THE MAUDSLEY PERSONALITY INVENTORY

ARTHUR R. JENSEN, Ph.D.1

Znstitute of Psychiatry, Maudsley Hospital, University af London

Most of the research coming from tbe Psychology Department of the

Instittne of Psychiatry in recent years has been oriented in terms of
Eysenck’s conception of dimensions of personality (9). H. J. Eysenck is
Professor of Psychology and Director of Research in the Institute of
Psychiatry. His philosophy of research and the results of research con-
ducted in his department have been summarized in three monographs
(9, 10, 13). Eysenck’s prlmary contentIon has been that the taxonomical
problems of description, classification, and meti-urement must be worked
out in the personality field before worth while attempts can be made to
explain the underlying causes of differences in personality.
In reviewing practically the entire literature of objective personality
reser. rh before 1953 based on rat&s, questionnaires, objective behavior
tests, analysis of physique, physiological measures, and analysis of in-
terests and attitudes, Eysenck 2 found considerable evidence for at least
three pervasive and relatively independent “dimensions” in the personality
domain. He has identified these dimensions as Introversion-Extraversion,
Neuroticism, and Psychoticism. Eysenck’s own factor analytic studies
(9, 10) have further substantiated this dimensional hypothesis.
Recently Eysenck and his co-workers have been experimentally testing
hypotheses concerned with the “dynamics” or underlying cause of dif-
ferences on the Introversion-5xtraversion (I-E) dimension. Much of this
work has been reported in Eysenck’s latest monograph (13). It has been
necessary in this work, to have convenient criterion measures of the per-
fionality dimensions under investigation. For this purpose the Maudsley
Personality Inventory (MPI) has been developed. It is intended to measure
on the verbal level two dimensions of personality: Introversion-Extraver-
sion and Neuroticism. This inventory is being used extensively in the
Maudsley research and already many n:ferences to it have appeared in the
1 Research Fellow of the Natimal Institute oi Mental Health, United States Public
Health Seruk
a The Structure of Human Pmomzlity. London: Mmthuen, 1953.
314
 THE MAUDSLEY PERSONALITY INVENTORY 315
literalture. Its significance is enhanced by its correlations with a number
of e:uperimcntal and psychiatric variables. Thus it seerus worthw
this stage to present for more widespread use in personality research
instrument which has undergone elaborate development and has already
proved useful in research. Most of the existing normative data are also
presermd.

DEVJZLOPMEIW OF THE MPI

The development of the MPI has been described in great detail by

Eyscnck (12). The E (extraversion) and N (neuroticism) scales o:
MFI were derived from rather elaborate procedures involving item analysis
and factor analysis of other personality inventories, principally the Guilfcrd
inventory of factors STDCR and the Maudsley Medical Questio
The two scales, E and N, have high “construct validity,” that is, the i
making up the scales are highly correlated with the factor they are
to measure and they have insignificant correlations with other factors.
The items have been selected so as to minimize the correlati
the E and N sc:sles. The two factors are thus represented as
i.e., uncorrelated with one another.

DHCRIPTIONOF THE MPI

The complete MFI is given below. It consists of 24 E-scale items, 24
N-scale items, 20 Lie-scale items, and 12 “buffer” items which help in
concealing the nature of the questionnaire from the subject. The Lie sc
was intended to detect subjects who tend to present the
favorable light to such an extent as to make the validity o
questionable. A record may be regarded as definitely suspect in ‘his res
if more than 10 of the Lie scale items are answered in the: keyed
Scoring. Two points are given to the designated scale for
responses, and one point to the designated scale for the “‘7”. Thus
possible range of scores on the E and N scales is from 0 to 48.
Short Form, A short form of the MFI was prepared by Eysenck (14)
for use in market research, short interviews, and similar situations in
which there is limited time for testing. Eysenck (1
method of selecting items for the short MPI. There
each scale. These have been denoted by asterisks on
correlations (for both the E and N scales) between the short MPI
the total MFI have been found to be practically as high as the split-
reliabihty coefficient of the total MPI.
