models in any meaningful manner. Even the number of subjects
used in our study, much larger than is usual in this field, is still
too small to have complete confidence in the results. We believe
that replication is not likely to reverse our major conclusions, but
it may well fail to confirm some of our minor ones.
The evidence seems clear that the main factors of sexual attitudes
are partly determined by genetic causes, and it seems likely that
these effects are mediated through personality traits such as P, E
and N, at least in part. However, clearly behaviour is the end pro¬
duct of a chain of chemical reactions which start in the DNA but
are mediated through biochemical, endocrine and neurological funct¬
ions before finally being experienced as behaviour. The question
of the biological basis of personality can be answered only by
psychophysiological experimentation directly relating personality
measures and such factors as cortical arousal (for E) and autono¬
mic activation (for N), perhaps somewhat along the lines suggested
by Eysenck (1967). There is some evidence to suggest that these
physiological factors are relevant to extraversion and neuroticism;
P is too new and unexplored a variable to have any such stable
reference. The only suggestion that has come forward for a biolo¬
gical basis for this trait is grounded on the observation that P seems
to be closely connected with masculinity, a relationship that appears
particularly close in the sexual sphere (Eysenck and Eysenck 1976).
This links up with the suggestion by Money (1961a, 1961b)
that ‘in adulthood, androgen is the libido hormone for both men and
women. The androgens of eroticism in normal, untreated women
may be of adrenal origin or may derive metabolically from the
closely related progestins.’ However, we shall see that there are
some arguments against such an identification (Davidson 1972).

214
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

It is known that progestin, usually given as a ‘fertility pill’, has

masculinising effects morphologically (Wilkins 1959, 1960; Young,
Gay and Phoenix 1965); it also appears to have such masculinising
effects as far as behaviour is concerned (Ehrhardt and Money 1967).
These authors studied ten girls whose mothers received progestin
while pregnant; nine were born with abnormal genitalia. ‘From
the sex-role preference tests and the interview material, nine of the
ten girls were tomboys. The criteria were play with boys’ toys;
athletic energy, outdoor pursuits; and minimal concern for feminine
frills, doll play, baby care and household chores’ (p. 99). These
results are in line with similar findings in animal work (Money
1965, Phoenix 1966). These studies suggest strongly a correlation
between morphological and behavioural effects of sex hormones.
These studies, while relevant and supportive, nevertheless are
only of indirect concern; what is needed is clearly a more direct
link between hormones and P+ behaviour, both in the sexual
and non-sexual spheres. Unfortunately, there is a dearth of such
studies; a review by Eysenck and Eysenck (1976) found the evidence
suggestive but certainly not conclusive. The same must be said
about the recent work of Daitzman (1976), which should certainly
be repeated on larger numbers of subjects, but which constitutes
the most direct and relevant study to date of this important problem.
Daitzman (1976) carried out a pilot and a main study, using
25 and 51 male subjects respectively, all of them university students;
he also used a very small sample of 7 female students in addition.
Measuring level of androgen and estrogen secretion on two occasions
(in order to rule out temporary effects and establish the reliability
of the hormonal measure), he correlated the results with a variety
of questionnaire data. Among his main findings were the following.

1 Androgen secretion is positively correlated with neuroticism,

correlations in the two samples being 0-53 and 0-27.
2 Androgen secretion is positively but insignificantly correlated
with psychoticism, correlations being 0T0 and 0 07.
3 Correlations with various measures of extraversion tend to be
positive (Sensation Seeking disinhibition scale, P.E.N. extraversion,
C.P.I. sociability, M.M.P.I. social introversion (negative)).
These results, given the small numbers involved and the pioneer¬
ing nature of the experiment, tend to fit in quite well with our
conceptualisation; high libido is correlated with P, E and N, and all

215
 SEX AND PERSONALITY

three show (although not always at a significant level) positive cor¬

relations with androgen level.
4 However, the heterosexual experience scale used by Daitzman
(which resembles our sexual behaviour scale) correlates negatively
with androgen level (—0-45, —0-49). This is certainly contrary to
expectation, and would require replication before being accepted.
(Daitzman gives hundreds of correlations in his thesis, and by chance
the conventional level of significance would be reached or exceeded
by quite a number of these.)
5 Another finding is more in line with expectation, namely that
favourable parental attitude to sex in children is positively correlated
with androgen secretion (0-31, 0-36).
6 When we turn to estrogen levels, we find that these too show
a powerful correlation with neuroticism (0-45).
7 The correlation with psychoticism is positive but quite insig¬
nificant, but that with the M.M.P.I. schizophrenia scale is highly
significant (0-48).
8 With femininity, using the C.P.I. scale, the correlation is sur¬
prisingly negative (—0-33).
9 Estrogen correlates positively with the general Sensation Seek¬
ing Scale (0-29) and significantly with the S.S. disinhibition scale
(0-37), which is probably the scale most closely related to extra¬
version in the S.S.S.
10 The correlation of estrogen secretion with heterosexual ex¬
perience is negative, although quite insignificantly so. With parental
attitude towards sex in children it is strongly positive (0-59).
These results suggest, if anything, that estrogen secretion has on
the whole similar relations with personality and sexual attitudes
and behaviours as does androgen secretion; both seem to relate
positively to P, E and N (although these relations are not always
clear or significant). Both have negative correlations with sexual
experience and positive ones with parental attitudes.
A last set of correlations is offered by Daitzman, using the
androgen/ estrogen balance.
11 This balance correlates positively with femininity, which is
perhaps unexpected.
12 It also correlates positively with extra version, and negatively
with the M.M.P.I. social introversion scale.
13 Correlations with sexual experiences are negative, very signi¬
ficantly so. All the correlations mentioned so far are not changed

216
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

significantly by controlling for such variables as weight, age, height,

or recency of orgasm.

The sample of girls was, of course, too small to give very mean¬
ingful results, but androgen levels were associated significantly
with neuroticism, and negatively with masculinity; estrogen levels
with the M.M.P.I. schizophrenia scale and the psychopathic deviant
scale, the S.S. disinhibition scale, and (negatively) with the C.P.I.
socialisation and communality scales, and the ego-strength scale.
Thus, in women too there appears to exist a similar relationship
between high P, high E and high N on the one hand, and high
androgen and estrogen levels on the other. It would almost appear
as if high male and female hormone levels both act in a synergistic
fashion, rather than in an antagonistic one, at least as far as per¬
sonality and sexual behaviour are concerned. The picture would
be completely consistent if only the sexual experience scale in this
study correlated positively rather than negatively with sex hormone
secretions. This failure of prediction constitutes a definite and
serious theoretical anomaly which must be left to future research to
clear up.
One further set of studies must be mentioned because it suggests
a tangible way of measuring an anatomical feature probably deter¬
mined by antenatal androgen secretion, and in turn correlated with
high libido v. low libido behaviour. Schlegel (1969, 1975) has
carried out large-scale measurements of the size of the pelvic outlet,
using the distance between the ischial tuberosities (Sitzbeinhocker)
as his guide. It is well known that the pelvis of the male tends
towards a funnel shape, that of the female towards a tube shape,
but that intermediate types are frequent in both sexes. Typically,
males have a small pelvic outlet, females a large pelvic outlet;
this is of course in line with the biological purpose of this particular
anatomical feature. Now it appears that, while the size and shape
of most parts of the pelvis are properly determined only at puberty,
the distance between the ischial tuberosities is already seen to be
differentiated between males and females prior to birth (Boucher
1957), suggesting an effect of gonadal excretions at this decisive
period.
Schlegel argues that it is possible, using this anthropometric
measure, to grade both males and females along a continuum from
andromorphic to gynekomorphic; this continuum, he argues (on

217
 SEX AND PERSONALITY

the basis of some not very convincing familial correlations), is

based on a single-gene type of inheritance, located on the x-chromo-
some, and plentiful mutation. Schlegel’s work is of interest because
he cites quantitative evidence from very extensive studies of high
correlations between size of pelvic outlet and sexual and social
behaviour which fits our high P, high libido picture quite well. Thus
he finds a correlation of — 0-90(!) between the gynekomorph out¬
let type and preference for active, dominant types of sexual con¬
duct, irrespective of sex. He also quotes a correlation of —0-77
between gynekomorph outlet type and preference for younger, par¬
ticularly adolescent, sex partners. On the social side, he gives a cor¬
relation of —0-43 with social leadership and dominance. It is
interesting that these are typically P rather than E behaviours; with
E-type social behaviours the correlations are in the opposite direc¬
tion: 0-38 with sociability, 035 with communication, and 0-31 with
ability to make social contacts.
The use of the ischial tuberosities’ distance as a measure is
given support by the fact that the distribution in the population
is bimodal, in both men and women. This is very unusual pheno¬
menon, and suggests that some strong biological, sex-linked factor
may indeed have been at work here to produce these effects. Another
pointer in this direction is the fact that Schlegel found strong cor¬
relations between gynekomorph pelvic outlet types and male homo¬
sexuality.
Schlegel goes on to argue, on the basis of much animal work, that
in the higher mammals there are two quite distinct components
of male sexual behaviour: sexual satisfaction, and sexual domin¬
ance, linked with the pecking order. It is the latter that is related
to the anatomical and morphological differences in pelvic shape
observed, and hence it is not surprising that P-type behaviour is
associated with small pelvic outlets. Schlegel has also observed other
anatomical correlates of sexual behaviour, e.g. a highly significant
correlation of 0-31 between frequency of sexual intercourse and
circumference of the hand (this would appear to be a measure of
Kretschmer’s and Sheldon’s athletic-mesomorphic type). These cor¬
relations are of less interest because there is no obvious relationship
with hormonal secretions of the androgen type, although quite
possibly such correlations will be discovered in the future. The area
of research opened up by Schlegel promises to be of considerable
interest, and it is to be regretted that his work is not better known

218
 GENETIC FACTORS IN SEXUAL BEHAVIOUR

m the English-speaking countries. At present, however, it would

be premature to put too much faith in his results, well though they
agree with our own work and theorising, without the necessary
replication.
We must conclude this brief discussion of the hormonal effects
on sexual and non-sexual behaviour with a question mark. The
evidence is suggestive of a causal relation between androgens and
P + , high libido behaviour,(4) but it seems certain that any theory
simply equating androgens with masculinity and estrogens with
femininity will run into serious difficulties. It is possible that we
may have to discriminate more carefully between the different andro¬
gens, attributing different effects to different hormones; it may be
that there are synergistic as well as antagonistic interaction effects
between androgens, estrogens, progesterons, etc. It would be idle
to guess at the complexities that future research may encounter. We
seem to have isolated the complex of variables that may be relevant
to P+ and high libido behaviour, but we have only taken the
most preliminary steps to study this relationship in detail.

1. This chapter was jointly authored by N. G. Martin (Department of

Genetics, University of Birmingham) and H. J. Eysenck; a more detailed
discussion of the results and issues arising will be found in Martin
(1976). We wish to thank Mrs J. Kasriel for her invaluable assistance in
locating and contacting the twins on whom the research was done; she
coordinated the collection of data and managed to keep straight the
questionnaire returns that were collected on several different occasions.
Without her help this chapter would never have seen the light of day!
2. Ectomorph body-build refers to long, lean physique; mesomorph
to squat, powerful physique, and endomorph to large, fat physique.
3. Of the twins, 95 pairs were M.Z. females, 39 were M.Z. males,
54 D.Z. females, 17 D.Z. males, and 41 D.Z. twins pairs of opposite
sex. Mean ages ranged from twenty-three (for the unlike sex group) to
twenty-eight. Around 60 per cent were middle class.
4. There are two sets of findings that are especially relevant. Ehren-
kranz and others (1974), Persky and others (1971), Kreuz and Rose
(1972) and others have demonstrated a correlation of plasma testos¬
terone levels with aggressive behaviour and social dominance in man;
these are P+ activities. Shepherd, Lader and Rodnight (1968) have
summarised literature to show that phenothiazines, which have curative
powers for schizophrenia, include estrogenic activities; these reported
endocrine effects include delayed ovulation and menstruation, oligo-
menorrhoea and amenorrhoea and breast swelling with lactation, as
well as impotence in men and increased libido in women. Drugs which
counteract high P are therefore found to have estrogenic properties.

