See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/259635836

The Socioemotional Well-Being Index (SEWBI): Theoretical Framework and

Empirical Operationalisation

Article  in  Social Indicators Research · November 2013

DOI: 10.1007/s11205-013-0528-z

CITATIONS READS

12 2,619

1 author:

Eduardo Bericat
Universidad de Sevilla
51 PUBLICATIONS   1,046 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

THE QUALITY OF EUROPEAN SOCIETIES: COMPOSITE INDICATORS FOR ITS MEASUREMENT AND MONITORING. View project

All content following this page was uploaded by Eduardo Bericat on 10 April 2018.

The user has requested enhancement of the downloaded file.

Soc Indic Res (2014) 119:599–626
DOI 10.1007/s11205-013-0528-z

The Socioemotional Well-Being Index (SEWBI):

Theoretical Framework and Empirical
Operationalisation

Eduardo Bericat

Accepted: 25 November 2013 / Published online: 15 December 2013

! Springer Science+Business Media Dordrecht 2013

Abstract This article presents the design, process of construction, content and validation
of the Socioemotional Well-Being Index. This index is a composite indicator of subjective
well-being, and has been designed with the aim of providing a measurement device for the
sociological analysis of the subjective components of quality of life and social quality.
Two spheres of knowledge have been combined in its construction: research in social
indicators, the recent development of which has been oriented toward the elaboration of
composite indicators, and the theoretical content developed in recent decades by the
sociology of emotions. As a composite indicator, the index presented in this article offers a
hierarchical and multidimensional alternative to the univariate scales measuring happiness
and satisfaction most often used in social research. In addition, in comparison to measures
of subjective well-being grounded in cognitive evaluations, this index is based on the
evaluation of a series of emotional states recently experienced by individuals. The con-
ceptual definition of socioemotional well-being is based on Thomas Kemper’s social
interactional theory of emotions and Randall Collins’ theory of interaction ritual chains. A
‘‘4 factor, 10 variable’’ solution has been obtained by applying common factor analysis to
the data of the European Social Survey, 2006.

Keywords Subjective well-being ! SWB ! Socioemotional well-being ! Sociology

of emotions ! Composite indicators ! Social indexes ! Quality of life ! Social
quality ! Status ! Power ! Life satisfaction ! Happiness ! Emotional energy

E. Bericat (&)
Department of Sociology, Faculty of Communication, University of Seville, Avda. Américo Vespucio,
s/n, Isla de la Cartuja, 41092 Seville, Spain
e-mail: [email protected]

123
600 E. Bericat

1 Introduction

This article presents the design, process of construction, content and validation of the
Socioemotional Well-Being Index (SEWBI)1. This index is a composite indicator of sub-
jective well-being aimed at the sociological study of quality of life and, specifically, the
analysis of the social stratification of socioemotional well-being.
The concept of quality of life refers to both objective and subjective dimensions of
human existence. The former incorporate normatively determined objective measures of
well-being (such as employment opportunities, income and wealth, education level,
medical attention and quality of housing), while the latter reflect individually perceived or
subjectively experienced well-being (Huppert et al. 2005, 2010). ‘‘Subjective well-being
(SWB) comprises people’s evaluative responses to their lives’’ (Viterso 2004:299). Ve-
enhoven (1984) defines subjective well-being as the degree to which an individual judges
the overall quality of his or her life favourably. This judgment or evaluation can have two
components, one emotional, which may be expressed, for example, by the degree of
happiness an individual experiences, and the other, cognitive, which may be expressed by
the degree of satisfaction an individual feels regarding the life he or she is leading (Diener
1984, 1994).
Over the last decade, social researchers and statistical institutions, as well as politicians
and public officials, have shown great interest in the scientific conceptualisation and
measurement of subjective well-being (SWB) (Noll 2013; De Smedt 2013; Lhéritier 2012;
OECD 2012; ONS 2011; Michaelson et al. 2009; Krueger 2009). In the 1960s pioneering
contributions to its study began to appear (Wilson 1967; Veenhoven 1968; Bradburn 1969;
Easterlin 1974; Diener 1984). Since then, the number of articles published in academic
journals on the subject has grown exponentially. Ever since Richard Easterling showed, in
1974, through his now famous paradox that increases in a country’s gross domestic product
did not involve a parallel increase in happiness as declared by its population, the reasons
for studying subjective well-being and emotional prosperity (Oswald 2010) have continued
to grow. Currently, the need to go beyond GDP as an indicator to measure the progress of a
society makes it even more important to develop adequate models for measuring subjective
well-being (CMEPSP 2009).
However, the urgent need for adequate measurement instruments contrasts with the
enormous methodological difficulty that its measurement presents. And perhaps this
explains the intense academic activity currently taking place in this field, such as, for
example, analysing different measurement perspectives (Tay et al. 2013), or establishing
new forms of measurement (Diener et al. 2010; Hicks et al. 2013). In this context, and in
particular taking into account the relative abandonment in sociology of the study of sub-
jective well-being (Veenhovens 2008), the Socioemotional Well-being Index (SEWBi) has
been conceived with the aim of facilitating and encouraging its incorporation in socio-
logical studies carried out through survey technique. In any case, the index constitutes a
new model for measuring subjective well-being and requires additional development,
further tests of its validity and, above all, future applications to demonstrate its utility in
social analysis.
Two spheres of knowledge have been combined in the construction of this index:
research in social indicators, the recent development of which has been oriented toward the

1
This research has been financed by the Spanish Ministry of Economy and Competitiveness. Research
Project: ‘‘Social Quality in Europe. Design and Development of Composite Indexes for the Measurement
and Monitoring of the Quality of European Societies’’ (CSO2012-35032).

123
The Socioemotional Well-Being Index (SEWBI) 601

elaboration of composite indicators, and the theoretical content developed in recent dec-
ades by the sociology of emotions. As a composite indicator, the index presented in this
article offers a hierarchical and multidimensional alternative to the univariate scales
measuring happiness and satisfaction most often used in sociological research. In contrast
to scales composed of a single variable, the index was created to offer a measurement
model that is both multivariable and parsimonious. The parsimony of the model is
essential, reducing to the minimum the number of questions on the index facilitates its
incorporation into survey questionnaires.
In addition, in comparison to measures of subjective well-being grounded in cognitive
evaluations, this index is based on the evaluation of a series of emotional states recently
experienced by individuals, drawn from their responses to a survey. In this sense, the
SEWBI measures the ‘‘emotional’’ component of subjective well-being exclusively. In
addition, within the broad range of emotional states that can condition an individual’s
emotional well-being, those that, in agreement with sociological theories of the emotions,
can be most related to the position that a specific individual occupies in the social structure
have been chosen. In this sense, the SEWBI exclusively measures the ‘‘social’’ component
of subjective well-being.
This article includes, first, a brief exposition of the two sociological theories of emotions
that provide the theoretical framework of the index. Secondly, through the design of a map
of emotional states, it offers a theoretical definition of the concept of socioemotional well-
being. Third, using data from two different surveys, it shows the results of exploratory
analyses carried out applying principal component analysis (PCA) and common factor
analysis (CFA). Fourth, the solution obtained by applying common factor analysis to the
European Social Survey (ESS 2006) is evaluated; the four factors extracted are interpreted
analytically and a confirmatory factor analysis (CFA) is carried out of the selected mea-
surement model. The last section provides some final comments and information regarding
the nature and utility of the Socioemotional Well-Being Index.

2 Theoretical Framework

The sociology of emotions provides the theoretical framework within which subjective
well-being is conceptualised as a phenomenon dependent on the result of social interac-
tions linked to the positions individuals occupy in the social structure. Thomas Kemper’s
social interactional theory of emotions (1978, 1990, 2006), and Randall Collins’ theory of
interaction ritual chains (1981, 1990, 2004), constitute two pioneering and fundamental
theories that enjoy wide acceptance and are common references in the field of the soci-
ology of emotions (Turner and Stets 2005, 2006).

2.1 A Social Interactional Theory of Emotions

The social interactional theory of emotions is based, first, on the idea that ‘‘a very large
class of emotions results from real, imagined, or anticipated outcomes in social relation-
ships’’ (Kemper 1978:48). In the course of each interaction, actors can maintain, obtain or
lose specific benefits or rewards. If actors obtain a reward or benefit they experience
pleasurable or satisfying emotions; if they lose benefits or rewards they experience
unpleasant or unsatisfying emotions. Secondly, the theory posits that the emotional states
actors experience essentially depend on their relative positioning on two basic dimensions
of sociability, the power dimension and the status dimension.