316 A. R. JENSEN

MMJDsUY~INva~Y
IRSTR~~XXOM: Please answer each question by putt& a circle round the “Yea” or
the “No” following the question; if you simply cannot make Up your mind, encircle
the rr?t’l_ Work quickly and do not ponder too long about the exact shade of meaning
of each question. There are no right or wrong answer%, aad IU) t&k questions.
Rcnrcmbcrto answer e& question.
Keyed Response
E 1. Ateyouindinedtolimityouracquaintancestoaselectfew?. . Yes ? No
E 2.*DoyouFreferactiontopl~foraction?. . , . . . . . . Yes ? Ng
E 3. Do you nearly always have a “ready aaswer” for remarks dire&A
at you? . . . . . . . . . . . . . . . . . . . _ . . . . Yea ? No
N 4. Are your day- frequently about things that con never come
trllel . . . . . . . . . . . . . . . . . . . . . . . . . Yes ? No
L 5. As a child, did you always da as you were told, immediately and
w&houtgrumbling? . . . . . . . . . . . . . . . . . . . . Yea ? No
E 4.*AreyouincDinedtobeauickaadsureinyouractions?. . _ . Yes ? No
7. Do~oubavedifEcultyinmakingnewfri&Is? , . . . . . . . Yes 4 No
L 8. Do you sometimes put off until tomorrow what you ought to do
today?..........................Yes? No
E 9. Am you inclined to take your work casually, that is, as a matter
ofcour§e? . . . . . . . . . . . . . . . . . . . . . . . . Yrs ? No
N 10. Do you often feel disgnmtkxi? . . , . . . . . . . . . . Yer 7 NO
N 11. A-eyouinclinedtopunderoveryourpast?. . . . . . . . . . Yta ? No
L 12. If you say you will do something do you always keep your promise,
~~~ma#er~~inconvenientitraiShtbetodoso?. . _ . . . . Yea 7 No.
E 13. Doyouliketomixsaciallywitbpeople?. . . . . . . . . . . Yea ? No
E 14. k~youiaclinedtaheshyinthe~ofthenppositc~? Yed 7 No
L 15. %yousometimar@aoss?. . . . . . . . . . . . . . . . Yes 7 No
N 16. Doyouoftenexperienceperiodsoflanelineas? . . . . . . . . ‘Iles ? No
N 17. Areyoutoucbyonvarioussubjects?. . , . . . . . . . . . . Yes ? No
N 18. Do you often find that you have made up your mind too late? Yea 4 No
L 19. Are you compleMy free from pre~uiice of any kind?. , . . . . Yea 7 No
E 20. Areyou&clinadtobe over@mAerltious? . . . . . . . . . . Ye+l ? No
21. Do you often ‘have +&atime of your life” at sccial affaiia? . . Yea ? No
22. Doyouevercbangefromhappinea5tosadness,orviaversa.
witboutgoodreason? . . . . . . . . . ~ . . . . . . . . . Yes ? No
E 23. Doyouliketoplaypranksuponotbers?. . . . . . . . . . Yea 2 No
L 24. Doyouelaughatadutyjoke?. . . . . . . . . . . Yea 7 No
N 25.#Does your mind often wander +ile you are trying to umcentrate? Yea ? No
N 26. Would you rate yourself as a teme or %igb-strung” individual? Yea ? No
N 27. AftcracMcal moment is over, do you usually think of sometbinp
youabouldhavedontbutfailedFQdo? . . . . . . . . . . . Yea ? No
L 28. Wou!dlyoumucbratberwin,tbanlose,agame? . . . . . . . Yea 1 No
29. DoyauLdit~,araiule,tomalrenew~~~?. . . Yw ? No
30. DO YOU WCS &WC a qucq feeling t&at you arc Mt your old self? Yes ? No
---
1 Tbcen&cledanawersbavebcenset.intypeboldface.
E=Ext.ravenGon. N=Naur&i&u L=Lie. * Short Form.