219
 7

Sexual attitudes and

social consequences

We have so far been concerned with the trees, rather than the wood;
in this final chapter we shall attempt to take stock of what we have
found and discuss, at least briefly, the kinds of consequences that
might follow from our findings. In summarising our main findings
we shall commit two solecisms. In the first place, we shall assume
that these findings are genuine and replicable. Now, clearly, our
data were collected at a particular place and time, from small
though varied samples, in a manner that does not rule out falsifica¬
tion on the part of the respondents; there are many good reasons
why errors might have slipped into the experiment, and why the
results might not apply in another place or at another time. We
acknowledge the difficulties but will not reiterate them each time a
finding is mentioned; for the purpose of the discussion we shall
assume that the major outlines of our findings are not likely to be
disproved in subsequent researches. We are encouraged to believe
this because we have found similar results with quite divergent
groups (students, normals, adults, prisoners, Broadmoor patients),
because the results from many different instruments converge on
certain broad generalisations, and because the results largely
agree with predictions arrived at from general and experimental
psychology, thus forming an attractive example of intergration of
theory and fact. Many details of the picture are almost certain to be
wrong still, and to require change; much further work will certainly
be needed to put this preliminary sketch on an acceptable footing;
nevertheless, the broad outlines, we believe, are likely to survive.
This belief may not carry conviction for the reader, and we gladly

220
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

acknowledge that we may be quite wrong; yet for the purpose of

this chapter we shall proceed as if the belief were justified.
In the second place, we shall paint this broad picture without
paying too much attention to minor anomalies; we shall not enter
a detailed discussion of each experiment, carefully setting off the
results of one against the other. This we have already done in the
major part of this work; it would not be appropriate here. For a
more detailed (and consequently more accurate) account the reader
must consult the original data; he is of course free to disagree with
our summary conclusion. What such a picture loses in detail it gains
in intelligibility; scientific work often resembles in its technique the
pointillisme of the French painters - the resulting picture must be
seen at a distance in order to achieve its effect. This distancing is
what we are attempting here.
The first finding is that attitudes towards sex fall into groups that
have clear-cut psychological meaning. Items denoting sexual satis¬
faction cluster together; so do items relating to permissiveness,
excitement, nervousness, curiosity, repression, experimentation,
censorship, promiscuity, hostility, guilt, inhibition, impersonal sex,
pornography, dominance, disgust and so forth. Most of these factors
can be measured reliably, and would seem to constitute a meaning¬
ful way of describing sexual attitudes.
The second finding is perhaps of greater psychological interest.
Apparently these factors or clusters of items themselves are cor¬
related, rather than independent; they define two very broad, clear-
cut, independent factors which we have called Satisfaction v.
Pathology, and Libido v. Restraint. These factors emerge equally
clearly in men and women, and in students, random samples of
adults, Broadmoor patients and other groups; it would appear that
they have general significance. Males, as expected, have more libido,
but women may, if anything, be more satisfied with their own sex
fives; however, the difference for satisfaction is marginal, while that
for libido is very marked.
While sexual performance declines with age in both men and
women, libido actually increases in men, while it decreases in
women; thus libido in women parallels the decline in sexual per¬
formance, but goes counter to it in men. The reasons for this curious
phenomenon are unknown, and in view of the odd nature of this
finding it might be advisable to wait for a replication before taking
it too seriously. There is little change in satisfaction with age, except

221
 SEX AND PERSONALITY

at the very young end of the age scale; at sixteen men and women
are less satisfied than at later ages. This is not surprising; society
does not make it easy for adolescents to find suitable outlets for
their sex drive.
Male-female differences on items and scales are very marked,
with men more frequently endorsing items dealing with porno¬
graphy, orgies, voyeurism, prostitution, impersonal sex, promiscuity,
pre- and extra-marital sex, sexual excitement and other indications
of libido. Women are more prudish and show more guilt. A special
masculinity-femininity scale made up of items giving the largest
differences between sexes behaved very much like the libido scale,
and indeed featured much the same items.
Personality differences profoundly influenced the views held on
the many issues canvassed in the sex inventory. High P scorers
endorsed items that featured on the masculinity and libido scales;
in particular, they showed lack of concern with virginity, liked
impersonal sex, pre- and extra-marital sex and prostitution, and
experienced strong sexual excitement. They disliked marriage,
censorship and restraint. High P scorers tended to have low scores
on satisfaction; they tended to be dissatisfied with their sex lives.
There is a definite tendency for high P scorers to indulge in what
are commonly called perverted practices, including group sex.
Quite generally, high P scorers take a very toughminded, biological
attitude to sex, and are opposed to current morality and customs
and to idealistic and romantic notions.
Extraverts are similar to high P scorers in their high libido level,
but unlike them in their much greater satisfaction. They tended to
have intercourse earlier than introverts or ambiverts, have it more
frequently, with more partners, in more different positions; they
tended to indulge in more varied sexual behaviour outside inter¬
course, in more prolonged precoital love play, and to habituate more
quickly to sexual stimuli, with consequent search for change. A
happy philanderer, the extravert has considerable social facility
with the opposite sex, likes and enjoys his sexual activity, is con¬
tented with it and has no worries or anxieties regarding it. His
promiscuity lacks the pathological touches of the high P scorer, or
the insistence on perverted forms of satisfaction. The extravert’s
attitudes seem to derive from hedonism, those of the high P scorer
from protest and hostility.
High N scorers have greater libido than low N scorers, but have

222
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

many hangups which interfere with a peaceful sex life; hence their
satisfaction is particularly low. They have strong guilt feelings,
worry about sexual activities, have fears and difficulties associated
with contact with the opposite sex and often see sex as both trouble¬
some and disgusting. They tend to blame their parents for their
inhibitions, and there is an aura of abnormality about their
expressed desires. In line with this is the finding that high N
scorers show more clearly pathological behaviour, e.g. lack of
orgasm and frigidity in women, ejaculatio praecox and impotence
in men.
Subjects scoring high on the L scale tend to be conformist, as
expected from the nature of the scale. They endorse orthodox,
socially approved views and attitudes, and deny any thoughts of
perverted practices and other socially undesirable habits. They have
first intercourse relatively late in life, and they do not approve of
pre- or extra-marital sex, the strength of inhibiting influences is
high, they take an idealistic rather than a purely biological view of
sex, and they approve of censorship, monogamy and faithfulness
in marriage and oppose permissiveness. High L scorers are low on
the libido factors, but are not dissatisfied with their sex lives. As in
the case of P, E and N, here too the pattern of sexual attitudes that
applies to men also applies to women; one of the strongest supports
for the reliability of our findings comes from the marked agreement
of the results obtained for men and women of the various
personality types.
Social attitudes in general correlate meaningfully with sex atti¬
tudes. Conservatism correlates negatively with permissiveness and
stress on physical sex, pornography and libido generally. Tough¬
mindedness correlates with permissiveness, impersonal sex, sexual
excitement and quite generally the libido factor and all the items
contained in this. It is not surprising that conservatives (with a
small c!) endorse orthodox social viewpoints and condemn modern
trends, and the established relationship between toughmindedness
and P explains the observed relations for T.
Assortative mating is indicated for the sexual attitude factors,
and also for the social attitudes factors; in both cases, quite high
correlations are found for males and females either married or
living together in semi-permanent unions. The correlations are
almost as high as those usually found for intelligence, i.e. around
0-5 to 0-6; correlations for personality variables are much lower, and

223
 SEX AND PERSONALITY

do not deviate significantly from zero in most cases.

Sexual behaviour, as distinguished from sexual attitudes, falls
into much the same pattern; several findings, such as the greater
indulgence of high P scorers in sexual perversions, have already
been mentioned, as has the greater and earlier participation in
sexual practices of extraverts, as compared with introverts. Of
particular interest to us has been the discovery that, although people
may indulge in certain practices, they may actually dislike this
participation. Only just over 50 per cent of females who had
indulged in fellatio had actually enjoyed doing so; even for such
items as manual genital manipulation, male by female, or for
mutual genital manipulation, less than 70 per cent of females
enjoyed the experience. Males reported enjoyment in almost 100
per cent of all cases. Altogether, we found that females had done
things they disliked almost four times as frequently as males. It is
not unexpected that dislike of activities indulged in is correlated
with sexual shyness, prudishness and disgust; correlations with
scales entering into the libido factor are negative - in other words,
people high on libido tend to like what they do by way of sexual
experimentation. There is little by way of relationship between
personality and dislike of sexual activities indulged in. This may be
due to the fact that there were too few males in this category to run
meaningful correlations, and that women are much more prone
to follow the lead of the male in such things, irrespective of their
personality configuration almost. This whole area is worthy of much
more detailed study than we were able to give it.
Among the abnormal group we tested, Broadmoor patients came
out as low on libido and also dissatisfied with their sex lives. The
former finding might seem surprising, but it confirms frequent
psychiatric observations to this effect. The unhappiness and dis¬
satisfaction of psychiatric patients with their sex lives is not surpris¬
ing; these results very much bear out expectations based on less
systematic research along psychiatric fines done through interviews
and other somewhat subjective channels.
Age was found to be an important factor in the sexual attitudes
adopted by our subjects. As might have been expected, older sub¬
jects tended to be more conservative and less permissive; there was
also a general decline of libido for women. This would seem to
parallel the decline in the number of times sexual intercourse
occurs. Satisfaction however does not seem to decline. It is difficult

224
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

to say whether these effects are due to actual age-related changes,

or whether they merely mirror the social climate in which individuals
of a different age grew up.
Social class is another important variable that has perhaps not
been studied as much as it deserves; our own findings are incidental,
and probably not as trustworthy as those relating to personality. On
the whole, the working-class subjects emerge as more ‘earthy’ and
the middle-class groups as more ‘moral’ in their attitudes; this is not
unexpected in view of the greater toughmindedness usually found
in working-class groups. Thus working-class people seem to have
higher libido, but not greater satisfaction; the former conclusion is
perhaps in accord with widespread social beliefs for which hitherto
there has not been much empirical support.
The attitudes and behaviours discussed above, as well as the
personality traits associated with them, have a clear genetic basis;
about half of the variation in attitudes and behaviour that we can
observe in our society is due to heredity, the other half to environ¬
ment. This is a very rough-and-ready summary of our findings; for
libido, for example, the males showed a heritability of 67 per cent,
which is well above the 50 per cent level. However, such estimates
have high fiducial limits, and it would not be realistic to attribute
too much accuracy to the actual values found; they can best be used
as estimates of the likely region within which the true values may
lie. In addition, the genetic models fitted to our data suggest certain
complex interactions with within- and between-families environ¬
mental influences, as well as competition (as in the case of sexual
satisfaction) for limited resources. Our genetic findings are perhaps
the most challenging, as well as the most in need of replication, of
all our discoveries.
Before turning to a discussion of the social relevance and mean¬
ing of these facts, and the consequences that might be thought to
flow from them, it may be useful to take a position on a controversy
that has for long bedevilled any attempt at reasonable interpretation
of psychological data, in this as in any other field. Psychology
often seems to be divided into biotropes and sociotropes; into people
who stress biological determinants to the exclusion of social ones,
and others who reverse this process. Sociologists and social
psychologists, as well as psychoanalysts, usually belong with the
sociotropes, as do anthropologists. Physiological and experimental
psychologists often take the opposite side, as do zoologists and