123
602 E. Bericat

For Kemper (1978:28) the two possible answers to the following question reveal the
analytical difference between ‘‘power’’ and ‘‘status’’: In a given social situation, why does
A do what B wants him to do? First answer: ‘‘A does what B wants because A is actually or
potentially being coerced to do so by B…’’ Second answer: ‘‘A does what B wants because
A wants to do it as a benefit to B.’’ In the first case, A attributes greater power to B; in the
second case, A attributes greater status to B and is therefore willing to voluntarily
accommodate B’s wishes, needs, demands or requests.
Kemper uses a Weberian definition of power: ‘‘…the probability that one actor within a
social relationship will be in a position to carry out his own will despite resistance,
regardless of the basis on which this probability rests’’ (Weber 1978:51). However, in the
status relationship, the key is the willingness with which an actor complies with or offers
benefits or rewards to another. Status is ‘‘the mode of relationship in which there is
voluntary compliance behaviourally with the wishes, desires, wants, and needs of the
other’’ (Kemper 1978:378). The rewards of status, such as admiration, respect, affection
and disinterested support, are given not because the actor is forced to do so, but because of
the social bond created by the deference, esteem, love, appreciation and respect that one
actor feels for another.
The theory predicts the emotional states that actors experience based on their relative or
comparative levels of power and status in relation to other actors. An individual will, in
general, be content and satisfied when he or she considers that his/her power and status are
adequate, and will be discontented or dissatisfied when he or she feels his/her power and
status to be insufficient or excessive. Combining these three possible situations with the two
basic dimensions of sociability, we obtain six different types of structural emotions.
Security is the emotional state that actors experience when they have or believe they
have sufficient power resources to face a specific situation. Adequate power generates
confidence because it permits an individual to have sufficient control over his/her envi-
ronment. It also guarantees that the individual will continue to enjoy the benefits and
rewards that come with power. Its inverse, fear, is the prototypical emotional state of an
actor that faces a dangerous or threatening situation with insufficient power resources.
Limited control over a situation provokes feelings of anxiety and uncertainty. The inability
to force the compliance of others is an obstacle to achieving in this manner certain benefits
or rewards. Lastly, guilt is the emotional state of actors that obtain reward using excessive
power. When the intensity of coercion violates the moral codes regarding the legitimate
use of power, individuals not only feel remorse, but also anxiety and fear because of
possible reprisals resulting from the anger that abuse of power always awakens in those
subject to it.
The feeling of happiness, of being content, of joy, of being esteemed, accepted, loved
and valued by others, is the prototypical emotional state of an actor, who, in the course of a
social interaction receives adequate reward willingly granted by others; ‘‘when one
receives status in adequate amounts one should ‘feel good’’’ (Kemper 1978:59). In con-
trast, an individual will experience emotional states of depression if others do not grant
him or her adequate rewards from status. Given the inherent social nature of human beings,
a more or less persistent deficit of affection, esteem, respect and recognition will bring with
it, under normal conditions, feelings of loneliness, sadness and depression. Lastly, an actor
will feel shame if he or she receives and accepts benefits from status that are not merited, as
‘‘status is given for meeting standards of competence or achievement in the division of
labour or in social relationships’’ (Kemper 1978:59).

123
The Socioemotional Well-Being Index (SEWBI) 603

2.2 Interaction Ritual Chains and Emotional Energy

In The Elementary Forms of the Religious Life, Durkheim presented his studies on the
religion of aboriginal tribes in central Australia. Analysing their ritual practices, he
believed he had found the essence of the sacred. All sacrificial rituals, such as the In-
tichiuma of the Arunta, are structured through a combination of two basic and comple-
mentary acts: an act of alimentary communion, and an act of oblation or offering. A similar
circular logic affects both the religious and the social. Human beings create their gods, but
their gods create human beings. Individuals receive from society that which makes them
human—language, art and morality—but society demands certain sacrifices and
renouncements, certain offerings without which the society could never exist. Rituals are,
for Durkheim, the social institutions which best incarnate the circular logic of the sacred
and the social. Social rituals are quintessential institutions that produce society. Thanks to
the process of cognitive, valuational and emotive synchrony/attunement activated by the
ritual, the sacred emerges as the collective conscience through which individuals experi-
ence the power of the social. This power manifests itself in the intense emotion or col-
lective effervescence that participants in a ritual experience (Durkheim 1951).
Collins’ theory adopts Durkheim’s conception of ritual, but following in Goffman’s
footsteps (1967) it extends its application to the social interactions that are part of daily
life. ‘‘Ritual is a mechanism of mutually focused emotion and attention producing a
momentarily shared reality, which thereby generates solidarity and symbols of group
membership’’ (Collins 2004: 7). The ingredients of ritual provoke a collective intensifi-
cation of emotional experience, producing the following effects: ‘‘(1) group solidarity, a
feeling of membership; (2) emotional energy [EE] in the individual: a feeling of confi-
dence, elation, strength, enthusiasm and initiative in taking action; (3) symbols that rep-
resent the group; … these are Durkheim’s ‘‘sacred objects’’. (4) feelings of morality: the
sense of rightness in adhering to the group…’’ (Collins 2004:49)
This conception of ritual underpins the key concept in Collins’ sociology of emotions:
emotional energy (EE), a lasting mood that occurs in the individual after having collec-
tively shared the same emotional state with others. Which concrete emotional ingredients
(sadness, horror, pride, happiness, etc.) feed the collective effervescence of a ritual (Collins
2004: 107–108) are not important; what is important is the emotion that persists beyond the
ritual. Collins uses the concept of emotional energy to refer to these persisting emotions; in
other words, EE is a basic psycho-physiological pattern associated with the humours,
lasting moods and deep feelings that compose individuals’ daily lives. A successful ritual
elevates the emotional energy of the participants, while a failed, empty or forced ritual
diminishes it (Collins 2004:50). The concept of emotional energy is linked to the primary
emotions of happiness or joy and sadness or disappointment. High emotional energy
involves exhilaration, joy, enthusiasm, effervescence, vitality, feeling good about one-self
or confident, while low emotional energy involves disappointment, depression, lack of
initiative and negative feelings toward one-self. ‘‘Emotional energy is like the psycho-
logical concept of ‘drive’, but it has a specifically social orientation. High emotional
energy is a feeling of confidence and enthusiasm for social interaction’’ (Collins 2004:
108).
The theory of Interaction Ritual Chains proposes that individuals acquire or lose
emotional energy in both power and status interactions. ‘‘Order-givers maintain and
sometimes gain EE, order-takers lose it; being in the focus of attention and thereby suc-
cessfully enacting group membership raises EE, experiencing marginality or exclusion
lowers it. Interaction rituals are connected in chains over time, with the results of the last

123
604 E. Bericat

interaction (in emotions and symbols) becoming inputs for the next interaction; thus EE
tends to cumulate (either positively or negatively) over time’’ (Collins 2004:118). In short,
‘‘High and low EE come from the entrainment of communicative gestures and emotions
rhythms that are distinctive to human intersubjectivity; from an individual viewpoint, they
are tightly woven together into the human self’’ (Collins 2004:107).

3 The Conceptual Definition of Socioemotional Well-Being

Kemper and Collins’ theories of the emotions provide a suitable framework to develop a
sociological-emotional conceptualisation of subjective well-being.
The two theories are essentially compatible and complementary. First, both consider
social interactions to provoke most of the emotions that really affect and are of importance
to us. Secondly, both consider individuals’ emotional states to be conditioned by the results
they obtain in processes of interaction. Third, despite their micro-sociological character,
both theories can be projected onto the macro-sociological sphere as they adopt a struc-
tural perspective ab initio. And finally, both articulate the emotional dynamic of actors
based on two basic dimension of sociability: power and status (Kemper and Collins 1990).
Both the theory of interaction ritual chains and the social interactional theory of
emotions maintain that we experience a more or less stable general emotional mood in our
lives. The first distinguishes between long-term and transitory emotions, and the latter
between structural and momentary emotions. Although individuals can obtain or lose
emotional energy in the course of each social interaction, according to Collins accumulated
emotional energy does not dissipate instantaneously but lasts over time. For his part,
Kemper argues ‘‘that each actor is either satisfied or dissatisfied in some degree with his
own and the other’s positions on the power and status dimensions’’, and this is expressed in
structural emotions, which result from the relatively stable structure of a specific social
relationship (Kemper 1978:49). ‘‘From the structural perspective, the amount of status an
actor receives may have a certain degree of stability and consist of a probability that certain
types and amounts of rewarding behaviour will be accorded to him’’ (Kemper 1978:384).
In addition, a certain probability exists that an actor will receive specific rewards in
function of his/her positioning in the power dimension. In short, this general and lasting
emotional state, posited by both theories, would correspond with the degree of socio-
emotional well-being (SEWB) an individual experiences.
At this point, with the aim of operationalising the concept and selecting concrete
emotions which, in the judgment of both authors, are most closely linked with this general
state of emotional well-being, an exhaustive analysis of the emotional content of both
theories has been carried out, in other words, of the emotions specifically associated with
the two theoretical dimensions that Kemper and Collins respectively emphasise. From this
analysis we have constructed a map of emotions that characterises socioemotional well-
being, which can be seen in Fig. 1.
The emotional states that Collins mentions in referring to emotional energy (EE)—
vitality/enthusiasm/effervescence versus dejection/depression on one side, and content-
ment/happiness versus sadness/shame on the other—correspond to the two dimensions of
this concept, in other words, a certain level of energy on the one hand, and a certain level
of emotional well-being on the other. Following Kemper’s theory, which predicts the
emotional states that individuals experience based on the degree of relative power and
status they have in the context of a given social situation or interaction, we have the
emotions of the two other axes of socioemotional well-being: Individuals can feel