 THE MAUDSLEY PERSONALITY INVENTORY 317

Keyed R
W 3 1I Do you ever iti?&? your work as if it were a matter of life or death7 ?
;V 32.*&e you frequently “lost in thought” even when supposed to be
t,.tig part Bfka conversation? . . . . . . . , . . . . . . ? No
L 33. Do you always feel genuinely pleased when a bitter enemy achieves
fi merited success?. . . . . . . . . . . . . . . . . . . . Ya ? NO
34. Do you derive more real satisfaction from social activities than
from anything else? . . . . . . . . . . . . . . . . . . _ Yea ? No
N 35. Do ideas run through your bead so that you cannot sleep? . . . ? No
L 36. Do you sometimes boast a little? . . . . . . , . . . . . . . TN0
E 37. Can you usually let yourself go and have an hilariously good time
atagayparty7.. . . . . . _ _ . . . . . . . . . . . . . Yea ? No
N 38. Doyoubketoindulgeinareverie(daydreaming)? . . . . . . Yer ? No
N 39. Have you often felt Listless and tired for no good reason?. . . , Yea ? No
L 40. Are 011your habits good and desirable ones? . . . . . . . . . Yea ? No
E 41. Are you inclined to keep quiet wben out in a social group? . . . Yes ? No
N 42.*Are you sometimes bubbling over with energy and sometimes very
sluggish?. _ . . . . . . . . . . . . . . . . . . . . . . . Yes 9 No
L 43. Do you always answer a personal letter as soon as you can after
youbavereadit?, . . . . . . . . . . . . . . . . . . . Yes ? No
E 44. Would you rate yourself as a talkative individual?. . . . . . . Yes ? No
L 45. Do you occasionally have thoughts and ideas that you would not
likeotherpcople to know about? . . . . . . . . . . . . . Yes ? No
E 46.*Would you be very unhappy if you were prevented from making
numerous social contacts? . . . . . . . . , . . . . . . . . Yes ? No
E 47.*Are you happiest when you get involved in some project that calls
for rapid action? . _ . . . . . . . . . . . . . . . . . . Yes ? NO
N 4% Do you spend much time in thinking over good times you have
hadinthepast?. . . . , . . _. . . . . . . . . . . . ? NO
L 49. Do you sometimes talk about things you know nothing about? YeS ? No
N ii@- Have you ever been bothered by having a useless thought come
into your mind repeatedly?. . . . . . . . . . . . . . . . . No
E 91. Do other people regard you as a lively individual?. . . . . . . YeJ NO
L 52. Doyousometimesgossip?. . . . . . . . . . . . . . . . . Yt.3 No
53. Do you usually keep in fairly uniform spirits?. . . . . . . . . YeS No
N 54. Areyourfeelingsratheressilyhurt?. . . . . . . . . . . . . YCS NO
L 55. Attimes,haveyouevertoldalie?. . . . . . . . . . . . . . YeS NO
E 56. Do you generally prefer to take the lead in group activities7 YtXl NO
E S7. Would you rate yourself as a happy-go-lucky individual?. . . . YCS No
L 58. Have you money worries at times? . . . . . . . . . . . . . YCS No
N 59. Do you have periods of such great restlessness that you cannot
sitlonginachair7 . . . _. . . . . . . . . . . . . . , . NO

60. Areyouusuallya”‘goodmixerT’. . . . . . . . . . . . . . No
E 61, l Would you rate yourself as a lively individual? . . . . . . . . No
L 62. Have youeverbeen late for an appointment or work7 . . . . . No