225
 SEX AND PERSONALITY

ethologists. Both sides may pay lip-service to the other, but they
nevertheless tend to stress one or the other alternative. The position
here taken is that so well put by E. O. Wilson in his monumental
book on Sociobiology (1975); we believe that man is a biosocial
organism whose behaviour is determined both by genetic and
environmental forces, acting in combination, and that any attempt
to understand his behaviour exclusively or even largely in terms of
the one determinant or the other is ultimately doomed to futility.
At the moment the danger of one-sided interpretation comes much
more from the side of the sociologically-minded environmentalists,
who often refuse to recognise any biological factors whatever; it
would be much more difficult to think of biologically sophisticated
writers who would deny the importance of environmental and social
factors in human behaviour.
Difficult as it may seem to deny the importance and relevance of
hormonal and general physiological differences between the sexes,
many modern writers seem to stress overwhelmingly the importance
of ‘sex roles’ and their determination by society - without asking
themselves whether perhaps society imposes these roles because
nature has so ordained. It cannot be the purpose of this chapter to
enter into this controversy, which has generated a huge literature.
The reader is referred to Hutt’s (1972) book on Males and Females
for a general introduction and short bibliography; to Ford and
Beach’s (1952) book on Patterns of Sexual Behaviour for a survey
of work directly concerned with sex differences in this field; and
Levine’s (1972) Hormones and Behavior for an account of what is
known about differential hormonal secretions in the two sexes.
Readers interested in the way biological differences between the
sexes may determine social behaviour may with advantage study
Goldberg’s (1973) The Inevitability of Patriarchy, subtitled ‘Why
the biological difference between men and women always produces
male domination’. The thesis advanced by the author may not be
popular, but the empirical support that it receives would make it
very difficult to argue with.
The main difference between men and women, in the sexual field,
appears to lie in the area covered by our factor of libido, and the
differences documented in this book are amusingly illustrated by a
recent survey conducted in the United States in which 2,000 men
and women, constituting a random sample, were asked to rank the
pleasurability of twenty-two everyday activities. Among the men

226
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

aged between 26 and 39, sex was a clear leader, and among men
aged between 40 and 55 it came first or second. But women between
the ages of 26 and 39 rated music, nature, family and travel more
fun than sex, and among women over 39 even such activities as
snoozing, watching television and doing housework came out ahead.
Figure 6.1 illustrates this difference very clearly. In this figure we
have plotted first of all the decline in actual physical performance
for both men and women as they get older; the differences are small
and quite insignificant, although pervasive, i.e. apparent of all ages.
In so far as they represent a factual difference, they may be due to
the existence of prostitutes, none of whom were included among
the women questioned, but who might have been used by some
of the men, thus swelling the male figures. (The figures are derived
from a random national sample questioned in the U.K.) Also given
in the figure are data derived from our twin sample; here we have
used their factor scores for the libido factor, and drawn regression
lines through the resulting scores; only the lines are given in the
figure. The ordinate is subdivided only according to the performance
data, i.e. in terms of number of times per week that intercourse was
had on the average. We have not given the factor scores for the
libido data as these have no meaning by themselves; all the data
are meant to illustrate is that for women libido declines pari passu
with performance, while for men there is no decline in libido, and
there may be an actual increase. (Actually the decline for women

Figure 6.1 Sexual performance of males and females at various ages;

also degree of libido

227
 SEX AND PERSONALITY

does not set in until after the age of thirty-five; before that there
appears to be an actual increase.)
These figures should not be over-interpreted. The number of
males in the sample was much smaller than that of females, so that
the actual increase in libido must be regarded with suspicion, par¬
ticularly as in our larger adult sample (see chapter 3) we did not
observe such an increase. However, there is a statistical difference
between the sexes, not only with respect to libido but also with
respect to slope; in other words, women decline in libido with age,
while men either actually increase (the ‘dirty old man’ syndrome),
or at least do not decline at anything like the same rate.
Again, there are dangers in interpreting cross-sectional data; as
already pointed out in relation to our social attitude data, differences
between age groups may not be the effect of ageing as such, but may
rather reflect social norms at the time different individuals grew
up. That this may have been responsible in part for our findings
here is indicated by a series of studies showing that while

... today’s adolescents and young adults have a sex-specific view

of sexuality which is clearly related to the traditional gender roles:
girls behave sexually as if they had less sexual drive than boys; girls
show fewer signs of sexual frustration when they abstain sexually;
girls behave as if they should show less sexual initiative than their
male partners or at least not more than he; girls behave as if their
sexuality is more dependent on love, personal relations, and fidelity
than do boys. [Schmidt 1975]

Yet during recent years, and especially during the last ten or fifteen
years, these differences have decreased considerably (Bell and
Chasker 1970, Christensen 1971, Hunt 1974, Israel and others 1970,
Sigusch and Schmidt 1973, Sorensen 1973, Schmidt and Sigusch
1971, 1972, Zetterberg 1969). Such changes indicate that social
factors have considerable importance; they do not indicate that
social factors are all-important. Even among our youngest groups
there were large differences in libido between the sexes, and even
in the countries that have shown the fastest approach to equality
(Denmark, Sweden, West Germany and the metropolitan areas
along the East and West Coasts of the United States), differences in
attitudes entering our libido factor are still considerable. As Gold¬
berg (1973) points out in his book on The Inevitability of Patriarchy,
although there may be differences between different societies in the

228
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

degree of male dominance, yet patriarchy is universal; there is a

vital role to be played by social factors, against a backdrop of
biologically determined differences.
It seems logical that the most relevant biological factors in this
connection should be the sex hormones, with the androgens favour¬
ing masculine attitudes and behaviours, high libido and P, while
the estrogens might favour feminine attitudes and behaviours, low
libido and low P. As regards P (and psychotic behaviour in general)
the evidence has been reviewed by Eysenck and Eysenck (1976),
and there appears to be some ground for assuming that this rather
simple-minded hypothesis is not entirely along the wrong lines. A
more general review of the influence of hormones on behaviour is
given in Levine (1972); this is more likely to convince the reader
of the incredible complexity of the field, and our lack of detailed
knowledge (particularly in relation to humans), than to suggest that
rash formulations such as that relating androgens and estrogens to
male and female behaviours respectively have much empirical sup¬
port. The true relationship is almost certainly much more complex
than this, and it is unfortunately true that there is very little direct
evidence in humans on which to build more appropriate hypotheses.
We have already discussed Money’s (1961a, 1961b, 1965) hypothesis
that androgens are the active cause of libidinal behaviour in men
and women, and the evidence suggesting that this hypothesis too is
probably overly simple, although it may be along the right lines.
Male and female hormones almost certainly interact in complex
ways, and the time at which they exert their influence seems to be
crucial. There almost certainly exists a strong causal relationship
between androgens and P+ behaviour, and between androgens and
high libido; the influence of estrogens is less clear, and the inter¬
action between the two sets of hormones is still a mystery. This
area of interdisciplinary research between psychology, biochemistry
and medicine promises to be one of the most interesting and fruitful
in the immediate future.
We have thus far discussed certain deductions from our general
theory, and attempted in various ways to find support or dis-
confirmation of these deductions. But of course there are very
many more deductions that follow from our theoretical framework,
and it would be a task of supererogation to list these here. However,
in order to give the reader a taste of the sort of detailed prediction
that can be made from personality theory on to sexual behaviour.

229
 SEX AND PERSONALITY

it may be worthwhile to mention en passant the fact that extraverts

seem to experience their highest state of cortical arousal in the
evening, introverts in the morning (Colquhoun 1960, Colqhuoun
and Corcoran 1964, Blake 1967). It would be tempting from this
well-established fact to make certain predictions which might serve
to explain the sad face of Strephon and poor Chloe in Charles
Hanbury-Williams’ poem, written around 1750:

Ye famed physicians of this place.

Hear Strephon’s and poor Chloe’s case
Nor think that I am joking;
When she would, he cannot comply;
When he would drink, she’s not a-dry.
And is this not provoking?

At night, when Strephon comes to rest,

Chloe receives him on her breast.
With fondly folding arms:
Down, down he hangs his drooping head.
Falls fast asleep, and lies as dead.
Neglecting all her charms.

Reviving when the morn returns.

With rising flames young Strephon burns.
And then would fain be doing:
But Chloe, now asleep or sick.
Has no great relish for the trick.
And sadly balks his wooing.

O cruel and disastrous case.

When in the critical embrace
That only one is burning!
Dear doctors, set this matter right;
Give Strephon spirits over night.
Or Chloe in the morning

The implicit hypothesis that Strephon was introverted, Chloe

extraverted, might with advantage be put to a more general test,
involving couples where one partner was introverted, the other
extraverted; the prediction would be that troubles similar to those
afflicting the lovers celebrated in the poem could set the matter
right by following the lead given by 1,600 members of the local
goldminers’ union in Vetikoula (Fiji), who demanded that their

230
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

present lunchtime be extended to include ‘a thirty-minute midday

sex break’. According to their secretary, they believed ‘that mid¬
day is the best time for sex’, neatly cutting the Gordian knot of
personality differences in circadian rhythms.
If we accept the fact that both social and biological factors pro¬
duce considerable diversity in the sexual attitudes and behaviours
of men and women in our (or any other) society, even though our
understanding of either the social or the biological factors is
deplorably weak and partial at best, then we may be able to say
something about the current (and in fact age-old) debates raging
between apostles of permissiveness and protagonists of conserva¬
tive values, or libertines and puritans, in a short phrase. It will
be clear from our general discussion that there are four major
groups formed in terms of our factors. First, there are those who
have little libido, but considerable satisfaction from their sex lives;
on the whole, stable introverts with little evidence of P. These
people would probably constitute the monogamous section of the
population, and their contentment testifies to the fact that this is a
viable and perhaps enviable position. Advocates of permissiveness
often by implication deny the possibility of achieving this com¬
bination, but the facts clearly are against them.
Next we have the happy philanderer, the stable extravert in the
high-libido, high-satisfaction group; devoting much of his energy
to sexual adventures with many different women, never settling
down permanently, or, if apparently settled down, nevertheless
always on the lookout for extra-marital experiences. Although
puritans might wish to deny the possibility of regarding this as a
feasible pattern, nevertheless the evidence suggests that it does
occur quite frequently, and does not produce less satisfaction to the
person involved than does the first pattern considered. There may
be an age shift, in the sense that the high-libido pattern is more
appropriate to the young, the low-libido pattern to the old, with the
middle-aged in between. The introvert may start out with the sow¬
ing of at least some wild oats; the extravert may end up happily
married in his dotage. Our samples were not extensive enough to
test this hypothesis, and in any case it would probably need follow¬
up studies to investigate such theories properly.
The other two patterns both show dissatisfaction with the sex
lives experienced. The low-libido male or female tends to be the
unstable introvert, with strong inhibitions, feelings of guilt, diffi-