123
The Socioemotional Well-Being Index (SEWBI) 605

Fig. 1 Socioemotional well-being (SEWB). Conceptual definition

confidence/security versus fear/anxiety/worry on the one hand; and pride/being esteemed/

being respected versus feelings of loneliness/shame on the other.
As can be seen in Fig. 1, the map of emotional states linked to the concept of socio-
emotional well-being is structured by four axes: (a) the general level of vital energy the
individual operates with, which can be high or low; (b) the general quality of emotions
affecting each individual, which can be satisfying or unsatisfying, positive o negative, good
or bad; (c) the emotions associated with the power dimension, such as confidence or fear,
and (d) the emotions associated with the status dimension, such as pride or shame.
The configuration of this map of emotional states and the definition of socioemotional
well-being proposed are also based on Scheff’s theory of pride and shame (1988, 1990,
2000). Scheff considers both pride and shame to be the quintessential social emotions, as
both indicate the state of our social bonds: if the bond is secure, we feel pride, which is a
pleasant emotion; if it is insecure, we feel shame, a painful emotion. Collins explicitly
recognises the value of this theory: ‘‘Scheff’s model is a valuable complement to IR theory
because it specifies emotions generated by both high and low levels of Durkheimian
solidarity. Successful interactional attunement or an intact social bond generates pride;
breaking the bond generates shame. In the terms of IR theory, pride is the emotion attached
to a self energised by the group; shame is the emotion of a self depleted by group
exclusion’’ (Collins 2004:120).
In short, security, vitality, happiness and pride, on the one hand, versus fear, depression,
sadness and shame, on the other, constitute the basic emotions used to measure the

123
606 E. Bericat

emotional components of individuals’ subjective well-being. Socioemotional well-being is

a general and relatively stable emotional state associated with a determined level of
vitality (high or low), and the experience of different emotions, which can be positive or
negative. Along with this definition of the emotional side, the sociological or socio-rela-
tional side must be defined. From this perspective and considering that ‘‘emotions con-
stitute the bodily manifestation of the importance that an event in the natural or social
world has for a subject’’ (Bericat 2012b:2), we can conclude that socioemotional well-
being is a general and relatively stable emotional state that indicates the emotional
evaluation, positive or negative, that an individual makes of the results of the totality of
his/her social interactions.

4 The Operational Definition of Socioemotional Well-Being

Once a conceptual definition of socioemotional well-being is reached, it is necessary to

establish an empirical operationalisation that permits us to measure and validate the
concept. To do this, the 2006 European Social Survey was used. This survey incorporated,
as proposed by a team directed by Felicia Huppert (Huppert et al. 2005, 2010), an excellent
module on Personal and Social Wellbeing that contained a significant number of questions
on emotions. To choose the questions to include in this module, the team of researchers
carried out a prior study in which they took into account diverse empirical approaches to
the measurement of well-being. From a hedonist perspective on the measurement of well-
being they considered in particular multivariable scales that measure positive and negative
emotions, such as the Affect Balance Scale that Bradburn introduced in 1969 and Watson’s
Positive and Negative Affect Scale (Watson et al. 1988).
Based on the series of questions on emotional states that were finally included in this
survey module, as well as the emotional map based on the analysis of Kemper and Collins,
we have developed an empirical operationalisation selecting eight questions from the
module. First, taking into account the widely demonstrated differential response that
respondents provide to questions on positive and negative emotions (Bradburn 1969;
Carmines and Zeller 1979; Rosenberg 1965; Diener and Lucas 1999), both positive and
negative emotional states were chosen. Secondly, those emotional states most closely
linked to the sociological dimensions of power and status were selected, also including
emotions associated with Collins’ concept of emotional energy, in other words, related to
individuals’ levels of energy and vitality, as well as their happiness and enjoyment of life.
Lastly, following the principle of parsimony, necessary so that the index can be incor-
porated into surveys and sociological analysis, the number of questions have been reduced
as much as possible, taking into account the theoretical dimensions discussed. Figure 2
shows the eight emotional states finally selected from the European Social Survey to
empirically operationalise the index.
The exact formulation of the questions for measuring six of the emotional states in the
ESS-2006 survey was the following: ‘‘I will now read out a list of the ways you might have
felt or behaved in the past week. Please tell me how much of the time during the past week:
(1) you felt depressed, (2) you were happy; (3) you felt lonely, (4) you enjoyed life, (5) you
had a lot of energy, (6) you felt calm and peaceful’’. The possible answers were: None or
almost none of the time; Some of the time; Most of the time; All or almost all of the time.
For the two other emotional states the questions were in a different format, asking
respondents for their degree of agreement (Agree strongly, Agree, Neither agree nor

123
The Socioemotional Well-Being Index (SEWBI) 607

Fig. 2 Emotional content of the

ESS-2006 measurement model

Fig. 3 Emotional content of the

CIS-2011 measurement model

disagree, Disagree, Disagree strongly) with the following two statements: (7) ‘‘I’m always
optimistic about my future’’, (8) ‘‘In general I feel very positive about myself’’.
With the aim of empirically validating the conceptual definition of socioemotional well-
being through the application of methodological triangulation, a second operationalisation
approach using another survey, different emotional states and a different question and
answer format was developed. Figure 3 includes the emotional states that we proposed to
include in the questionnaire of a survey carried out in Spain by the Centro de Investi-
gaciones Sociológicas (CIS)2 in December 2011 (CIS 2011).
The exact formulation of the questions in the CIS-2011 survey was the following: ‘‘In
what follows I am going to ask you questions about how you currently feel. Can you tell
me to what extent you lately feel very, quite, little or not at all…?’’ The eight emotional
states were the following: (1) Proud of yourself, (2) Worried about the things happening to
you, (3) Full of energy and life, (4) lonely, (5) That you are really enjoying life, (6)

2
The CIS (Centro de Investigaciones Sociológicas) is Spain’s leading public institution in the area of social
and policy research. The author is grateful to its prior president, Ramón Ramos, and research director, Javier
Callejo, for the inclusion of the battery of questions on emotional states in the survey referred to in this
study.

123
608 E. Bericat

Depressed with no desire to do anything, (7) Stressed because of all the things you have to
do, and (8) Satisfied with the life you are leading.
The application of different exploratory analyses to both data sets (ESS-2006 and CIS-
2011), using both principal component analysis and common factor analysis, permits us to
evaluate the consistency and coherency of the developed construct.

4.1 Exploratory Analysis

In a first phase, the exploratory analysis applies principal component analysis to both the
Spanish CIS-2011 survey,3 as well as the Spanish sub-sample of the ESS-2006. From these
preliminary analyses only the 3 component, 8 variable solution is presented. The 2 factor, 8
variable solutions, as theoretically represented in Figs. 2 and 3, do not fit the empirical
model. In a second phase, data exploration is carried out applying common factor analysis
to both the CIS-2011 and the European sample of the ESS-2006. A 4 factor, 8 variable
solution was finally chosen. With the objective of validating the substantive interpretation
of this 4 factor solution, a similar extraction using 19 emotional items from the ESS-2006
was obtained.
In comparing the results, it is very important to keep in mind that the format of
questions and answers used in the two surveys is very different. The CIS-2011 asks about
the recent intensity of the emotional states experienced by the respondent. The ESS-2006
asks about the frequency of the feelings experienced during the past week. It is also
necessary to emphasise that both the time when the surveys were carried out and their
samples are different.

4.1.1 Principal Component Analysis

Through this initial exploration the intention was to verify the following: first, up to what
point the variation in responses given by the respondents to the eight questions could be
explained by fewer dimensions; second, to see if the interpretation of these dimensions was
consistent with the conceptual definition of socioemotional well-being; and, third, to
investigate the possibility of a valid composite indicator of socioemotional well-being.
Tables 1, 2, 3 and 4 show the total variance explained by the first three components
(63.5 % for the CIS-2011, and 66.8 % for the ESS-2006-Spain), as well as the loadings of
the rotated component matrix. This first analysis reveals a high level of consistency,
although we do find some important discrepancies. Exactly the same variables (depression
and loneliness) saturate the second component in both models. In addition, three of the four
variables that saturate the first component in both models are the same (enjoyment of life,
satisfaction/happiness and vitality). However, while the feeling of pride saturates the first
component in the CIS-2011, it is the feeling of calm and peacefulness that saturates this
component in the ESS-2006. Lastly, while the third component of the CIS-2011 is satu-
rated by worry and stress, which are two emotional states linked to the social dimension of
power, the third component of the ESS-2006 is saturated by two different emotions, self-
esteem and optimism, emotions linked to pride.

3
The CIS-2011 is a representative survey of the Spanish population 18 years of age or older of both sexes,
with a sample size of 2,483 respondents.