N 63. Do you ever feel “just miserable” for no good mason at all? . . NO
N 64. Areyouoftentroubledwithfeelingsofgnilt?. . . . . . . . . No
N 65.+A_myouinclin&tobemoody?. . . . . , . . . . . . . . . NO
318 A. R. JENSEN
Keyed Rqxmse
E 66. Doyouliketahavemanysocialengagements?. . . . . . . . Yes ? No
L 67. ~inawhii::,doyouloseyourtanperandgetangry?. . . . Yes ? No
N 68.*Do you sometmes feel happy, sometime8 depressed. witbout any
appaxntmn? . . . . . . . . . . . . , . . . _ . . . . Yes ? No
E 69. Is it dif&ult tci %seyou&f’* evenat alively party? . . , . . Yes 7 No
70. Areyouordia~itrilyacarefreeindividual?. . . . . . _ _ _ . . Yes ? No
N 71.sDo you have requent ups =d downs in mood, cipherwith or
without appan:at cause? . . . . . . . . . . . . . . . . . . Yea 1 No
L 72. W&d you aI5lays declare evcryfMzg at the Customs, even if you
kncwthatyoucouldneverbefoundout? . . . . , . . . . . Yea 7 No
E 73. Do you like w xk that nauires considezable attentkm to details? Ye-s 7 No
74. Are there time!; when you seek to be alone and you cannot bear
the corn-y alanyone? . . . . . . . . . . . . . . . , . . Yea ? No
E 75. Are you inclia~.:dto keep in the background on social occasions? Yes 1 No
N 76. Have you oftetl lost sleepover your worries? . . . . . . . . . Yes 1 No
L 77. Of all the peo:i:lleyou know are there 8ome whom you definitely
do not like? . . . . . . . . . . . . . . . . . . . . . . Yes ? No
78. Do you usual]:,?feel disappointments so keenly that you cannot
gettbemoutol”yourmind? . . . . . . . . . . . . . , . . Yes i! No
E 79.*Do you usually take the Stiative in making new friends? . . . Yea ? No
80. Do you enjoy participating in a show@ of “Rah Rab” e-u-
thus%zsm? . . . . . . . . . . . . . . . . . . . . . . . ” Yea 7 No
NORMATIVE DATA
Table .I shows the means, standard deviations, reliability, and car-
relations between the N and E scales in various samples. The Table is
quite self-explanat8~ry. The age of the subjects has not been given since
even in large sam >les in which the ages ranged from 17 to 6.5, no cor-
relatlon has been found between age and the MPI scales, Correlations
with sex have been negligible in all studies, though there is a slight tendency
for women to Scot e on the average about one point higher than men on
both the iE and PI scales. The reliabilities of the scales are high for a
personality inven~r~ and compare favorably with the reliability of
cognitive ‘tests such as the Stanford-Hinet and Wechsler intelligence tests.
Table II gives similar data for the Short Form of the MPI.
COLLATIONS w~xx 03-iz~ ~ASURES
Table III shows the correlations of the MPI scales with a number of
other psychological measures. A brief description is here given of each of
these measures. For further details the reader is referred to the original
articles.
1. The Short MPI has been described above.
2, The Heron Two-Part Personality Measure (15) consists of an
Introversion (or “sociability”) scale and a Neuroticism (or “emotional
 THE MAUDSLEY PERSOhJALITY INVENTORY

T-LB I
Mean, Standard Deviation, Reliability and Intercorrelation of tbe Extraversion
Neuroticism Scales of the MPI for Various Groups

E-Scale N&ale
SampIe N Mean SB Mean SD