231
 SEX AND PERSONALITY

culties in meeting people of the opposite sex and generally pre¬

cluded from adequate sexual responses. The high-libido male or
female tends to be high on the P scale, and may also add the
quasi-hysterical combination of high N and high E (Eysenck 1971c).
Pseudo-nymphomanic behaviour in the female, Don Juanism in the
male characterises people in this group; they might be said to be
those who give permissiveness a bad name. There is something
attractive about the permissiveness of the extra vert; there is little
that is attractive about the permissiveness of the high P scorer, who
combines hostility, aggressiveness and impersonal attitudes with
his sexual desires, reducing sex to a biological need, on a level with
eating and defecating. The interesting point, and one that will give
support to the puritans, is that this pattern of sexual behaviour is
self-defeating, leading to low satisfaction.
We thus have two patterns of behaviour that are acceptable
because they lead to personal satisfaction and happiness - the
introvertive traditional one of monogamy and family life, the extra-
verted one of wine, women and song. Either pattern may lead to
dissatisfaction and unhappiness, however, when there is an admix¬
ture in the personality of psychiatric, abnormal, pathological
characteristics, whether of the quasi-neurotic or the quasi-psychotic,
quasi-psychopathic variety. (The term ‘quasi-’ is used here to make
it clear that none of our subjects were clinically ill, or were in
psychiatric treatment.) In more extreme cases, this pathological
admixture may lead to definite physical symptoms, such as frigidity,
lack of orgasm, ejaculatio praecox and impotence. We may thus
distinguish between normal and abnormal patterns of adjustment,
i.e. patterns where men and women receive satisfaction from their
course of conduct and others where they do not, and where they
may suffer from actual physical symptoms of psychiatric disorder
in relation to their sex lives. Within either pattern, we have a per-
missive-libidinal form and an inhibited-traditional form. Whether
a person falls into one or other of these patterns will depend to
some extent on his personality, and on his genetic makeup; it will
also depend, of course, on myriad other factors, ranging from
accidents of upbringing and physical endowment to peer influences
and unpredictable events of adult life. An inch added to Cleopatra’s
nose, it is said, might have altered the course of history; an inch
added to, or taken away from, a girl’s breast measurements may
change her destiny in equally marked fashion. Personality factors

232
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

do not determine a person’s fate; they merely nudge it along in

certain directions; nevertheless, the nudge can be pretty compelling
at times.
The fact that these personality and behaviour differences are
so marked, and so firmly established on a genetic basis, makes the
task of society very difficult in laying down rules and laws that
would lead to the greatest happiness of the greatest number and
at the same time safeguard the interests of third parties, such as
children and other dependents. Clearly the greatest possible per¬
missiveness on the part of the State, compatible with these safe¬
guards, is most likely to achieve these objectives; our own society
still is curiously impermissive and downright vengeful when facing
such issues as abortion and prostitution. Religious people and
others who judge such matters in terms of prejudged principles
derived from putatively divine sources of course can not be
expected to share such views; it is still apparently very difficult for
most people not to try and impress their own value judgements on
others. Yet this is what the principles of freedom demand, and the
political forms of democracy are intended to give expression to
such principles.
It might be thought that this endorsement of permissiveness on
the part of the government in dealing with personal and private
matters such as sex could and should be extended to what is
customarily called ‘permissiveness’ in the media, in social life and
in society generally. This is not necessarily so. The term ‘permissive¬
ness’ is probably badly chosen and does not adequately portray what
goes on in society; so-called permissiveness can be as dictatorial
and non-permissive, in one direction, as was puritanism. To take
but one example, lessor and lessor (1976) studied 600 adolescents
and found that sexual habits had changed so much that now it is
the boy or girl who does not lose his or her virginity at an early
age who is looked upon as psychologically abnormal. There is thus
an enormous pressure on young people to behave in a manner that
may be contrary to their personality makeup and their value judge¬
ments; this is not true permissiveness, which would enable
different persons to express in their conduct the outlook and the
values that their heredity, their personality and their upbringing
have produced in them. So-called permissiveness is not permissive
towards traditional values; it is simply a fine-sounding name given
to a value system equally irrational, equally dogmatic and equally

233
 SEX AND PERSONALITY

strictly enforced by peer sanctions as the traditional system. True

permissiveness would recognise the relativity involved in such
value judgements, and would not put pressure on young people
(or their elders), to fall in line with one or the other alternative
as the only system worthy of consideration.
By the same token, permissiveness as here understood does not
imply agreement with the proposition that there should be no
censorship over obscene and pornographic material in the printed
word, television, the cinema or the theatre. The argument may
perhaps first be stated in a general form, to be followed by a
more detailed look at some of the facts. It is fairly universally
agreed that freedom of expression is a vital strand in the general
fabric of modern democracy, and that censorship is an evil that
requires adequate reasoning and potent demonstration of its neces¬
sity before it can be tolerated. We do admit the necessity of censor¬
ship and the need for interference of the law in the freedom of
expression when we deal with such issues as race; incitement to
racial hatred and expressions of racial prejudice and discrimination
are rightly forbidden by law. Limitations of the freedom of
expression may therefore be tolerated in a democratic society,
however permissive, in order to safeguard the rights of minority
groups.
Now it can be said that much of what passes in modern society
as pornography is in fact serving the cause of male chauvinism.
Just as Nicolas Chauvin, who became immortal through his
boastful, warlike, over-enthusiastic patriotism, abused freedom of
expression and turned it into licence to insult and maltreat those
of another nationality, so pornographic writings and films incite
men to insult and maltreat women. The values expressed in such
works, whether intentionally or not, are male values, particularly
those of the high-P, high-libido male; women serve only as sex
objects in these works, and may be maltreated, raped, beaten and
generally humiliated - in short, treated no better than second-class
citizens. If we object to seeing blacks as slaves, entertainers and
generally servants of the dominant whites, then we should equally
object to seeing women as slaves, entertainers and generally servants
of the dominant males. Permissiveness, as we have argued, must
stop at that point where it interferes with the rights of others; here
we would seem to encounter such a limitation to general per¬
missiveness.

234
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

Readers who may find that this view exaggerates the dehumanis¬
ing and downgrading aspects of pornographic films, as far as women
are concerned, may like to consider some findings from our own
thematic study of a random sample of such films, confiscated by the
customs authorities or the police. In the majority of cases, the film
ends by the male withdrawing his penis and ejaculating the sperm
all over the face of the female. This has no erotic appeal (to judge
by the comments of our subjects), but simply signifies, in a symbolic
form, the subjection of the woman; it recalls the victor in tribal
warfare urinating over the face of the vanquished. Sado-masochistic
episodes proliferate, always with the woman the first unwilling, then
willing victim. Scenes include rape, as in a film showing a burglary;
the burglar is discovered by the woman, sleeping alone in her bed.
He jumps on her and rapes her, in spite of her struggles, which
gradually diminish and end in enthusiastic cooperation. In another
film, the boss finds fault with the work of his secretary; he gives her
the option of dismissal or being spanked. She choses the latter, is
put across his knees, has her panties removed and is spanked; this
quickly leads to cooperative fornication. The general moral of these
and other films is: ‘Treat them rough, and they’ll get to like it.’
Only the most occasional film showed scenes of tenderness, wooing
or feeling of any kind. It is interesting to note that, among our sub¬
jects at least, the dehumanising and downgrading parts of the films
did not produce a greater physiological reaction, or a higher erotic
rating, than did more acceptable scenes; thus the excuse that these
aspects of biological sex are emphasised because they produce
strong erotic reactions is not valid. The only apparent purpose is to
emphasise the superiority of the male and the inferiority of the
female.
It is often argued that in fact these objections are immaterial
because the evidence does not seem to show that pornographic
films have any observable effects. Usually the conclusions of the
American Presidential Commission Report on Obscenity and
Pornography are cited in this connection, where it is stated that ‘in
sum, empirical research designed to clarify the question has found
no evidence to date that exposure to explicit sexual material
plays a significant role in the causation of delinquent and criminal
behaviour among youths and adults’. The general conclusion that
‘there is no evidence’ of the effects of pornography on sexual
behaviour is often repeated as if it were true; yet in fact there is a

235
 SEX AND PERSONALITY

large body of such evidence, so much so in fact that at present there

can be very little doubt about the matter. Cline (1974) has brought
together both the positive evidence on this point, and also the
criticisms of the Commission’s Report, and there would be little
point here in trying to summarise all the wealth of material referred
to by him.
What happened, as Cline (who also assisted in the preparation of
the Minority Report appended to the Majority Report) shows clearly,
was that the membership of the Commission was determined largely
on ideological grounds, with the express aim of justifying the end
of censorship. The report itself neglected to mention nearly 90 per
cent of the research carried out for the Commission, misrepresented
much of what was quoted and over-interpreted restricted data,
collected on small and unrepresentative samples. As Cline put it,
there was a great variety of common errors, omissions and drawing
of incorrect conclusions from flawed research, as well as a manifest,
systematic bias in marshalling evidence and reporting’. All this is
clearly documented, and anyone familiar with the literature, and
having read the ten volumes of detailed evidence produced by the
research teams of the Commission (and not relying on the biased
summaries and potted versions contained in the Commission Report
itself), will have no doubt about the determination of the majority
of the Commission members to arrive at the conclusion to which
they were committed from the beginning. Contrary to popular
opinion, then, there is much direct evidence of the enormous power
that even quite short snippets of film can have in altering attitudes
and behaviours over quite lengthy periods; furthermore, these
experimental results are in good agreement with theory regarding
the efficacy of ‘modelling’ (imitation) and desensitisation, and the
very large clinical and experimental literature that has been built
up around these concepts. To disregard all this evidence is to fly
in the face of scientific fact; it is simply not feasible any more to
dismiss the view that pornographic and obscene films and publi¬
cations have an effect on the behaviour of people, or on their
attitudes, and it is not even open to anyone to claim that the verdict
should be ‘not proven’.
It is of course much more difficult to prove beyond the shadow
of a doubt that the recent tremendous increases in the incidence
of rape, molestation, indecent assault, unlawful intercourse and
violence generally are linked causally with the presentation of such

236
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

activities on the screen, and are caused directly by such presenta¬

tions. It would seem unreasonable to demand such direct proof; in
over fifty years the medical profession has not been able to prove
indisputably that cigarettes act as a carcinogenic agent and cause
lung cancer in smokers. The correct position to adopt, in my sub¬
mission, is the one taken by the Government Committees on Drug
Safety; new drugs have to be shown to be safe by the manufacturers
(within human reason), and it is not required of the Committee to
prove them to be lethal! Similarly, grossly pornographic and violent
pictures, showing in detail rape and other unlawful activities, should
be required to prove their innocence, and not allowed to be shown
until they had done so. It is unreasonable to require private persons
who suspect (probably rightly) that such pictures have a very des¬
tructive influence on society to finance the very time-consuming,
expensive and highly technical research that is needed to prove the
point. Such research should be financed from the profits of the
pornographic industry itself, in order to support, if possible, their
plea of harmlessness.
Such financial support should of course be indirect, i.e. be given
to an authoritative body (such as the British Psychological Society,
or the Medical Research Council), and disbursed through them.
The reasons for this are obvious; pornographic film-makers are
interested parties, and would do their best to influence the member¬
ship and the research activities of any body set up to study these
questions. This is what happened, for instance, in the famous case
of the U.S. Surgeon General’s report on television and social
behaviour. The committee set up to carry out the work had some
of its proposed members vetoed by the large television companies,
on the grounds that they were expected to make the report too
realistic and unfriendly to the television companies! The story is
told in detail in Cline’s book; no wonder that the Committee finally
appointed ‘misrepresented some of the data, ignored some of it and
buried all of it in prose that was obviously meant to be unreadable
and unread’, as Newsweek so admirably put it. Independence of
fact-finding and of conclusion-drawing bodies from the practitioners
of the arts of pornography and violence is an overriding need in the
scientific study of this field.
It is sometimes said that artistic integrity demands the public
showing of what is in fact pornography. Psychologically and artisti¬
cally this is not only not true, but it is exactly the opposite of the