123
The Socioemotional Well-Being Index (SEWBI) 609

Table 1 Principal component analysis (explained variance. 3 factor, 8 variable solution. CIS-2011)
Component Initial eigenvalues Rotated solution; eigenvalues

Total % variance % cumulative Total % variance % cumulative

1 2.997 37.458 37.458 2.326 29.074 29.074

2 1.205 15.058 52.516 1.491 18.639 47.713
3 .880 11.000 63.517 1.264 15.803 63.517
4 .803 10.037 73.554

Table 2 Principal component analysis (explained variance. 3 factor, 8 variable solution. ESS-2006-Spain)
Component Initial eigenvalues Rotated solution; eigenvalues

Total % variance % cumulative Total % variance % cumulative

1 3.475 43.441 43.441 2.192 27.395 27.395

2 .980 12.256 55.697 1.612 20.144 47.539
3 .890 11.120 66.817 1.542 19.278 66.817
4 .668 8.356 75.173

Table 3 Principal component

Emotional states Components
analysis (rotated component
matrix. 3 factor, 8 variable solu- 1 2 3
tion. CIS-2011)
Proud of yourself .728 .074 .061
Worried about the things happening to you -.165 -.117 .852
Full of energy and life .718 -.236 .090
Lonely -.195 .827 -.067
Enjoying life .713 -.321 -.134
Depressed with no desire to do anything -.453 .616 .255
Stressed because of all the things you have .163 .429 .604
to do
Bold values indicate variables Satisfied with the life you are leading .689 -.253 -.272
with the highest factor loading

Table 4 Principal component

Emotional states Components
analysis (rotated component
matrix. 3 factor, 8 variable solu- 1 2 3
tion. ESS-2006-Spain)
Were happy .644 -.399 -.232
Felt depressed -.309 .730 .171
Had lot of energy .760 .005 -.244
Felt lonely -.127 .850 .119
Enjoyed life .742 -.219 -.210
Always optimistic about my future -.197 .036 .859
In general feel very positive about myself -.153 .298 .776
Bold values indicate variables Felt calm and peaceful .689 -.243 .016
with the highest factor loading

123
610 E. Bericat

4.1.2 Common Factor Analysis

With the aim of resolving the inconsistencies detected, we continue the validation process
applying exploratory factor analysis to the both Spanish CIS-2011 and the European
sample of the ESS-2006.
Common factor analysis (principal axes) is the proper technique when the intention is to
validate a measurement model consistent with a previously established conceptual defi-
nition. In contrast to principal component analysis, which aims to maximise the expla-
nation of total variance, factor analysis distinguishes between common variance and
unique variance, exclusively maximising the explanation of the common variance of the
variables introduced in the analysis (Cea D’Ancona 2002; Gorsuch 1983; Nardo et al.
2005). Given that we assume the existence of a reality that corresponds to the construct of
socioemotional well-being, the measurement that establishes the composite indicator has to
exclusively consider the variance that is common to all the variables and not that which
may be due to other factors, foreign to the concept intended to be measured. Thus, for
example, we are not interested in explaining all the variance of ‘‘feelings of loneliness’’,
but only that part linked to greater or lesser socioemotional well-being. In short, it is clear
that in the construction of indices or composite indicators, we want the factors to maximise
the explanation of the common variance, in other words, the variance associated with the
construct being measured.
After carrying out and evaluating the different exploratory analyses, the four factor,
eight variable solution was chosen. This solution minimises the inconsistencies detected in
the previous phase, improves the ability to distinguish the emotional content of the factors
and makes a more coherent theoretical explanation of them possible.
Despite the significant differences between the surveys (date, universe, sample,
emotional content, question format, etc.), the results obtained are very similar, revealing
the potential coherency, validity and reliability of a composite index for a set of emo-
tional states linked to the concept of socioemotional well-being. The percentages of
variance explained are included in Tables 5 and 6, and the factor loadings in Tables 7
and 8.
In the four factor and eight variable solution for both surveys, the first factor is now
saturated almost exclusively by the emotional states of happiness (or satisfaction) and
enjoyment. In both surveys a factor clearly saturated by depression and loneliness also
exists, as does a factor saturated by feelings of pride or self-esteem and optimism. Finally,
in the CIS-2011, there is a factor saturated by worry and stress, negative emotional indi-
cators of the power dimension. In the ESS-2006, feeling calm and peaceful, as well as
having energy, positive emotional indicators of the power dimension, saturate the last
factor. In short, the degree of consistency obtained in this four factor and eight variable
solution is very high.
Lastly, common factor analysis was applied to a total of 19 questions on emotional
states included in the Personal and Social Well-Being Module of the ESS-2006. The
intention of this analysis was, first, to improve the construct validity of the measurement
model without substantially altering its degree of parsimony and, secondly, to improve the
discriminant validity of the factor linked to the power dimension (Batista-Foguet et al.
2004), given that the two variables that saturate the fourth factor in the European survey
are closely correlated with the first factor. Table 9 shows the factor loadings of the rotated
matrix.

123
 The Socioemotional Well-Being Index (SEWBI)
Table 5 Common factor analysis (explained variance. 4 factor, 8 variable solution. CIS-2011)
Factor Initial eigenvalues Extraction sums of squared loading Rotations sums of squared loadings

Total % variance % cumulative Total % variance % cumulative Total % variance % cumulative

1 2.997 37.458 37.458 2.567 32.092 32.092 1.250 15.627 15.627

2 1.205 15.058 52.516 .555 6.932 39.024 1.030 12.869 28.496
3 .880 11.000 63.517 .324 4.047 43.071 .817 10.212 38.708
4 .803 10.037 73.554 .258 3.229 46.300 .607 7.592 46.300
5 .666 8.323 81.877
Extraction method: principal axis factoring
123

611
 612
123

Table 6 Common factor analysis (explained variance. 4 factor, 8 variable solution. ESS-2006)
Factor Initial eigenvalues Extraction sums of squared loading Rotations sums of squared loadings

Total % variance % cumulative Total % variance % cumulative Total % variance % cumulative

1 3.286 41.073 41.073 2.780 34.748 34.748 1.222 15.278 15.278

2 1.047 13.091 54.164 .538 6.726 41.473 1.022 12.776 28.053
3 .835 10.444 64.607 .298 3.727 45.200 .940 11.753 39.807
4 .701 8.762 73.369 .175 2.189 47.389 .607 7.583 47.389
5 .639 7.983 81.352
Extraction method: principal axis factoring

E. Bericat
The Socioemotional Well-Being Index (SEWBI) 613

Table 7 Common factor analysis (rotated factor matrix. 4 factor, 8 variable solution. CIS-2011)
Emotional states Factor

1 2 3 4

Enjoying life .835 -.269 .187 -.118

Satisfied with the life you are leading .497 -.280 .296 -.280
Proud of yourself .182 -.118 .710 .002
Worried about the things happening to you -.107 .021 -.015 .500
Depressed with no desire to do anything -.244 .691 -.201 .344
Lonely -.205 .508 -.090 .090
Full of energy and life .400 -.327 .376 .027
Stressed because of all the things you have to do -.005 .151 .011 .372

Extraction method: principal axis factoring

Rotation method: Varimax with Kaiser normalisation
Bold values indicate variables with the highest factor loading

Table 8 Common factor analysis (rotated factor matrix. 4 factor, 8 variable solution. ESS-2006)
Emotional states Factor

1 2 3 4

Always optimistic about my future -.175 .643 .143 -.112

In general feel very positive about myself -.099 .642 .138 -.148
Felt depressed, how often past week -.211 .172 .598 -.271
Were happy, how often past week .570 -.193 -.329 .255
Felt lonely, how often past week -.203 .136 .539 -.080
Enjoyed life, how often past week .769 -.158 -.229 .185
Had lot of energy, how often past week .349 -.223 -.232 .319
Felt calm and peaceful, how often past week .243 -.191 -.195 .540

Extraction method: Principal axis factoring

Rotation method: Varimax with Kaiser normalisation
Bold values indicate variables with the highest factor loading

After the observation and evaluation of the results, it was decided that the incorporation
of two new emotional states could favour both the correct substantive interpretation of the
factors, as well as the discriminant validity of the fourth factor. The items incorporated are
the feeling of ‘‘sadness’’, associated with loneliness and depression, and the sensation of
‘‘being rested when waking up in the morning’’, associated with the absence of worries and
with energy. This 4 factor, 10 variable solution is the one finally used to interpret the four
dimensions of the Socioemotional Well-being Index.