1. Normal adult males (12) . 200 24.62 10.04 17.81 11.32 -.15+
2. Normal adult females (12) 200 25.17 9.33 19.45 11.02 -.04
3. Total of 1 & 2 (12) . . . 409 24.89 9.67 18.63 11.19 -.09
4. English university stu-
uznts. Male (11) . . . 50 28.86 8.36 19.04 11.24 .I2
5. English university stu-
dents. Male (18) . . . 213 25.26 8.85 23.23 11.27 -.07
6. English university stu-
dents. Mixed (20) . . . 64 25.16 10.22 26.78 9.23
7. Student mrrses. Female(2O) 22 23.82 9.71 30.64 9.22
8. Totalof6&7(20) . . 86 24.81 10.11 27.77 9.42 -.30 * *
9. Polytechnic & art school
students. Mixed (7) . . 68 24.57 19.04 27.06 11.56 -.OS
10. American university stu-
dents. Male (4) . . . . 714 28.4) 8.06 20.19 10.71
I 1. P~nrerican university stu-
dents. Female (4) . . . 350 29.41 8.37 21.63 10.45
12. American university stu-
dents. Mixed (4) . . . 145 27.77 7.60 21.57 9.75 -.20+* .741 .W
13. Iudustrial apprentices.
Male(18) . . . . . . 100 29.34 9.00 21.20 10.26 -.OQ
14. Industrial apprentices.
Male (16) . . . . . 76 29.18 8.19 21.2: i0.41 -.14
15. Neurotic patients. Male . 83 19.W 10.13 32.98 10.78
16. Neurotic patients. Female 65 18.67 9.21 34.75 11.83
17. Total of 15 t 16 . . . 148 18.91 9.86 33.75 11.29 -.30”*
18. Dysthymicss (17) . . . . 25 21.OC 11.96 36.80 10.48
19. H:ysterics & psychopaths 27 25.22 9.96 28.82 12.76
(17). . . . . . ‘ . .
20. R~ecidivists. Male (1). . . 72 24.76 10.08 32.18 10.38 -.32++

l p < .05.
**p < .Ol.
r Corrected split-half reliability.
a Kuder-Richardson “Formula 20”.
s *‘~Dysthymic” is Eyscnck’s term for neuroses of the anxiety, depression. o vc-
compulsive, and phobic types.
320 A. IR.JENSEN

Meart, SGmkd Deviation, Reliability, and Intezco~on of the Extrav~oo and

Neuroticism scales of the Short Form of the MPI.

E-scale N-SW.@ Reliability

Sample N Mean SD h&au SD r, E N
-
1. Quota sample of English
urban and nmi dwellers
(14) . . . _ . . . , _ 1600 7.36 2.97 6.15 3.33 -.05 .fl .79
2 Neurotic patients. Male . 83 !i.41 3.04 8.. 4 3.19
3. Neurotic patients. Female 65 6.03 2.80 9.00 3.75
4.Totalof2&3 . . . . . 148 !;A8 2.95 8.86 3.45 -.44”
** p< -01.

maladjustment”) scale. The scales are quite short, together requiring only
about 20 minutes of the subject’s time. The reliabilities are high (-74 and
.8f respectively) and the Neuroticism scale has Bern shown to differentiate
normals from hospitalized neurotics.
:3. Cattell’s CPF (Contact Personality Factor) scale (6) is made up
largely of items from five scales of Cattell’s 16 Personality Factor
Questionnaire. The CPF is essentially a measure ol’ social extraversion and
was designed for employment selection purposes as a measure of the
amount of social contact a person needs in his work in or&r to find it
personally satisfying. Cattell states also that the test measures “general
adjustment to social demands and ability to adapt to people” (6). Suc-
cessful salesmen, for example, would be expected to obtain high scores,
while bookkeepers or research scientists would obtain comparatively low
scores. There are two equivalent forms of the CPF, Forms A and B, which
are reported by Cattell (6) as beiig correlated .86 in a sample of 125
subjects. In a sample of 134 neurotic plktients tested by the writer, the
correlations between Forms A and B was .69.
4. The Minnesota TSEm (Thinking, !SociaI, E.uptional) Introversion-
Extrave~rsicn Scales (8) were based on Jung’s &lea that introversion-
extraversion manifests itself in thinking, in social or interpersonal behavior,
and in feeling or emotional behavior, and that these three aspects of in-
troversiion-extraversion are not necessarily correlated. The itms of the
TSEr scales were selected so as to represent these three aspects of I-E
and the three scales were so constructed as to be not significantly cor-
related with one another.
5. The Taylor MA8 (Wanifer f Anxiety Scale) (19) was devised as a
measure of the kind of neurotic state diagnosed by psychiatrists as anxiety
and ‘has been shown to be correlated with psychiatrists* ratings of anxiety.
 THE MAUDSLEV PEBSONALITY INVENTORY

The MAS is composed of 42 items selected from the Minnesota Multiph~i~

Personality Inventory.