237
 SEX AND PERSONALITY

truth. Consider the plea recently made that Lady Macbeth should
be allowed to sleepwalk in the nude ‘because in her time she would
not have worn a nightgown’. It is not only that this plea itself is
probably untrue and certainly artistically irrelevant (should she also
sleepwalk unwashed and evil-smelling?); it should not need a
psychologist to point out that the appearance of a well-known
actress in the nude would break the spell of the play completely, and
invite rude comments on the size of her breasts, and the shape of
her behind. Plays and films are meant to create emotional reactions
by art and indirection, not by blatant sensory appeals; such direct
sensory appeals detract from the message of the medium.
The general argument here put forward is a very simple one.
Given that films, plays, television performances and books have a
powerful effect on viewers, listeners and readers, a point not now
seriously doubted by competent psychologists, then the public show¬
ing of scenes of rape, degradation and general abuse of women, and
their portrayal as sex objects pure and simple, is likely to have
effects on male attitudes and behaviour that go counter to the trend
towards greater sexual equality which has been one of the more
acceptable and desirable tendencies characteristic of this century.
It is not necessary to imagine that all modern pomographers are
motivated by a desire to destroy modem civilisation, although some
no doubt believe, with Richard Neville, that the weapons of revolu¬
tion are obscenity, blasphemy and drugs’. A far more likely motive
is simple greed. But in any case motives are largely irrelevant;
effects are the important consideration. Ideally one might like to
appeal to all concerned to heed the sentiments expressed so power¬
fully in Masefield’s poem addressed to his mother:

What have I done, or tried, or said

In thanks to that dear woman dead?
Men triumph over women still.
Men trample women’s rights at will,
and man’s lust roams the world untamed.

O grave, keep shut lest I be ashamed!

The probability of such an appeal being heeded is not high, and

in consequence one must seek more powerful help in the form of
censorship, however agonising the choice may be. It is interesting
to note, in this connection, that although it is usually men of the

238
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

left who demand ‘pornographers’ licence’, it is the communist

countries, from the U.S.S.R. to China, that are least tainted by the
pomographer’s trade, and which uphold old-fashioned virtues of
virginity and marriage. Perhaps Richard Neville had a point, after
ah.
It will be seen that I have been dealing largely with pornographic
products, which would be related particularly to our fourth quad¬
rant, i.e. the high-libido-high-satisfaction, pathological one. It
is possible to produce light-hearted, gay (in the proper meaning of
the word!) and amusing pornography which would be suitable for
the first quadrant, i.e. the extraverted, high-libido-high-satisfaction
one, and which would give expression to healthy male and female
sensuality. Contrary to some anti-pornographers, I do not feel that
the same objections apply to works of this nature, particularly if
produced in a tasteful and artistic fashion. Clearly, in the words
used by Cline (1974) in the title of his book, the question is: Where
do you draw the line? Arguments range all the way from the
completely puritan and censorious to the completely libertarian,
but in practice every society draws the line somewhere, and the
position here suggested is determined by the general thrust of our
experimental results; that which is pathological, and likely to pro¬
mote pathological activities, is out, while anything that does not fall
into this category is in. It is not suggested that all potential viewers
of pathological material are likely to be tempted to indulge in
pathological practices; there is good evidence that some at least,
usually already psychiatrically ill to some extent, will be so
tempted, and it seems likely, on the existing evidence, that for many
people, if not the majority, there will be a profound impact in the
direction of devaluation of female sexuality, feminity and female
ideals. In the same way, racialist propaganda and incitement to
racial violence are ruled out by law, not because everyone who
is exposed to such propaganda will become a racist, but because
some, usually already psychologically flawed in certain ways, will
be encouraged to live out their phantasies. (The concept of the
‘marginal viewer’, corresponding to the notion of marginal utility
of the economist, might be useful here.)
Our results may also carry a message to those who would enforce
sex education on often unwilling schoolchildren, and to those who
would teach everyone to ‘do sex by numbers’, and assiduously read
the many sex manuals now available. Such an approach clearly

239
 SEX AND PERSONALITY

mirrors attitudes characteristic of the ‘high-libido’ group, but is

antagonistic to the views of the ‘low-libido’ group. To the tough-
minded person, sex is largely a biological activity, impersonal and
designed to produce a maximum of sensuous pleasure; like any
other sport, it demands competence, training and competitive
excellence. What more appropriate than to teach sex at school,
like football, and to publish manuals teaching people the various
positions which past experience has found to be useful and profit¬
able? In the same way, the gourmet ‘lives to eat’; both worship at
the shrine of old-fashioned hedonism, of a particularly simple-
minded kind.
To the tenderminded sex is only in part a biological fact; they
would rather see it in terms of Plato’s beautiful parable of the
circle cut in two, and reconstituted when the male and female half
are again united. Sex to them has a spiritual, value-laden, com¬
panionate side which may be far more important than biological
factors such as number of orgasms achieved, or variety of positions
mastered. It is wrong to impose the sexual teaching appropriate to
the toughminded child on the tenderminded; this is propaganda at
its worst, directed at defenceless children, often against the wishes
of their parents. It would be equally wrong, of course, to make
illegal the sale of sex manuals; they serve a useful function for those
who feel in need (although the factual information given in many
of these manuals must be said to be of very doubtful value, and
often downright wrong in terms of our present-day physiological
knowledge). But there is an important difference. Sex manuals are
bought only by those who wish to read them; sex education is
enforced on children who may not wish for it, and whose parents
may be opposed to such teaching. When it is realised that most
teachers are even less equipped for this task than are the writers
of sex manuals for theirs, one must conclude that the good that may
be achieved is probably cancelled out by the bad.
Although it is doubtful if the vaunted liberation from Victorian
repression that modern society has achieved is entirely to the good,
the disturbing increase in male impotence (difficult to prove beyond
doubt, yet attested to by many experts working in this field) may
be a direct consequence of modern stress on performance. Many
men not up to Wimbledon standards feel inferior, and such feelings
only make matters worse, and may produce impotence. Similarly,
many women believe that failure to achieve instant orgasm

240
 SEXUAL ATTITUDES AND SOCIAL CONSEQUENCES

reflects adversely on them, making the achievement of orgasm

even more difficult. Or else they may blame their unfortunate
husbands, inflicting guilt feeling on them which impede their per¬
formance in turn. There is a double bind involved in sexual rela¬
tions which is not solved by the added stresses of modern insistence
on ‘performance’. So-called permissiveness in fact imposes rules and
standards that are as strident and compelling as were the entirely
different ones imposed by puritanism and Victorian orthodoxy;
neither set of standards is suitable for everyone. True permissive¬
ness involves an all-embracing tolerance for human diversity, and
a recognition that toughminded and tenderminded males and
females have different ideals, different standards, different ways of
reaching happiness. These differences are only in part susceptible
to social influence and manipulation; in large part they are anchored
in heredity, in personality, in the very physiology of the person
concerned. We should allow maximum freedom to all to develop
along the lines of their biological programming, provided only
that the rights of others are protected. Permissiveness in this sense
is clearly the essence of democratic society; its achievement can
be advanced only by gaining a much better understanding of sexual
behaviour than we possess at the moment, and in particular a better
understanding of the relation between sex and the mysterious
processes of personality.

241
 References and
Author index

Bartlett, M. S. (1947) ‘The use of transformations’, Biometrics 3,

39-52 200
Bell, R. R. and Chasker, J. B. (1970) ‘Premarital sexual experience
among coeds: 1958 and 1968’, Journal of Marriage and the Family
32, 81-4 228
Bentler, P. M. (1968) ‘Heterosexual behaviour assessment - 1: males’,
Behaviour Research and Therapy 6, 21-5 77
Blake, M. J. F. (1967) ‘Relationship between circadian rhythms of
body temperature and introversion-extra version’, Nature 215, 896-7
230
Boneau, C. A. (1959) ‘The effects of violations of assumptions under¬
lying the t test’. Psychological Bulletin 57, 49-64 140
Bouscher, B. J. (1957) ‘Sex differences in the foetal pelvis’, American
Journal of Physiology and Anthropology 15, 581-600 217
Bowerman, C. and Day, B. (1956) ‘A test of the theory of comple¬
mentary needs as applied to couples during courtship’, American
Sociological Research 21, 602-5 109
Brady, J. P. and Levitt, E. E. (1965) ‘The scalability of sexual ex¬
periences’, Psychological Record 25, 275-81 77
Bragan<ja, G. (1972) ‘A comparative study of attitudes to sex of a
sample of sexual offenders, arsonists and non-sexual offenders, resi¬
dent at Broadmoor Hospital’ (Southampton: unpublished B.Sc.
dissertation) 177
Burgess, E. W. and Wallis, P. (1944) ‘Homogamy in personality
characteristics’, Journal of Abnormal and Social Psychology 39,
475-81 109
Bynner, J. M. (1969) ‘The association between adolescent behaviour
and attitudes as revealed by a new social attitude inventory’ (London :
unpublished Ph.D. thesis) 14, 69
Byrne, D. (1971) The Attraction Paradigm (New York: Academic
Press) 139
Byrne, D., Cherry, F., Lamberth, J. and Mitchell, H. E. (1973)
‘Husband-wife similarity in response to erotic stimuli’, Journal of
Personality 41, 385-94 139, 140

243
 SEX AND PERSONALITY

Cattell, R. B. (1952) Factor Analysis (New York: Harper) 87

Cattell, R. B. and Nesselroade, J. R. (1967) ‘Likeness and com¬
pleteness theories examined by sixteen personality factor measures in
stably and unstably married couples’, Journal of Personality and
Social Psychology 7, 351-61 109
Chilton, B. (1972) ‘Psychosexual development in twins’, Journal of
Biosocial Science 4, 277-86 196
Christensen, H. T. (1971) Sexualverhalten und Moralishe Reinbeck
(Hamburg: Rowohlt) 228
Cline, V. B. (1974) Where do you draw the line? (Provo, Utah:
Brigham Young University Press) 236, 237, 239
Cochrane, R. and Duffy, J. (1974) ‘Psychology of scientific method’,
Bulletin of the British Psychological Society 27, 117-21 22-7
Cohen, M. L. and others (1971) ‘The psychology of rapists’, Seminars
in Psychiatry 3 no 3
Colquhoun, W. P. (1960) ‘Temperament, inspection efficiency and
time of day’, Ergonomics 3, 377-88 230
Colquhoun, W. P. and Corcoran, D. W. J. (1964) ‘The effects of
time of day and social isolation on the relationship between tempera¬
ment and performance’, British Journal of Social and Clinical
Psychology 3, 226-31 230
Comrey, A. L. and Levonian, E. (1958) ‘A comparison of three paired
coefficients in factor analysis of MMPI items’, Educational Psycho¬
logical Measurement 18, 739-48 87
Conner, R. L. (1972) ‘Hormones, biogenic amines and aggression’, in
Levine (1972) 77
Cooper, A. J. (1969) ‘Some personality factors in frigidity’, Journal of
Psychosomatic Research 13, 149-55 17
Coppen, A. (1965) ‘The prevalence of menstrual disorders in psychiatric
patients’, British Journal of Psychiatry 111, 155-67 17
Daitzman, R. J. (1976) ‘Personality correlates of androgens and estro¬
gens’ (Delaware: unpublished Ph.D. thesis) 215-16
Davidson, J. M. (1972) ‘Hormones and reproductive behavior’, in
Levine (1972) 77, 214
Eaves, L. J. (1970) ‘The genetic analysis of continuous variation’, British
Journal of Mathematical and Statistical Psychology 23, 189-98 202
Eaves, L. J. (1976) ‘A model for sibling effects in man’. Heredity
36, 205-14 194, 206
Eaves, L. J. 214
Eaves, L. J. and Eysenck, H. J. (1974) ‘Genetics and the development
of social attitudes’, Nature 249, 288-9 150, 171, 196
Eaves, L. J. and Eysenck, H. J. (1975) ‘The nature of extraversion:
a genetical analysis’, Journal of Personality and Social Psychology
32, 102-12 195, 201
Eaves, L. J. and Eysenck, H. J. (1976) ‘Genetic and environmental
components of inconsistency and .unrepeatability in twins’ responses
to a neuroticism questionnaire’, Behaviour Genetics 6, 145-60 211
Eaves, L. J. and Eysenck, H. J. 209, 210