5 Empirical Operationalisation

The empirical operationalisation of the measurement of socioemotional well-being was

obtained applying common factor analysis (principal axes) to the ESS-2006 survey (ESS-
2006, ed.3.3, 2011). Based on this analysis, four factors were extracted from a total of ten

123
614 E. Bericat

Table 9 Common factor analysis (rotated factor matrix. 4 factors, 19 variable solution. ESS-2006)
Emotional states Factor

1 2 3 4

Always optimistic about my future .151 -.175 .605 .115

In general feel very positive about myself .113 -.110 .671 .141
At times feel as if I am a failure -.220 .198 -.321 -.127
On the whole life is close to how I would like it to be .296 -.291 .411 .061
Felt depressed, how often past week .658 -.222 .182 .182
Felt everything did as effort, how often past week .509 -.148 .092 .292
Sleep was restless, how often past week .393 -.066 .110 .402
Were happy, how often past week -.330 .608 -.232 -.086
Felt lonely, how often past week .546 -.202 .145 .047
Enjoyed life, how often past week -.276 .671 -.191 -.084
Felt sad, how often past week .695 -.198 .172 .127
Could not get going, how often past week .505 -.182 .095 .272
Had lot of energy, how often past week -.204 .474 -.167 -.380
Felt anxious, how often past week .514 -.032 .223 .237
Felt tired, how often past week .354 -.071 .120 .547
Absorbed in doing, how often past week -.035 .307 -.061 -.090
Felt calm and peaceful, how often past week -.227 .328 -.223 -.351
Felt bored, how often past week .411 -.159 .102 .113
Felt rested when woke up in morning, how often past week -.124 .251 -.145 -.574
Extraction method: principal axis factoring
Rotation method: Varimax with Kaiser normalisation
Bold values indicate variables with the highest factor loading

variables that record the emotional states that respondents stated they experienced. In the
appendix, the sociodemographic characteristics of the sample used are described
(Table 13), as well as the responses given by the respondents to each of the ten questions
included in the measurement model (Tables 14, 15, 16, 17).
In this section we first provide and evaluate the basic parameters of the exploratory
factor analysis (EFA) carried out. Secondly, we interpret the four factors of this factorial
solution in terms of the basic dimensions involved in the measurement of socioemotional
well-being. Lastly, we carry out a confirmatory factor analysis (CFA) with the aim of
validating the measurement model.

5.1 Description and Analytical Evaluation of the Measurement Model

The factorial solution was obtained with the data from a sample from the European Social
Survey in which 20 countries were included (Austria, Belgium, Bulgaria, Switzerland,
Cyprus, Germany, Denmark, Estonia, Spain, Finland, France, United Kingdom, Ireland,
Netherlands, Norway, Poland, Portugal, Sweden, Slovenia and Slovakia). The sample size
was 37,043 respondents, and the sample was weighted using the variables ‘‘design weight’’
and ‘‘population size weight’’. The process of extraction of the four factors required 35
iterations. Given the low number of lost values (0.5 %), these values have been replaced

123
 The Socioemotional Well-Being Index (SEWBI)
Table 10 Common factor analysis (variance explained. 4 factor, 10 variable solution. ESS-2006)
Factor Initial Eigenvalues Extraction sums of squared loading Rotations sums of squared loadings

Total % Variance % Cumulative Total % Variance % Cumulative Total % Variance % Cumulative

1 3.920 39.202 39.202 3.427 34.275 34.275 1.575 15.750 15.750

2 1.110 11.098 50.301 .608 6.084 40.359 1.269 12.687 28.437
3 1.004 10.036 60.337 .492 4.918 45.277 1.034 10.337 38.775
4 .819 8.188 68.525 .380 3.796 49.073 1.030 10.299 49.073
5 .652 6.521 75.045
Extraction method: principal axis factoring
Rotation method: Varimax with Kaiser normalisation
123

615
616 E. Bericat

with the average. The Varimax Method (Varimax with Kaiser Normalisation), an
orthogonal rotation method that extracts uncorrelated factors and facilitates their inter-
pretation minimising the number of variables that have high loadings on each factor, has
been used for the rotation of factors. The rotation converged after 5 iterations. For the
estimate of the factor score coefficients the regression method was used.
The eigenvalues are included in Table 10. The four common factor solution explains 49.07
per cent of the total variance. The correlation matrix shows that approximately half of the
correlation coefficients are close to 0.3, that a third are above 0.3, and that a fourth are above
0.45. In a factor analysis the variables should be correlated, but their correlation should not be
extremely high (Field 2000:444). Bartlett’s test of sphericity is statistically significant to
0.000, which proves the existence of correlation between the variables. At the same time, as
the determinant of the correlation matrix is greater than 0.00001 (determinant = 0.68), this
reveals that no multicollinearity exists and that linear combinations of the correlated variables
can be obtained (Field 2000: 445). The Kaiser–Meyer–Olkin measure of sampling adequacy,
or the KMO index, has a value of 0.860, much higher than the minimum limit required ([0.5)
in the academic literature (Field 2000: 446). In the anti-image correlation matrix we can also
confirm that the KMO values for all the individual variables are higher than 0.50. In concrete,
the KMO values for the variables introduced in the analysis range from 0.808 to 0.911. In the
reproduced correlation matrix no redundant residual with an absolute value greater than 0.05
is found, which indicates a good fit for the model (Cea D’Ancona 2002:472).
In short, all these statistical parameters demonstrate the suitability of applying factor
analysis to the chosen emotional states. In the following section this measurement model is
substantively evaluated.

5.2 Interpretation of Common Factors

Considering the factor loadings included in Table 11 we can conclude that the four ana-
lytical dimensions of socioemotional well-being, based on the measurement model
obtained, are the emotional evaluation of status, of situation, of self and of power.

Table 11 Common factor analysis (rotated factor matrix. 4 factor, 10 variable solution. ESS-2006)
Emotional states Factor loadings

Status (f1) Situation (f2) Self (f3) Power (f4)

Optimistic about my future .150 -.176 .634 -.122

Very positive about myself .139 -.102 .644 -.156
Depressed .619 -.234 .169 -.212
Happy -.304 .629 -.190 .210
Lonely .518 -.212 .130 -.094
Enjoy life -.242 .709 -.158 .206
Sad .782 -.159 .134 -.167
Feel energetic -.221 .346 -.205 .383
Calm and peaceful -.238 .261 -.187 .452
Rested upon waking in the morning -.119 .128 -.122 .687

Extraction method: principal axis factoring

Rotation method: Varimax with Kaiser normalisation
Bold values indicate variables with the highest factor loading

123
The Socioemotional Well-Being Index (SEWBI) 617

Emotions, as Freud indicated in his discussion of anxiety, have a signalling function,

that is, they indicate something to the person that feels them (Freud 1948). This signalling
function has been emphasised by many sociologists of emotions, such as Hochschild
(1983) and Scheff (1990), and from this perspective, it can be stated that ‘‘emotions
constitute the bodily manifestation of the importance that an event in the natural or social
world has for an individual’’. ‘‘In its most basic expression this involves three elements:
(a) the assessment/appraisal; (b) of an event in the world; (c) made by an individual’’
(Bericat 2012b:2). Nussbaum argues that ‘‘emotions have to do with whatever I do value’’
(Nussbaum 2001: 49), in other words, emotions indicate how we evaluate that which
affects us.
Thus, the dimensions of the socioemotional well-being index indicate how individuals
evaluate their social status, general life situation, themselves or their ‘‘self’’, and their
social power. In this regard, it should be noted that the measurement model of the index
necessarily incorporates all the complexity involved in our emotional processes (Bericat
2012b). The four dimensions are:
(a) Status: sadness, depression and loneliness are the emotional states that saturate the
status factor to the greatest degree. These states occupy the lower right quadrant of the
conceptual map represented in Fig. 1, indicating a low level of vital energy and emotional
well-being. Although depression can be conditioned by an individual’s persistent failures
and by frustration due to a lack of power resources (Seligman 1975), in general it is closely
linked to the absence of the type of rewards characteristic of the relational dimension of
status (Scheff 1990). Respect, recognition, affection and love are rewards that others
voluntarily give to the individual (Kemper 1978). Scheff’s theory of pride and shame takes
into account the enormous importance that the quantity and/or quality of our social bonds
have on our emotional well-being (Scheff 1990).
(b) Situation: feelings of enjoying life and of happiness are those with the greatest
weight in the composition of the second factor, which indicates the emotional evaluation
the individual makes regarding his/her general life situation. Empirical studies carried out
measuring levels of happiness (Veenhoven 1984) or satisfaction (Diener et al. 1999, 2012)
show that these subjective assessments are correlated with the objective and external
conditions the individual is living through at the moment. The link between subjective
well-being and energy level, which characterises Collins’ concept of emotional energy, is
at least partially supported by the importance that feeling full of energy or vitality has on
this factor.
(c) Self: self-esteem and optimism regarding one’s own future are the emotional states
that saturate the third factor, which refers to the self, as it measures pride and the evalu-
ation the individual makes of him/herself. The emotional well-being of an individual is not
exclusively conditioned by the objective conditions of his/her life situation, but also by the
assessment that the individual makes of him/herself. Both self-esteem and optimism are
emotional indicators of the personal and psychological resources available to the individual
in facing life situations (Huppert et al. 2010; Stets 2010; Tinkler and Hicks 2011). Self-
esteem and optimism are essentially diachronic emotional states, as they feed on the
successes and failures experienced by the individual in the past, and they project into the
future in the form of accumulated energy, confidence and determination. As a result,
Kemper (2006: 101) argues that what he calls anticipatory emotions arise from individuals’
levels of optimism or pessimism and confidence or lack of confidence.
(d) Power: in the original approach, included in the CIS-2011 survey, the fourth factor is
clearly saturated by two emotional states: worry about the bad things happening to the
respondent, and stress over the quantity of things that have to be done. Both worry and

123
618 E. Bericat

stress are indicators of the lack of control or power that the individual has over his/her life
situation or over certain important aspects of life. Worry forms part of the emotional family
of fear and is provoked by the expectation that something bad or undesirable may happen.
Stress is the emotional consequence experienced by an individual who is obliged for some
reason to do more than he/she feels capable of doing. These negative emotions are located
in the lower left quadrant of Fig. 1 but are not included in the ESS-2006, although certain
positive states that are also indicators of the assessment individuals make in terms of the
sufficiency or insufficiency of their power resources are. Waking up with the feeling of
having rested during the night indicates that the worries an individual may have did not
alter his/her sleep. In addition, feelings of calm and peacefulness indicate that the indi-
vidual’s daily obligations do not produce stress nor do his/her worries provoke anxiety.
These emotions are located in the upper left quadrant of Fig. 1, forming part of the power
factor, as they also reflect the emotional assessment that the individual makes of the degree
to and mode in which he/she controls the situation.