6. Biis’ Index of Adjustment and Values (5) is an adject
consisting of 49 adjectives describing personal characteristics.
rates the adjectives according to the degree to which they desc
and also according to what he would regard as an ideal self, i.e., the way
he would like to be. The Index yields two scores - and acceptance of
self (AS) score and a score reflecting the discrepancy (D) between
ceived and ideal selves. The two scores (AS and Dj are highly intercor-
related (negatively), suggesting that they both measure the same tem-
perament variable, which is probably one of general self esteem.
7. The Copperative Vocabulary test was used as a measure of verbal
intelligenee. The non-significant correlations with the MPI scales are
consistent with all other studies in which the relationship ~tw~n in-
telligence and 1-E and N has been determined. The correlation is in
all cases negligible or non-significant. If there is any suggestion of a trend
in these correlations it is that verbal intelligence is negatively correlated
to a slight degree with both extraversion and neuroticism.
8. College Achievement was based on course gades and is not
significantly correlated with the MPI scales.
Tmtx III
Correlations of the MPI Extraversion and Neuroticism Scales With Other kales
-
correiatio
Scale Sample Iv E-scale H-
--- --
la. Short MPI, N-scale . . . Neurotics . . . . . . . 134 -.27 * * .86* *
b. Short MPI, E-scale. . . . Neurotics _ . _. _ 134 .87** -.29 * l
2s Heron, Neuroticism , . . Neurotics . . . . . . . 134 -.28 * @ .64**
b. Heron, Introversion . . . Neurotics . . . . . . . 134 -.8O+ * .28”
3a. Cattell, CPF. Fortu A . . Neurotics . . . . . . . 134 .65** -.34 * *
b. Cattell, CPF, Fotsn B . . Neurotics . . . . . . . 134 .67*+ -.53”C
c. Cattell, CPF, Form A . Industrial apprentices (16) 76 .38*’
4a. MinnesotaT-scale . . . English Uriv. students (20) 87 -.OS
b. MinnesotaS-scale . . . . English Univ. students (20) a7 .81 l l
c. Minnesota Em-scale . . . Englis!l Univ. students (20) 87 .21*
5. Taylor MAS . . . . . . American univ. studeuts (3) 254 -.3s l *
G. Bills’ Index of Adjustment : American univ. studeuts (2)
a. Self-acceptancescore . . . 96 .27+* -.2S
b. Discrepancy score . . . 96 -.29** .30**
7. Verbal Intelligence . . . . American univ. students (4) 254 -.I2 -.os
g. College Achievement . . . American univ. students (4) 189 -.I2 -.11
*, p < .05.
l + p< .Ol.
322 A. R. JENSE,N

Dr~ssi~
While the present paper is not intended as a critique of the MPI or of
the fac:tor analytic theory of personality underlying its development and
its use in Eysenck’s research program, a few comments are in order
concerning the data presented in Tables I, II, and HI.
(a) It can be seen from the correlations presented in Tables I and II
that the E and N scales are not orthogonal (i.e., independent or uncor-
related) in all samples. However, even where E and N are significantly
correlated, they have at most only about 10 per cent of their variance in
common. A significant negative correlation obtains between E and N
only in those samples which i2 some way represent some highly selected
(and therefore biased) element of the general population, and these biased
samples are generally higher on Neuroticism than the general population.
It appears that while in the general population there is only a slight
negative correlation, if there is any correlation at all, in groups composed
of more neurotic subjects there is a greater tendency to perceive the self
as having more introverted characteristics. This effect may be partly an
artifact due to a factor ,such as differences in the “social desirability” of
the introverted and extraverted items. Subjects who have less self esteem
or are less concerned with making a good impression may score higher
in introversion (as well as in neuroticism) if more of the introverted than
extraverted items have socially less desirable or less self-flattering con-
notations. Thus, more neurotic subjects, such as hospitalized neurotics
and prisoners, whose seJf esteem is at a loa ebb and who have little in-
centive to create a “good” impression in an institutional setting, would
be less apt to favor the items that create the most favorable self-picture.