244
 REFERENCES and AUTHOR INDEX

Ehrenkranz, J., Bliss, E. and Sheard, M. H. (1974) ‘Plasma testos¬

terone : correlations with aggressive behavior and social dominance
in man’. Psychosomatic Medicine 36, 469-75 219
Ehrhardt, A. A. and Money, J. (1967) ‘Progestin-induced hermaphro¬
ditism : IQ and psychosexual identity in a study of ten girls’, Journal
of Sexual Research 3, 83-100 215
Ellis, A. and Brancale, R. (1956) The Psychology of Sex Offenders
(Springfield, Mass.: C. C. Thomas)
Eysenck, H. J. (1947) Dimensions of Personality (London: Routledge
and Kegan Paul) 18
Eysenck, H. J. (1954) The Psychology of Politics (London: Routledge
and Kegan Paul) 24, 130, 150, 151, 156, 171, 173
Eysenck, H. J. (1964) (ed.). Experiments with Drugs (New York:
Pergamon Press)
Eysenck, H. J. (1967) The Biological Basis of Personality (Springfield,
Mass.: C. C. Thomas) 9-13, 23, 24, 214
Eysenck, H. J. (1970a) The Structure of Human Personality, 3rd ed.
(London: Methuen) 10, 18, 197
Eysenck, H. J. (1970b) Crime and Personality (London: Paladin) 12
Eysenck, H. J. (1970c) ‘Personality and attitudes to sex: a factorial
study’, Personality 1, 355-76
Eysenck, H. J. (1971a) ‘Personality and sexual adjustment’, British
Journal of Psychiatry 118, 553-608 12, 69, 75, 104, 187
Eysenck, H. J. (1971b) ‘Masculinity-femininity, personality and sexual
attitudes’. Journal of Sexual Research 7, 83-8 61, 104, 130
Eysenck, H. J. (1971c) ‘Hysterical personality and sexual adjustment,
attitudes and behaviour’, Journal of Sexual Research 7, 274-81 16,
65, 74, 232
Eysenck, H. J. (1972) ‘Personality and sexual behaviour’. Journal of
Psychosomatic Research 16, 141-52 68, 70, 133
Eysenck, H. J. (1973a) ‘Personality and attitudes to sex in criminals’.
Journal of Sexual Research 9, 295-306 73
Eysenck, H. J. (1973b) The Measurement of Intelligence (Lancaster:
Medical and Technical Publishers) 194
Eysenck, H. J. (1974) ‘Personality, premarital sexual permissiveness,
and assortative mating’, Journal of Sexual Research 10, 47-51 109
Eysenck, H. J. (1975a) ‘Genetic factors in personality development’, in
A. R. Kaplan (ed.), Human Behaviour Genetics (Springfield, Mass.:
C. C. Thomas) 18, 133, 195
Eysenck, H. J. (1975b) ‘The structure of social attitudes’, British
Journal of Social and Clinical Psychology 14, 323-32 151, 167
Eysenck, H. J. (1976) ‘Personality and participation in group sex: an
empirical study’, Revista Latinoamericana de psicologia (to appear)
139
Eysenck, H. J. and Coulter, T. (1972) ‘The personality and attitudes
of working-class British communists and fascists’, Journal of Social
Psychology 87, 59-73 174

245
 SEX AND PERSONALITY

Eysenck, H. J. and Eysenck, S. B. G. (1964) Manual of the EPI

London: University of London Press) 20, 74
Eysenck, H. J. and Eysenck, S. B. G. (1969) Personality Structure and
Measurement (London: Routledge and Kegan Paul) 10, 18, 31, 47,
87
Eysenck, H. J. and Eysenck, S. B. G. (1975) Manual of the E.P.Q.
(Eysenck Personality Questionnaire) (London: University of London
Press) 18, 19, 30, 127, 167, 198
Eysenck, H. J. and Eysenck, S. B. G. (1976) Psychoticism as a Dimen¬
sion of Personality (London: University of London Press) 18, 21,
30, 76, 180, 214, 215 229
Eysenck, H. J. and Wilson, G. (1974) The Experimental Study of
Freudian Theories (London: Methuen) 8
Eysenck, S. B. G. (1961) ‘Personality of pain assessment in childbirth
of married and unmarried mothers’. Journal of Mental Science 108,
417-30 69
Eysenck, S, B. G. and Eysenck, H. J. (1970a) ‘Crime and personality:
an empirical study of the three-factor theory’, British Journal of
Criminology 10, 225-39 19
Eysenck, S. B. G. and Eysenck, H. J. (1970b) ‘A factor-analytic
study of the lie scale of the Junior Eysenck Personality Inventory’,
Personality 1, 3-10 20
Eysenck, S. B. G. and Eysenck, H. J. (1971a) ‘Crime and personality:
item analysis of questionnaire responses’, British Journal of Crimin¬
ology 11, 49-62 19
Eysenck, S. B. G. and Eysenck, H. J. (1971b) ‘A comparative study of
criminals and matched controls on three dimensions of personality’,
British Journal of Social and Clinical Psychology 10, 362-6 19
Eysenck, S. B. G. and Eysenck, H. J. (1971c) ‘Attitudes to sex, per¬
sonality and lie scale scores’, Perceptual and Motor Skills 33, 216-18
74
Eysenck, S. B. G., Nias, D. K. B. and Eysenck, H. J. (1971) ‘The
interpretation of children’s lie scale scores’, British Journal of Educa¬
tional Psychology 41, 23-31 20
Ferguson, G. A. (1941) ‘The factorial interpretations of test difficulty’,
Psychometrika 6, 323-9 87
Fisher, S. (1973) The Female Orgasm (London: Allen Lane) 17 18
29
Ford, C. S. and Beach, F. A. (1952) Patterns of Sexual Behavior
(London: Eyre and Spottiswoode) 226
Foss, G. L. (1951) ‘The influence of endogens on sexuality in women’
Lancet 1, 667-9 77
Fuller, J. L. and Thompson, W. R. (1960) Behavior Genetics (New
York: Wiley) 192
Giese, H. and Schmidt, A. (1968) Studenten Sexualitdt (Hamburg:
Rowohlt) 14, 15, 16, 110
Giese, H. and Schimdt, A. 69

246
 REFERENCES and AUTHOR INDEX

Goldberg, S. (1973) The Inevitability of Patriarchy (New York:

Morrow) 226, 228
Gottesman, I. I. (1965) ‘Personality and natural selection’, in S. G.
Vandenberg (ed.). Methods and Goals in Human Behaviour Genetics
(London: Academic Press) 109
Gray, H. (1949) ‘Psychological types in married people’. Journal of
Social Psychology 29, 189-200 109
Guilford, J. P. (1975) ‘Factors and factors of personality’. Psycho¬
logical Bulletin 82, 802-14 139
Hackett, T. P. (1971) ‘The psychotherapy of exhibitionists in a court
clinic setting’. Seminars in Psychiatry 3 no 3
Havlicek, L. L. and Peterson, N. L. (1974) ‘Robustness of the
t test: a guide for researchers on effect of violations of assumptions’.
Psychological Reports 34, 1095-114 140
Hendrickson, A. and White, P. O. (1964) ‘Promax: a quick method
for rotation to oblique simple structure’, British Journal of Statistical
Psychology 17, 65-70 31, 87
Hewitt, J. H. (1974) ‘An analysis of data from a twin study of social
attitudes’ (Birmingham: unpublished Ph.D. thesis) 209, 210, 211
Hill, M. (1968) unpublished Ph.D. thesis quoted by H. J. Eysenck,
‘Genetic factors in personality development’, in A. R. Kaplan (ed.),
Human Behaviour Genetics (Springfield, Mass., 1974: C. C. Thomas)
109
Hunt, M. (1974) Sexual Behavior in the 1970s (Chicago: Playboy Press)
228
Hum, C. (1972) Males and Females (Harmondsworth: Penguin Books)
226
Insel, P. M. (1971) ‘Family similarities in personality, intelligence and
social attitudes’ (London: unpublished Ph.D. thesis) 110
Israel, J., Gustavsson, J., Eliasson, R. M. and Lindberg, G. (1970)
‘Sexuelle Verhaltensformen der schwedischen Grosstadtjugend’, in
M. B. Bergstrom-Walan (ed.), Modellfall Skandinavien? (Reinbeck:
Rowohlt) 228
Jaffe, Y., Malamuth, N., Feingold, J. and Feshbach, S. (1974) ‘Sexual
arousal and behavioral aggression’. Journal of Personality and Social
Psychology 30, 759-64 175
Jensen, A. R. (1973) Educability and Group Differences (London:
Methuen) 23
Jessor, S. and Jessor, R. (1976) ‘Sex: the first time’. Journal of
Personality and Social Psychology (in press) 233
Jinks, J. L. 214
Jinks, J. L. and Fulker, D. W. (1970) ‘A comparison of the biometrical
genetical MAVA and classical approaches to the analysis of human
behaviour’, Psychological Bulletin 73, 311-49 195, 196, 200
Joe, V. C. (1976) ‘Social attitudes and sexual behaviors of college
students’. Journal of Consulting and Clinical Psychology (to appear)
76