5.3 Confirmatory Factor Analysis

The final phase of the analysis is aimed at validating the measurement model of the
proposed index and consists in the application of confirmatory factor analysis (CFA). CFA,
according to many authors, is the most ideal technique for confirming the validity and
reliability of a measurement model (Long 1983; Batista-Foguet et al. 2004; Cea D’Ancona
2002; Albright and Park 2009; Bollen 1989; Kline 2011; Arbuckle 2010). A good fit for the
model would validate the four analytical dimensions of the concept of socioemotional
well-being, the proposed correspondence between variables and latent factors, the ade-
quacy of the number of variables finally used and their reliability.
Figure 4 presents the standardised parameters corresponding to the measurement model
of 10 variables and four factors. The specification of this model assumes that a correlation
exists between all the latent factors, that the error terms are uncorrelated and that each
empirical variable saturates a single factor. For the identification of the model the factor
loading of one of the variables for each latent factor has been set at 1.
Each two-headed arrow indicates the covariance between two latent factors, and the
figures located in the centre of these arrows, the correlation between them. The figures
alongside of the arrows that unite factors and variables are the standardised regression
weights, which can be interpreted as the factor loadings in the exploratory factor analysis.
Lastly, the figures located in the upper right of the rectangles indicate the communality or
proportion of variance for each variable that can be explained by the latent factor. This
latter parameter can be interpreted as an estimate of the reliability of the variables (Batista-
Foguet et al. 2004) (Arbuckle 2010).
With the aim of evaluating the fit of the measurement model corresponding to four
different factor solutions, Table 12 presents some of the most widely used tests: Root
Mean Square Error of Approximation (RMSEA); Incremental Fit Index (IFI); Tucker-
Lewis index (TLI); and Comparative Fit Index (CFI). The lower the value of the RMSEA
index, the greater the fit of the model, estimating that values equal to or below 0.05 indicate
a good fit. In the case of the IFI, TLI and CFI indexes, the majority of authors consider a
value over 0.9 to indicate a good fit (Cea D’Ancona 2002; Arbuckle 2010), although some
authors elevate this value to 0.95 (Halleröd and Seldén 2012; Wu and Yao 2007).
Based on Table 12, the initial theoretical approach represented in Fig. 2, with power
and status as the only two factors and with four emotional states linked to each factor, does
not show a good fit (RMSA = 0.114). The model with three factors reveals a good fit, but

123
The Socioemotional Well-Being Index (SEWBI) 619

Fig. 4 Confirmatory factor analysis. Standardised estimates. 10 variable, 4 factor model. ESS-2006

Table 12 Confirmatory factor analysis (model fit. Different models. ESS-2006)

Models RMSEA IFI TLI CFI

4 factors 10 variables (4f10v) 0.039 0.984 0.970 0.984

4 factors 9 variables (4f9v) 0.032 0.992 0.982 0.992
4 factors 8 variables (4f8v) 0.029 0.994 0.985 0.994
3 factors 8 variables (3f8v) 0.042 0.985 0.968 0.985
2 factors 8 variables (2f8v) 0.114 0.878 0.768 0.878

close to the maximum limit (RMSA = 0.042). Lastly, the three models of four factors fit
well. In short, this triple confirmation contributes to validating the structure of emotional
states and the dimensions of the measurement model for the SEWBI.

6 The Socioemotional Well-Being Index (SEWBI): Final Considerations

Based on the proposed measurement model, the SEWBI is the unweighted arithmetic
average multiplied by one hundred of the factor scores obtained in the rotated solution of
the applied common factor analysis. The factors were not weighted because the percentage
of variance they explain (Table 10) is similar: Status (15.75 %); Situation (12.69 %); Self

123
620 E. Bericat

(10.34 %); and Power (10.30 %). The negative signs for F1 and F3 serve to orient the
value of the four factors in the same direction, in other words, the higher the value on the
index, the higher the level of socioemotional well-being. Looking at Tables 14, 15, 16 and
17 we can see that the relationship of the scores on factors 1 (status) and 3 (self) to
socioemotional well-being are the inverse—the higher the value in this response category,
the lower the socioemotional well-being. In Table 13 in the appendix, we can see the
general scores for the index, as well as for its four dimensions for each one of the sample’s
sociodemographic categories.
! "
ð#F1Þ þ ðF2Þ þ ð#F3Þ þ ðF4Þ
SEWBI ¼ ' 100
4
Regarding the analytical structure of this measurement model, socioemotional well-
being is the vital balance that results from a set of emotional states experienced by the
individual, who evaluates his/her general life situation, the self that lives it, and his/her
position of status and position of power (Fig. 5). From a theoretical-practical perspective,
the index is a composite indicator of subjective well-being that reflects the emotional
consequences arising from both the social position occupied by the individual, as well as
the result of the totality of his/her social interactions.
All indices and all composites indicators are the result of a complex empirical capture
that, through a very specific set of observational operations, ultimately establishes the
measurement of a specific phenomenon. In this sense, the Socioemotional Well-being
Index has to be considered an initial approach, with methodological limitations and
weaknesses, that must be improved in the future and that, above all, must show its use-
fulness and applicability in sociological research. As a result, subsequent studies may
suggest the substitution of certain emotional states for others that offer greater levels of
validity and reliability, particularly in the power dimension, which is the dimension that
has presented the most methodological problems. In addition, it will be necessary in the
future to analyse the measurement invariance in different cultural contexts (Davidov et al.
2008); and the existing relationships between emotional and cognitive measurement
models should be studied in-depth. The fact that the metatheoretical capture represented in
Fig. 1 (the theoretical-conceptual definition), and the empirical capture, represented in
Fig. 5 (analytical-empirical), do not exactly match, makes it necessary to carry out a deep
critique. In science, both representations of reality should configure, through dialogue
between them, a productive critical symbiosis capable of producing new knowledge
(Bericat 2012a). In this sense, we think that the two definitions of socioemotional well-
being, although different, are also compatible and consistent, and can be the basis to
develop our social scientific knowledge of subjective well-being in society.

Fig. 5 SEWBI: index, sub-indexes and emotional states

123
The Socioemotional Well-Being Index (SEWBI) 621

6.1 The Social Structure and Stratification of Socioemotional Well-Being

We end this article again emphasising that the index has been conceived with the aim of
serving as an instrument for the analysis of the social structure and stratification of
socioemotional well-being. Although some initial studies have already been carried out
comparing, for example, the degree of socioemotional well-being among married men and
women, based on the partners’ social class (Bericat 2014a), and the degree of socioemo-
tional well-being of working women, based on their employment situation (Bericat 2014b),
the sociological applications of the index are much broader, being limited only by the
availability of empirical data and the creativity of the analyst.
Given that the selection of emotional states included in the measurement model has
been based on socio-structural theories of emotion—fundamentally by theories that
incorporate social power and status as key dimensions (Turner and Stets 2006)—the index
constitutes a measure of socioemotional well-being specifically and exclusively linked to
the different positions that individuals occupy in the social structure. Although the unit of
analysis of the index is the individual, and therefore it is primarily an index of individual
and subjective well-being, it is important to emphasise that both the theoretical framework
which inspires it and the emotional states that configure its content, give it an essentially
social nature.
Just as with the classic scales measuring satisfaction and happiness commonly used in
sociological analysis, the index provides a quantitative measurement of the emotional
content of subjective well-being. However, the SEWBI, in contrast with these univariate
scales (Huppert et al. 2010:13; Huppert and So 2013; Michaelson et al. 2009:55), results
from a much more robust and richer measurement model. Its hierarchical structure (Fig. 5)
permits the development of a programme of analysis that combines three levels of different
and complementary information: (a) the global score for the composite index; (b) the score
for each of its four dimensions; and (c) the degree to which a given group of individuals
experience certain emotional states. While the information offered by a scale of satisfac-
tion or happiness constitutes a type of black box that impedes the researcher from going
further into the study of subjective well-being, the measurement model of this index
permits the analysis of both the combination of values of the basic dimensions of socio-
emotional well-being, as well as the structure of the emotional content beneath a specific
subjective state of well-being.
The analytical model of the SEWBI permits the development of a research programme
on socioemotional well-being in the context of the very diverse personal, social, economic,
political and cultural conditions we live under. The current inability of societies to guar-
antee on-going material development and the increase in living standards, leads the social
sciences to take on the challenge of broadening our knowledge of subjective well-being. In
the context of today’s crisis-ridden and uncertain hyper-developed, postmodern, con-
sumerist, globalised societies, social scientists must analyse, with creativity and rigour,
how individuals feel, how they perceive and evaluate their well-being, and how they adapt
emotionally to different contexts and circumstances. Above all, we must investigate the
social and individual dynamics established between objective conditions and subjective
experiences.
In short, one of the first projects that this general research programme should adopt, is
the study of the social stratification of emotional well-being. Collins (2004: 180–183)
suggests that there is an unequal distribution of emotional energy. It is clear that not all
members of a society enjoy the same level of emotional well-being, nor do all live their