The E-scale items should be studied for this “social desirability” factor.
If it exists, its eliminaticn would be a distinct improvement of the scales.
Another property of the MPI that wi’qants critical examination is the
fact that all of the N-scaie items are keyed “Yes”, while only two-thirds
of the E-scale items are keyed ,,Yes”. If there is a generalized tendency
to agree with statements in a questionnaire regardless of their specific
content,, there would thus be a built-in negative correlation between E
and N. A person answering “Yes” to all the items in the MPI would obtain
the highest possible N score but would be only intermediate on E.
(b) It will be noted in Table I that samples No. 6, 7, and 9 have ap-
preciably higher N scores than the normal population samples or other
student samples. The reason is probably that subjects in these particular
samples were tested on a volunteer basis, which was not the case with
t,her samples. Many of the subjects in samples 6, 7, and 9 heard that a
 THE MAULXLEYPERSONALITYINVENTORY 323
psyChOlOgiCalinvestigation was beii conducted and they volunteered to
participate. It is not unlikely that such volunteer subjects might have
greater concerns about their own psychological problems and woul
constitute a somewhat more neurotic sample than would be obtained if
the sample were randomly selected from the total student population.
(c) Some of the data presented in Table I are not in accord with
aspects of Eysenck’s theory and no explanation for this fact seems m
reasonable than the suggestion that the the80ry may not be correct in
particulars. Eysenck (13) has stated that during periods of emotional in-
stability persons towards the extraverted end of the I-E continuum
develop symptoms of the hysterical type, while persons toward the extra-
verted end develop “dysthymic” disorders (anxiety, depression, and
obsessive-compulsive). In short, hysterics (and psychopaths) are said to
be extraverted neurotics, and dysthymics are said to be introverted
neurotics. In Table I, however, it is apparent that the dysthymics and
hysteric-psychopath groups (Samples No. 18 and 19) do n
significan:ly on the E scale, as would be predicted from Eysenck
nor do they differ significantly from the normal groups on the
The dysthymics, however, score higher on the N scale, though there is
nothing in Eysenck’s theory that would predict this finding. The c:rite:ria
and method of diagnosing these patients as dysthymic, hysteric, or
psychopath, as well as a discussion of the negative findings with res
to Eysenck’s theory, are to be found in a paper by Sigal et al. (17).
Another deduction from Eysenck’s theory that is not borne out in these
data is that prisoners, especially recidivists, should be more extraverted
than are more responsible, law-abiding persons. According to the theory,
because of a constitutionally greater tendency to develop cortical inhibition,
extraverts are less readily conditioned and hence less readily socialized.
They are therefore less apt to be constrained by the rules of society and in
general to show more psychopathic types of behavior. Thus one would
predict a higher degree of extraversion among recidivists than in the
general population. As can be seen from Sample No. 20 (Table I), however,
recidivists score no higher on the E scale than do the normal adult males
(Sample No. 1) and ar.: not as high on E as industrial apprentices and
American university students. But the recidivists are as high on neuot-
icism as the hospitalized neurotics. Furthermore, the same degree of
negative correlation between E and N is found in the recidivist group as
in the neurotic group.
(6) Some of the correlations of the MPI with other measures of extra-
version and neuroticism (Table 111) are based on a sample of neurotic
324 A. R. JENSEN

mt ‘ce assumed to represent the correlations that might

le of the normal population. Certain points are worth
e E scale correlates highly with other measures of
for the Minnesota scales of Think@ (T) and
m) extraversion. The fact that the Ml?1 E scale is not
signlfleantly correlated with the Minnesota T and Em scales but is cor-
.81 with the Minnesota social extraversion (S) scale indicates that
scale of l &e MPI is a measure of only one aspect or one type of
extraversion, viz., social extraversion. From a lo& at the items in the
this finding is not surprising. Most of the items concern forms of
&avior. It might also be pointed out that the MPI E scale coc-
relates as highly with the Cattell CPF measure of extraversion as the two
forms of the Cattell scale correlate between themselves in this sample.