247
 SEX AND PERSONALITY

Kagan, J. (1964) ‘Acquisition and significance of sex typing and sex

role identity’, in M. L. Hoffman and Lois W. Hoffman (eds), Review
of Child Development Research (New York: Russell Sage Founda¬
tion), 137-68 28
Kallman, F. J. and Mickey, J. S. (1946) ‘Genetic concepts and Folie
a Deux’, Journal of Heredity 37, 298-305 109
Kinsey, A. C. 5-8, 22, 25, 28, 130
Kirkpatrick, C. and Stone, S. (1935) ‘Attitude measurement and the
comparison of generations’, Journal of Applied Psychology 19, 569-82
175
Kolaszynska-Carr, 65
Kreitman, N. (1964) ‘The patient’s spouse’, British Journal of Psychiatry
110, 159-73 109
Kreitman, N. (1968) ‘Married couples admitted to mental hospital’,
British Journal of Psychiatry 114, 699-718 109
Kreuz, L. E. and Rose, R. M. (1972) ‘Assessment of aggressive
behavior and plasma testosterone in a young criminal population’,
Psychosomatic Medicine 34, 331-6 219
Ktsanes, T. (1955) ‘Mate selection on the basis of personality type: a
study utilising our empirical typology of personality’, American
Sociological Reports 20, 547-57
Landis, C., Bolles, M. and D’Esopo, D. A. (1940) ‘Psychological and
physical concomitants of adjustment in marriage’. Human Biology
12, 559-65 17
Levine, S. (1972) Hormones and Behavior (London: Academic Press)
226, 229
Lewis, N. D. C. and Yarnell, H. (1951) ‘Pathological firesetting’,
Journal of Nervous and Mental Disease, monogr. no 82 (New York)
McCloskey, H. (1967) ‘Survey research in political science’, in G. Y.
Glolk (ed.), Survey Research in the Social Sciences (New York:
Russell Sage Foundation) 28
McKerracher, D. W. and Dacre, A. J. I. (1966) ‘A study of arsonists
in a special security hospital’, British Journal of Psychiatry 112, 1151-4
McNeill, D. and Livson, J. (1963) ‘Maturation rate and body build
in women’, Child Development 34, 25-32 197
Martin, N. G. (1976) Unpublished Ph.D. thesis, Department of
Genetics, University of Birmingham 219
Mather, K. and Jinks, J. (1971) Biometrical Genetics (London:
Chapman Hall) 195
Mather, K. 214
Michaelis, W. and Eysenck, H. J. (1971) ‘The determination of
personality inventory factor patterns and intercorrelations by changes
in real life motivation’, Journal of Genetical Psychology 118, 223-34
20, 21, 76, 173
Milgram, S. (1974) Obedience to Authority (London: Tavistock) 25-7
Mischel, W. (1968) Personality and Assessment (New York: Wiley)

248
 REFERENCES and AUTHOR INDEX

Money, J. (1961a) ‘Components of eroticism in man: 1. The hormones

in relation to sexual morphology and sexual desire’, Journal of
Nervous and Menial Disease 132, 239-48 77, 214, 229
Money, J. (1961b) ‘Sex hormones and other variables in human
eroticism’, in W. C. Young (ed.). Sex and Internal Secretions (Balti¬
more : Williams and Wilkins) 77, 214, 229
Money, J. (1965) ‘Influence of hormones in sexual behavior’, Annual
Review of Medicine 16, 67-82 77, 215, 229
Morgan, C. L. and Remmers, H. H. (1935) ‘Liberalism and con¬
servatism of college students as affected by the depression’. School
and Society 41, 780-4 175
Murstein, B. I. (1971) (ed.) Theories of Attraction and Love (New
York: Springer) 109
Nelson, E. 13
Newcomb, T. and Svehla, G. (1937) ‘Intra-family relationships in
attitude’, Sociometry 1, 180-205 175
Nias, D. (1975) ‘Personality and other factors determining the re¬
creational interests of children and adults’ (London: unpublished
Ph.D. thesis) 110
Nichols, R. C. (1969) ‘The resemblance of twins in personality and
interests’, in Manosewitz, G., Lindzey, G. and Thiessen. G. (eds),
Behavioral Genetics: Method and Research (New York: Appleton-
Century-Crofts) 211
Persky, H., Smith, K. D. and Basu, G. K. (1971) ‘Relation of psycho¬
logic measures of aggression and hostility to testosterone production
in man’. Psychosomatic Medicine 33, 265-71 219
Phoenix, C. (1966) ‘Psychosexual organization in nonhuman primates’,
paper delivered to conference on ‘Endocrine and neural control of
sex and related behaviour’, Puerto Rico, Dorado Beach, Kray 215
Pickford, J. H., Signori, E. I. and Rempel, H. (1967) ‘Husband-wife
difference in personality traits as a factor in marital happiness’.
Psychological Reports 20, 1087-90 109
Popper, K. R. (1959) The Logic of Scientific Discovery (London:
Routledge and Kegan Paul) 23
Popper, K. R. (1972) Objective Knowledge (Oxford: Clarendon Press)
23
Purtell, J. J., Robins, E. and Cohen, M. E. (1951) ‘Observations on
clinical aspects of hysteria’, Journal of American Medical Association
146, 902-9 17
Quirion, N. F. (1970) ‘Extraversion, neuroticism and habituation of the
orienting reaction’ (University of Ottowa: unpublished Ph.D. thesis)
13
Rees, L. (1973) ‘Constitutional factors and abnormal behaviour’, in
H. J. Eysenck (ed.). Handbook of Abnormal Psychology (London:
Pitman), 487-539 197
Reiss, I. L. (1967) The Social Context of Premarital Sexual Permissive¬
ness (New York: Holt, Rinehart and Winston) 131

249
 SEX AND PERSONALITY

Reiter, H. (1970) ‘Similarities and differences in scores on certain

personality scales among engaged couples’, Psychological Reports 26,
465-6 109
Roberts, R. C. (1967) ‘Some concepts and methods in qualitative
genetics’, in J. Hirsch (ed.), Behavior-Genetic Analysis (New York:
McGraw-Hill)
Roos, D. E. (1956) Complementary Needs in Mate Selection (North¬
western University: unpublished Ph.D. thesis) 109
Schellenberg, J. A. and Bee, L. S. (1960) ‘A re-examination of the
theory of complementary needs in mate selection’. Marriage and
Family Living 22, 227-32 109
Schiller, B. (1932) ‘A quantitative analysis of marriage selection in a
small group’, Journal of Social Psychology 3, 297-319 175
Schlegel, W. S. (1969) ‘Konstitutionsbiologie und Verhaltensforschung
im Menschen’, in F. Keiter (ed.), Verhaltensforschung im Rahmen
der W issenschaften vom Meuse hen (Gottingen: Musterschmidt)
228-32 217
Schlegel, W. S. (1975) ‘Parameter Beckenskelett’, Sexualmedizin 4,
228-32 217
Schmidt, G. (1975) ‘Male-female differences in sexual arousal and
behaviour during and after exposure to sexually explicit stimuli’,
Archives of Sexual Behavior 4, 353-65 228
Schmidt, G. and Sigusch, V. (1971) Arbeitersexualitd (Berlin: Luchter-
hand) 228
Schmidt, G. and Sigusch, V. (1972) ‘Changes in sexual behavior
among males and females between 1960-70’, Archives of Sexual
Behavior 2, 27-45 228
Schofield, M. (1968) The Sexual Behaviour of Young People (Har-
mondsworth: Penguin Books) 14, 61, 69
Schooley, M. (1936) ‘Personality resemblances among married couples’.
Journal of Abnormal and Social Psychology 31, 340-7 175
Schuster, E., and Elderton, E. M. (1906) ‘The inheritance of psychical
characters’, Biometrika 5, 460-9 175
Scott, J. P. and Fuller, J. L. (1965) Genetics and the Social Behavior
of the Dog (Chicago: University of Chicago Press) 192
Shepherd, M., Lader, M. and Rodnight, R. (1968) Clinical Psycho¬
pharmacology (London: English Universities Press) 219
Shope, D. F. (1966) ‘A comparison of selected college females on sexual
responsiveness and non-responsiveness’ (Pennsylvania State Univer¬
sity: unpublished Ph.D. thesis), quoted by Fisher (1973) 17
Sigurt, V., Schmidt, G., Rheinfeld, S., and Weidemaun-Sutor, I.
(1970) ‘Psycho-sexual stimulation: sex differences’, Journal of
Sexual Research 6, 10-24 69
Sigusch, V. and Schmidt, G. (1973) Jugendsexualitdt (Stuttgart: Enke)
228
Simon, P. (1972) Rapport sur le comportement sexuel des Francais
(Paris: Julliard et Charron) 28, 136

250
 REFERENCES and AUTHOR INDEX

Sorensen, R. C. (1973) Adolescent Sexuality in Contemporary America

(New York: Ward Publishing Co.) 228
Tennent, T. G. and others (1971) ‘Female arsonists’, British Journal of
Psychiatry 119, 497-502
Tennent, G., Loucas, K., Fenton, G. and Fenwick, P. (1974) ‘Male
admissions to Broadmoor Hospital’, British Journal of Psychiatry
125, 44-50 177
Terman, L. M. (1938) Psychological Factors in Marital Happiness
(New York: McGraw-Hill) 17
Terman, L. M. (1951) ‘Correlates of orgasm adequacy in a group of
556 wives’, Journal of Psychology 32, 115-72 17
Tharp, R. G. (1963) ‘Psychological patterning in marriage’, Psychological
Bulletin 60, 97-117 109
Thomas, D. R. (1975) ‘Conservatism and premarital sexual experience’,
British Journal of Social and Clinical Psychology 14, 195-6 175
Thomason, D. B. (1951) ‘Differential non-sexual and sexual behavior in
the marital adjustment of Penn State alumni’ (Pennsylvania State
College: unpublished Ph.D. thesis), quoted by Fisher (1973) 17
Thorne, F. C. (1966) ‘The Sex Inventory’, Journal of Clinical Psycho¬
logy Monograph Supplement no 21 30
Tisserand-Perrier, M. (1953) ‘Etude comparative de certains processes
de croissance chez les jumeaux’, Journal genetique humain 2, 82-98
197
Udry, J. (1967) ‘Personality match and interpersonal perception as
predictors of marriage’, Journal of Marriage and the Family 29, 722-5
109
Welch, B. L. (1937) ‘The significance of the difference between two
means when the population variances are unequal’, Biometrika 29,
350-62 140
Wells, B. W. P. (1969) ‘Personality characteristics of VD patients’,
British Journal of Social and Clinical Psychology 8, 246-52 69
Wherry, R. J. and Gaylord, R. H. (1944) ‘Factor patterns of test items
and tests as a function of the correlation coefficient; content difficulty,
and constant error factors’, Psychometrika 9, 237-45 87
White, O. 77
Wilkins, I. (1959) ‘Masculinization of the female fetus due to the use
of certain synthetic oral progestins during pregnancy’, Archives
d’Anatomie Microscopique et de Morphologie Experimentale 48,
313-30 215
Wilkins, I. (1960) ‘Masculinization of female fetus due to use of orally
given progestin’, Journal of the American Medical Association 172,
1028-32 215
Wilson, G. (1973) (ed.). The Psychology of Conservatism (London:
Academic Press) 76, 124, 130, 150, 173
Wilson, G. D. and Brazendale, A. H. (1973) ‘Social attitude correlates
of Eysenck’s personality dimensions’, Social Behavior and Person¬
ality 1, 115-18 173
Wilson, E. O. (1975) Sociobiology (Cambridge: Belknap Press) 226