123
622 E. Bericat

lives experiencing the same emotional states: some enjoy life more than others, some
suffer more than others. Sennett and Cobb (1972) carried out an excellent study on the
emotional experiences of working class individuals. The title of the book is sufficiently
telling: The Hidden Injuries of Class. Within this research programme, we can also address
the necessary development of a sociology of suffering (Plummer 2012; Wilkinson 2005).
We know that people suffer, but we must study who suffers, how much and why.

Appendix

See Tables 13, 14, 15, 16, and 17.

Table 13 Socioemotional well-being (SEWBI) of European population, by sex, age, educational level, and
main activity
Weighted Dimensions

n Status Situation Self Power SEWBI

Sex
Male 15,406 12.3 0.3 6.2 7.2 6.5
Female 16,755 -11.2 -0.3 -5.7 -6.6 -6.0
Age
15–24 4,697 6.2 18.4 7.8 -11.1 5.3
25–34 4,737 5.9 8.6 4.6 -8.6 2.6
35–44 6,039 5.5 1.6 0.3 -3.4 1.0
45–54 5,752 0.4 -5.9 -3.1 3.2 -1.4
55–64 4,868 -0.9 -7.3 -4.8 8.1 -1.2
65–74 3,519 -9.8 -5.4 -2.1 11.4 -1.5
75–84 2,012 -20.7 -19.0 -5.6 4.4 -10.2
85 and ? 555 -22.1 -17.8 -0.3 12.3 -7.0
Educational level
Less lower secondary 4,769 -20.5 -7.9 -8.3 -4.6 -10.3
Lower secondary 6,210 -5.1 1.0 -4.6 -5.4 -3.5
Upper secondary 12,317 1.8 -3.0 1.8 6.1 1.7
Advance vocational 2,287 9.2 3.8 3.9 6.9 5.9
Tertiary 6,460 13.9 9.3 5.6 -5.2 5.9
Others 137 -22.5 -5.3 7.4 -9.2 -7.4
Main activity
Paid work 16,530 10.8 4.9 4.4 -0.5 4.9
Education 2,833 5.3 18.5 8.0 -11.6 5.1
Unemployed 1,602 -24.2 -22.9 -11.4 1.4 -14.3
Permanently sick or disabled 803 -60.3 -36.6 -42.7 -36.6 -44.1
Retired 6,586 -9.5 -8.7 -3.6 10.7 -2.8
Housework 3,344 -11.2 -3.6 -5.1 -1.6 -5.4
Total 32,180 0.0 0.0 0.0 0.0 0.0
Source ESS (2006)

123
The Socioemotional Well-Being Index (SEWBI) 623

Table 14 Status dimension: emotional states of workers from 15 to 65 years of age, by sex, in Europe
Status Male (%) Female (%) Total (%)

Felt sad None or almost none of the time 64.2 50.6 58.1
How often past week Some of the time 32.8 43.3 37.5
Most of the time 2.2 4.3 3.1
All or almost all of the time 0.8 1.8 1.2
Felt depressed None or almost none of the time 68.8 58.7 64.3
How often past week Some of the time 27.0 34.6 30.4
Most of the time 3.0 4.8 3.8
All or almost all of the time 1.1 1.9 1.5
Felt lonely None or almost none of the time 76.5 71.5 74.3
How often past week Some of the time 19.3 22.8 20.9
Most of the time 3.0 3.8 3.4
All or almost all of the time 1.1 1.9 1.5
Source ESS (2006)

Table 15 Situation dimension: emotional states of workers from 15 to 65 years of age, by sex, in Europe
Situation Male (%) Female (%) Total (%)

Enjoyed life None or almost none of the time 3.7 4.4 4.0
How often past week Some of the time 24.2 26.1 25.0
Most of the time 46.0 44.4 45.3
All or almost all of the time 26.1 25.1 25.7
Were happy None or almost none of the time 3.1 3.7 3.3
How often past week Some of the time 23.0 24.6 23.7
Most of the time 49.7 47.3 48.6
All or almost all of the time 24.3 24.4 24.3
Source ESS (2006)

Table 16 Personal (self) dimension: emotional states of workers from 15 to 65 years of age, by sex, in
Europe
Self Male (%) Female (%) Total (%)

In general, feel very Agree strongly 17.8 14.3 16.2

positive about myself Agree 64.3 60.9 62.7
Neither agree nor disagree 12.8 15.8 14.1
Disagree 4.5 7.8 6.0
Disagree strongly 0.6 1.3 0.9
Always optimistic Agree strongly 15.7 13.2 14.6
about my future Agree 56.7 52.4 54.8
Neither agree nor disagree 17.3 20.3 18.7
Disagree 8.9 12.6 10.6
Disagree strongly 1.3 1.5 1.4
Source ESS (2006)

123
624 E. Bericat

Table 17 Power dimension: emotional states of workers from 15 to 65 years of age, by sex, in Europe
Power Male (%) Female (%) Total (%)

Felt rested None or almost none of the time 16.3 22.3 19.0
When woke up in morning, Some of the time 34.1 34.6 34.3
how often past week
Most of the time 36.0 31.6 34.0
All or almost all of the time 13.6 11.5 12.7
Felt calm and peaceful None or almost none of the time 6.7 11.6 8.9
How often past week Some of the time 31.5 36.0 33.5
Most of the time 46.7 40.7 44.0
All or almost all of the time 15.1 11.7 13.5
Had lot of energy None or almost none of the time 6.0 9.5 7.6
How often past week Some of the time 31.8 34.0 32.8
Most of the time 44.7 41.3 43.2
All or almost all of the time 17.5 15.1 16.4
Source ESS (2006)

References

Albright, J. J., & Park, H. M. (2009). Confirmatory factor analysis using Amos, LISREL, Mplus, and SAS/
STAT CALIS. Working paper. The University Information Technologies Services (UITS) Center for
Statistical and Mathematical Computing, Indiana University.
Arbuckle, J. L. (2010). IBM SPSS Amos 19. User’s guide. Chicago: SPSS.
Batista-Foguet, J. M., Coenders, G., & Alonso, J. (2004). Análisis factorial confirmatorio. Su utilidad en la
validación de cuestionarios de Salud. Medicina Clı́nica, 122(Suppl 1), 21–27.
Bericat. (2012a). The European Gender Equality Index: Conceptual and analytical issues. Sociol Indicators
Research, 108, 1–28.
Bericat. (2012b). ‘‘Emotions’’. Sociopedia, 2–13, International Sociological Association (ISA). http://www.
isa-sociology.org/publ/sociopedia-isa/sociopedia-isa-list-of-published-entries.htm.
Bericat, E. (2014a). Matrimonio, desigualdad de género y bienestar socioemocional de los miembros de la
pareja. In A. Garcı́a y O. Sabido (Eds.), Cuerpo y afectividad en la sociedad contemporánea. Algunas
rutas del amor y de la experiencia sensible en las ciencias sociales contemporáneas. Mexico: UAM-A
(in press).
Bericat, E. (2014b). The subjective well-being of working women in Europe. In M. Connerley, & J. Wu
(Eds.), Handbook on well-being of working women. The Quality of Life Research Series. Springer (in
press).
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Bradburn. (1969). The structure of psychological well-being. Chicago: Aldine.
Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment. Thousand Oaks: Sage.
Cea D’Ancona, M. A. (2002). Análisis multivariable. Teorı́a y práctica en la investigación Social. Madrid:
Sı́ntesis.
CIS. (2011). Estudio no. 2923. Centro de Investigaciones Sociológicas. http://www.cis.es/cis/opencm/ES/1_
encuestas/estudios/ver.jsp?estudio=12104&cuestionario=14278&muestra=19819.
CMEPSP. (2009). Report by the Commission on the measurement of economic performance and social
progress. www.stiglitz-sen-fitoussi.fr/documents/rapport_anglais.pdf.
Collins, R. (1981). On the microfoundations of macrosociology. American Journal of Sociology, 86(5),
984–1014.
Collins, R. (1990). Stratification, emotional energy, and the transient emotions. In Th. D. Kemper (Ed.),
Research agendas in the sociology of emotions (pp. 27–57). Albany, NY: State University NY Press.
Collins, R. (2004). Interaction ritual chains. Princeton, NJ: Princeton University Press.
Davidov, E., Schmidt, P., & Schwartz, S. (2008). Bringing values back in the adequacy of the European
Social Survey to measure values in 20 countries. Public Opinion Quarterly, 72(3), 420–445.
De Smedt, M. (2013). Measuring subjective well-being in the European Satatistical System (ESS). Social
Indicators Research, 114(1), 153–167.