Taylor’s Manifest Anxiety Scale (MAS) appears to be largely a measure
of the WI neuroticism factor, but it also has a slight correlation with
introversion.
It is interesting +o see that a’n Bills’ Index of Adjustment the extraverts
are more self satisfied and there is less discrepancy between the extravert’s
perceived self and his self ideal. In view of the IGW correlation between
Bills’ Ilndex and the N scale, it would seem that “Index of Adjustment”
is a misnomer. Whatever kind of maladjustment Bills’ Index measures it
is o3ly slightly related to Eysenck’s measure of neuroticism.

‘Ihe h&iwdslw FixsonaliQ rnvcntory (MPI), consisting

of two scales, oue measuring
intrbve&on (E scale) and the other mawwing neuroticism (N scale), has
been presented as well as

ion or sociability. the

of the brevity of the scalea, their high reliabiititu. their orthogonal&y in the
wnnal population, their bigb correlations with other measurea of these factors, their
blec4melationwithn on-penonality variables such as age, sex, aod intelligence,
Ehd their correlation with other

bencornfor reJearrh
purposea iMbeii perhaps the best queatiomlaire measure
of Wrwe&wxL~vti and neuroticism available at the pnwnt time.
 THE MAUDSLEY PERSONALITY INVENTORY

The writer is indebted to the foilowiq: The Statistics Dept. of the institute of
for emna of the statistical computations. Dr. J. Robinson, Director, and Mr. G
Psychologist, ?f tb.e Roffey Park Rehabilitation Center for their cooperati
the writer admhWer the MFI to the entire hospital population. Dr. A.
for obtaininS MPI data from subjects in a prison. Enter Treadwell for
scoring MPIs of university students.
Note : The MPI is now published by the University of London Press.

1. Bartholomew, A., Personal corommication, January 1958.

2. Bendi& A. W., and Hoffman, Jenny L., Bills’ Index of Adjustment and the Ma
Personality Inventory. Psychad. Reports, 1957, 3, 507.
3. -, Extravcrsion, neuroticism, and manifest anxiety. J. amu&. Psych& 1957.
21, 398.
4. -, Personal commuuication, December 1957.
5. Bills, R. E.. Vance, E L., and McLean, 0. S., An index of a
J. consult. Psych& 1951,. IS. 257-261.
6. Cattell, R. B., King, J. E., and Schuettler. A. R., Contact Perso
nalre. Champaign, Illinois: Institute for Pcmonali
1954.
7. Das, J. P., An expcrimeotal study of the relation between hypnosis, condi
and reactive inhibition. Ph. R. thesis, 1957, Univ. of London.
8. Evans, C., and McConnell, T. R., A new measure of extraversion-iutro
J. Psychol., 19.!1, 12, 111-124.
9. Eysenck, H. J., Dimensions of personality. London: Routlcdgc and
10. -, The scicntlfic study of personality. London: Routledge and
1952
11. -, Reminiscence, drive, and personality theory. J. &norm. SW.
1956. 53, 328-333.
12. - -, The questionnaire measurement of neuroticism and exmmrsi
di Psicologia, 1956, 54, 113-140.
13. -, The dynamics of anxiety and hysrerkx London: Routl
1957.
14. ----, A short questionnaire for the measurement oftwo
ty. J. appf. Psychof.. 1958. To appear.
15. Heron. A., A two-part personality measure for ~.se as a research criterion. &a. 1.
Psychof., 1956, 41, 243-251.
16. Holland, H., Pemonal communication, January 1958.
17. Sigal, J. J., Star, K. H., and Franks, C. M., Hysteria and dys
groups in the study of introversion-extraversion. J. a!
1958. To appear.
18. Star, K. H., An experimental study of ‘reactive inhibition’ and its relation
personality traits. D. Pb. thesis, 1957, Univ. of London.
19. Taylo,?, Janet A., A pcrsodty scale of manifestanxiety. J. abrtwn. See.
1953, 48, 285-m.
a. Treadwell, Bmer. Pcrwnal comm~tio& Decembar 1958.