251
 REFERENCES and AUTHOR INDEX

Winch, R. F. (1958) Mate Selection: A Study of Complementary Needs

(New York: Harper Bros) 109
Winokur, G., Guze, S. B. and Pfeiffer, E. (1958) ‘Developmental and
sexual factors in women: a comparison between control, neurotic
and psychotic groups’, American Journal of Psychiatry 115, 1097-100
17
Yom, B. L., Bradley, P. E., Wakefield, J. S., Kraft, I. A., Doughtie,
E. B. and Cox, J. S. (1976) ‘A common factor in the MMPI scales of
married couples’, British Journal of Social and Clinical Psychology (to
appear) 109
Young, W. C., Gay, R. W. and Phoenix, C. N. (1965) ‘Hormones and
sexual behavior , in J. Money (ed.), Sex Research: New Developments
(New York: Holt, Rinehart and Winston) 215
Zetterberg, H. L. (1969) Om Sexuallivet i Sverige (Stockholm: Statens
offentliga utredninger) 228
Zuckerman, M. (1973) ‘Scales for sex experience for males and
females’, Journal of Consulting and Clinical Psychology 41, 27-9
Zuckerman, M. (1974) ‘The sensation-seeking motive’, in B. Maher
(ed.), Progress in Experimental Personality Research vol. 7 (New
York: Academic Press) 15
Zuckerman, M., Bone, R. N., Neary, R., Mangelsdurff, P., and
Brustman, B. (1972) ‘What is the sensation seeker? Personality trait
and experience correlates of the sensation-seeking scales’. Journal
of Consulting and Clinical Psychology 39, 308-21 14
 Index

age dissimulation factor, 75-6

and sexual attitudes, 110-18, 125-9, dissimulation scale (see Lie scale)
224- 5 dominance-submission factor, 94
and sexual behaviour, 72 dual standard factor, 45
aggressive sex factor, 96-7 dysthymics, 50
ambiverts, 11
androgen, 77, 214, 215-19 ectomorphs, 197-8, 219
and high P behaviour, 229 ejaculatio praecox, 67
and masculine behaviour, 229 emphasis, 210
anti-pacificism (see pacificism) endomorphs, 197, 219
anti-religionism (see toughminded¬ English sexual behaviour and Euro¬
ness factor) pean norms, 136
assortative mating, 108-9, 132-3, environmental factors, 3, 9
193-5, 209, 224 environmental influences, 225
authoritarianism, 139-40 on a personality trait, 193
autonomic activation, 214 E.P.I. personality inventory, 13, 20,
74, 196
biological versus social determinants, estrogen, 216-17, 219, 229
225- 6 ' and low P behaviour, 229
Broadmoor patients, 177-91, 224 and feminine behaviour, 229
experimental methods, 22-8, 30, 31,
capitalism versus socialism factor, 36, 45-7
150, 156-9, 161-2 extraversion (E), 10, 30, 214
censorship, 234-40 extravert behaviour patterns, 13, 231
censorship factor, 42 prediction of, 12-13
circadian rhythms, 230-1 extraverts, 11, 222
competition model, 206-8 Eysenck Personality Questionnaire
conflict over sex factor (see neurotic (E.P.Q.), 19, 30, 105, 167, 198
sex factor) factor analysis of sexual attitudes,
conformity (see conservatism) 36-50
conservatism, 21, 123, 124, 140, 150,
223 female frigidity, 67, 68
and the high L scorer, 76, 123-4, Fisher’s assortative mating para¬
167, 223 meter, 194
conservatism factor, 156-60, 164 frequency of intercourse, 6-7
conservative values, 231
conventionality (see conservatism) genetic basis of sexual behaviour,
cortical arousal, 214 225
cultural factors, 29 genetic determination of sexual
attitudes, 3, 9
Darwin, Charles, 1, 6 genetic factors, 2, 9, 29
discrimination factor, 163 in sexual behaviour, 192-219

253
 INDEX

genetic influences on a personality low P scorers, behaviour patterns,

trait, 193 231-2
genotype-environment covariation,
194 male chauvinism, 234-5
genotype-environment interaction, male impotence, 66
194 increase in, 240
guilt factor, 43-4 marital status and sexual attitudes,
110-18, 129
habituation to sexual stimulus, 13 marriage, 129
Hanbury-Williams, Charles, 230 masculinity, 208-9
heritability, 196, 213-14 and libido, 213
of body and build, 196 and psychoticism, 214
of height, 196 masculinity-femininity scale, 104,
of libido, 225 156, 222
of personality traits, 211 Masefield, J., 238
of social attitudes, 196 mesomorphs, 197, 218-19
heterogam y, 109
higher-order factors neuroticism (N), 10, 15-16, 30, 214
sexual libido, 97, 99-101 and orgasm capacity, 17-18
sexual satisfaction, 97, 101-2 neurotic sex factor, 90-1, 183
high E scorers, 222 neurotic sexual behaviour, 16-18
high N scorers, 222-3 predictions of, 16
high P scorers, 222, 232 Neville, Richard, 238, 239
high scorers on the L scale, 223
occupation and sexual attitudes,
Holzinger’s formula, 196, 214
110-18, 125, 129-30
homogamy, 108-10, 133, 169, 175
orgasm, 67
homosexuality factor, 41, 181, 186-7
human diversity, necessity of, 2 pacifism, 156
Huxley, T. H., 5 pacifism factor, 163
hypothetico-deductive method, 9 panmixia, 109
‘pathological’ sexual behaviour, 59
idealism factor, 164-5 67-8
impersonal sex factor, 91-2, 183-4 Pavlovian conditioning, 12
individual differences in sexual be¬ pelvic shape and sexual behaviour,
haviour, 5-29 218
inhibition factor, 45 penis plethysmograph, 13
introvert patterns of behaviour, 231 permissiveness, 130-1, 231-41
introverts, 11 so-called, 233, 241
inventory of attitudes to sex, 32-5, permissiveness factor, 88-9, 182
79-87, 177, 198 personality and sexual attitudes,
inventory of social attitudes, 150, 50-61
151-6, 198 high E scorers, 54-6
ischial tuberosities of the pelvis, high N scorers, 56-9
217-18 high P scorers, 51-3, 56
the E factor, 59
Jensen, A. R., 26 personality factors
and enjoyment of sex, 191
libertine views, 2, 59, 231 and homogamy, 110
libido, 208-9, 221 and permissiveness, 132
and age, 221 and sexual attitude factors, 45-7;
heritability in, 211, 213 in twins, 198-200
libido scale, 106-8 and sexual behaviour, 72-6
Lie scale (L), 20-2, 74-6 and sexual disorders, 66-7

254
 INDEX

personality factors—cont. sex education, 239-40

and social attitudes in twins, 198- sex hormones
200 and behaviour, 214-15, 229
and the masculinity-femininity and pelvic shape, 217-18
scale, 105-6 and personality, 215-17
and voting behaviour, 171-3 sex (male-female) differences, 28-9
in twins, 210 and coloured sex partner, 37
personality questionnaire, 145-8 and hormonal secretions, 226, 229
key, 149 and libido, 50, 103, 213, 221, 226-8
personality scale scores of Broad¬ and marriage, 36
moor patients, 181 and permissiveness, 131-2
personality theory, 9 and popular wisdom, 104
and sexual behaviour, 10-21 and satisfaction, 103, 221
‘perverted’ sex attitudes, 138-9 and social attitude factors, 159-60
physical sex factor, 96, 186 and the P scale, 64-5
ponderal index, 196, 198 and toughmindedness, 162, 210
population parameters, 5, 79 cultural influences, 213
pornographic films, 235 determination by society, 226,
pornography, 234-9 228-9
American Presidential Commission environmental pressures, 208
Report, 235-6 genetic factors, 192, 209, 213
and artistic integrity, 237-8 in enjoyment of sexual behaviour,
and direct sexual activity, 138 134-7, 224
effect on sexual behaviour, 235-9 in personality, 20
pornography factor, 92-3, 184 in sexual attitudes, 3, 61-5, 103-4
premarital sex factor, 38-9 106
progestin, 214-15 in sexual behaviour, 14, 135
pro-government factor, 165 in sexual excitement, 97-8
Promax, 31, 46, 75, 87, 97, 156, 161 in social attitudes, 151, 156
promiscuity factor, 42-3 on masculinity-femininity scale,
prudishness factor, 40, 93-4 222
psychoanalytic views, 8 physiological, 226
psychotic behaviour patterns, 231-2 sex offenders, 65
prediction of, 19-20 sexual aggressiveness factor, 186
psychoticism (P), 10, 18-19, 30 sexual attitude factors
Public Opinion Inventory {see inven¬ analysis, 46-50
tory of social attitudes) and libido factor, 106-9
puritan views, 2, 59, 231 and satisfaction factor, 106-9
sexual attitude questionnaire {see in¬
radicalism, 209-10
ventory of attitudes to sex)
heritability of, 209
sexual attitudes
refusal rate, 6, 7
and personality, 110-27, 181-91
Reiss Courtship Inventory {see Reiss
and social consequences, 220-41
Permissiveness Scale)
and voting behaviour, 174-5
Reiss Permissiveness Scale, 131-2
in adults, 78-149
repression factor, 39-40
in psychiatric patients, 177-91
role-playing, 3, 210
in students, 30-77
sampling of Broadmoor patients, 177-80;
practice, 30, 68, 78-9, 151, 174, 177, factor analysis, 181-7
196, 198, 219 of high E scorers, 120-1
theory, 5, 7, 22-8 of high L scorers, 123-4
satisfaction factor, 47, 89-90, 182 of high N scorers, 121-3
satisfaction scale, 106-8 of high P scorers, 118-20

255
 INDEX

sexual behaviour and personality, 190

and high P scorers, 139 sexual shyness factor, 93
and social attitudes, 174-5 sibling effects, 206
disliked, 224 social attitudes
in adults, 78-149 and age, 167, 169
in psychiatric patients, 177-91 and assortative mating, 168-9
in students, 30-77 and personality factors, 165-7
sexual behaviour factor analysis, 70-1 and sexual attitudes, 167-8
sexual behaviour questionnaire, 14- and sexual behaviour, 150-75
15, 70 and social class, 166-7, 169
sexual behaviour scale, 177, 191 and voting behaviour, 169-71, 173-4
and libido, 138 factor analysis, 156-65
and satisfaction, 138 of high E scorers, 167
and sexual attitude scale, 134-5 of high L scorers, 167
sexual conflict in neurotic patients, 51 of high N scorers, 167
sexual curiosity factor, 38 of high P scorers, 167
sexual disgust factor, 94-5 of middle-class respondents, 167
sexual excitement factor, 37, 95, 185 social class and sexual attitudes, 110-
sexual experimentation factor, 41 18, 125-31, 225
sexual factor analysis, 87-98
sexual hostility factor, 43 tenderminded attitudes, 240
sexual libido factor, 50, 187-9 tendermindedness, 124, 130, 241
and personality, 190 testosterone, 77
sexual nervousness factor, 37-8, 185 tomboys, 215
sexual pathology and neuroticism, 67, toughminded attitudes, 240
69 toughmindedness, 130, 139-40, 150,
sexual pathology factor, 50 223, 241
‘sexual roles’, 3, 210, 226 heritability of, 210
sexual satisfaction, 206-8, 221 toughmindedness factor, 156-9, 160-3
and age, 221-2
and competition, 225 veridical reporting, 28, 31, 68-70, 136
environmental factors, 213 of Broadmoor patients, 180
genetic factors, 213
heritability of, 213 Wilson-Paterson Conservatism Scale,
sexual satisfaction factor, 36, 187, 209
189-90
4
 M&°, fl 7 2001
m«- fl 5 200|

DATE DUE
_ DATE de retour

* 5 2003-

oarr MCLEAN ----

38-296
 BF 692 .E95 1976
Eysenck, H. J. (Hans Jurg
010101 000
■||8[|llpfTfonal'|ty / H 3 Ev
.If .

0 1163 0085339
TRENT UNIVERSITY

BF692 .E95 1976

Jurgen.
Eysenck, Hans
sonality
Sex and per

312337
 :

■ . . .
; • ■.. ■'.

* -
- ■ :: .

- v
:=■-'■-■

■
:■;

K
:■ v’:

..;«?>• . .
.
-
;
- - •„ • - . . /y.'.'.i ’ •»:**4 *4+
. ,' ■-- '„ «*
:.®AS

■: - V; r.-i -4: ■-: - - -4 ' : ,::•■•:>'««,>*:•,•:••: ;•: ".. :::•-. •: ‘a

' ' ■