123
The Socioemotional Well-Being Index (SEWBI) 625

Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542–575.

Diener, E. (1994). Assessing subjective well-being: Progress and opportunities. Social Indicator Research,
31, 103–157.
Diener, E., Inglehart, R., & Tay, L. (2012). Theory and Validity of Life Satisfaction Scales. Social Indi-
cators Research,. doi:10.1007/s11205-012-0076-y.
Diener, E., & Lucas, R. E. (1999). Personality and subjective well-being. In D. Kahneman, E. Diener, & N.
Schwarz (Eds.), Well-being: The foundations of hedonic psychology. New York: Russell Sage
Foundation.
Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: Three decades of
progress. Psychological Bulletin, 125(2), 276–302.
Diener, E., Wirtz, D., Tov, W., Prieto, Ch., Choi, D., Oishi, Sh., et al. (2010). New well-being measures:
Short Scales to asses flourishing and positive an negative feeling. Social Indicators Research, 97,
143–156.
Durkheim, E. (1951). The suicide. A study of Sociology. New York: The Free Press.
Easterlin, R. A. (1974). Does economic growth improve the human lot? Some empirical evidence. In P.
A. David & M. W. Reder (Eds.), Nations and households in economic growth: Essays in honor of
Moses Abramowitz (pp. 89–125). New York: Academic Press.
ESS. (2006). Eurpean Social Survey, ESS3-2006 Edition 3.3. http://ess.nsd.uib.no/ess/round3/.
Field, A. (2000). Discovering statistics using SPSS for windows. London: Sage Publications.
Freud, S. (1948). Inhibitions, symptoms, and anxiety. London: Hogarth Press.
Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. Chicago: Aldine.
Gorsuch, R. L. (1983). Factor analysis. Hillsdale: Lawrence Erlbaum Associates.
Halleröd, B., & Seldén, D. (2012). The multi-dimensional characteristics of wellbeing: How different
aspects of wellbeing interact and do not interact with each other. Social Indicators Research. doi:10.
1007/s11205-012-0115-8.
Hicks, S., Tinkler, L., & Allin, P. (2013). Measuring subjective well-being and its potencial role in policy:
Perspectives from the UK office for national statistics. Social Indicators Research, 114(1), 73–86.
Hochschild, A. R. (1983). The managed heart: The commercialization of human feeling. Berkeley: Uni-
versity of California Press.
Huppert, F. A., Andrew, C., Frey, B., Marks, N., & Siegrist, J. (2005). Personal and social well-being:
Creating indicators for a flourishing Europe. http://www.europeansocialsurvey.org/index.php?option=
com_content&task=view&id=220&Itemid=309.
Huppert, F. A., Marks, N., Siegrist, J., Vázquez, C., & Vitterso, J. (2010). Personal and social well-being.
http://www.europeansocialsurvey.org/index.php?option=com_content&view=article&id=321&Itemid=
388.
Huppert, F. A., & So, T. T. (2013). Flourishing across Europe: Application of a new conceptual framework
for defining well-being. Social Indicators Research, 110, 837–861.
Kemper, Th. D. (1978). A social interactional theory of emotions. New York: Wiley.
Kemper, Th. D. (1990). Social relations and emotions: A structural approach. In Th. D. Kemper (Ed.),
Research agendas in the sociology of emotions (pp. 207–237). Albany, NY: State University NY Press.
Kemper, Th. D. (2006). Power and status and the power-status theory of emotions. In J. E. Stets & J.
H. Turner (Eds.), Handbook of the sociology of emotions. Springer Science: Boston, MA.
Kemper, Th. D., & Collins, R. (1990). Dimensions of microinteraction. The American Journal of Sociology,
96(1), 32–68.
Kline, R. B. (2011). Principles and practice of structural equation modeling. New York: The Guilford Press.
Krueger, A. B. (Ed.). (2009). Measuring subjective well-being of nations: National accounts of time use and
well-being. Chicago: University of Chicago Press.
Lhéritier, J. L. (2012). Measuring well-being in France. In H.-H. Noll (Ed.), Social monitoring and reporting
in Europe 2012. Subjective Indicators: Usefulness and Information Potential for Policy Making.
Long, J. S. (1983). Confirmatory factor analysis. Newbury Park: Sage.
Michaelson, J., Abdallah, S., Steuer, N., Thompson, S., & Marks, N. (2009). National accounts of well-
being: Bringing real wealth onto the balance sheet. London:NEF. http://www.unicef.org/lac/National_
Accounts_of_Well-being(1).pdf.
Nardo, M., Saisana, M., Saltelli, A., Tarantola, S., Hoffman, A., & Giovannini, E. (2005). Handbook on
constructing composite indicators: Methodology and user guide. Paris: OECD Statistics Working
Paper.
Noll, H.-H. (2013). Subjective social indicators: Benefits and limitations for policy making—An intro-
duction to this special issue. Social Indicators Research, 114(1), 1–11.
Nussbaum, M. C. (2001). Upheavals of thought. The intelligence of emotions. New York: Cambridge
University Press.

123
626 E. Bericat

OECD. (2012). European conference on measuring well-being and fostering the progress of societies. http://
www.oecd.org/site/progresseurope/europeanconferenceagenda.htm.
ONS. (2011). Measuring national well-being. Measuring what Matters. New Port: ONS. http://www.ons.
gov.uk/ons/guide-method/user-guidance/well-being/publications/measuring-what-matters-national-
statistician-s-reflections-on-the-national-debate-on-measuring-national-well-being.pdf.
Oswald, A. J. (2010). Emotional prosperity and the Stiglitz Commission. British Journal of Industrial
Relations, 48(4), 651–669.
Plummer, K. (2012). A manifesto for a critical humanism in sociology: on questioning the human social world.
http://kenplummer.wordpress.com/manifestos/a-manifesto-for-a-critical-humanism-in-sociology/.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
Scheff, Th. J. (1988). Shame and conformity: The deference-emotion system. American Sociological
Review, 53, 395–406.
Scheff, Th. J. (1990). Microsociology: Discourse, emotion and social structure. Chicago: The University of
Chicago Press.
Scheff, Th. J. (2000). Shame and the social bond: A sociological theory. Sociological Theory, 18, 84–99.
Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. New York: W.
H. Freeman.
Sennett, R., & Cobb, J. (1972). The hidden injuries of class. New York: W.W. Norton & Company.
Stets, J. E. (2010). Future directions in the sociology of emotions. Emotion Review, 2(3), 265–268.
Tay, L., Chan, D., & Diener, E. (2013). The metrics of societal happiness. Social Indicators Research.
doi:10.1007/s11205-013-0356-1.
Tinkler, L., & Hicks, S. (2011). Measuring subjective well-being. Office for National Statistics (ONS).
http://www.ons.gov.uk/ons/guide-method/user-guidance/well-being/publications/measuring-subjective-
well-being.pdf.
Turner, J. H., & Stets, J. E. (2005). The sociology of emotions. New York: Cambridge University Press.
Turner, J. H. & Stets, J. E. (2006). Sociological theories of human emotions. Annual Review of Sociology,
32, 25–52.
Veenhoven, R. (1968). Geluk als onderwerp van wetenschappelijk onderzoek [Happiness as a subject of
scientific research]. Sociologische Gids, 17, 115–122.
Veenhoven, R. (1984). Conditions of happiness. Dordrecht: Reidel.
Veenhoven, R. (2008). Sociological theories of subjective well-being. In M. Eid & R. Larsen (Eds.), The
science of subjective well-being: A tribute to Ed Diener (pp. 44–61). New York: Guilford Publications.
Viterso, J. (2004). Subjective well-being versus self-actualization: Using the flow-simplex to promote a
conceptual clarification of subjective quality of life. Social Indicator Research, 65, 299–331.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive
and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.
doi:10.1037/0022-3514.54.6.1063.
Weber, M. (1978). Economy and society. Berkeley: University of California Press.
Wilkinson, I. (2005). Suffering: A sociological introduction. Cambridge: Polity Press.
Wilson, W. (1967). Corrrelates of avowed happiness. Psychological Bulletin, 67, 294–306.
Wu, C. H., & Yao, G. (2007). Examining the relationship between global and domain measures of quality of
life by three factor structure models. Social Indicators Research, 84(2), 189–202.

123
View publication stats