Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Personality and Individual Differences 31 (2001) 1357±1364

www.elsevier.com/locate/paid

Emotional stability as a major dimension of happiness


Peter Hills *, Michael Argyle
The Oxford Happiness Project, School of Psychology, Oxford Brookes University, Headington Campus, Gipsy Lane,
Oxford OX3 0BP, UK

Received 19 June 2000; received in revised form 21 November 2000; accepted 30 November 2000

Abstract
Happiness is associated with both extraversion and neuroticism, and extraversion is generally considered
the more important. A recent study of happy introverts has shown that extraversion is not always an
essential correlate of happiness, and an extensive meta-analysis has found that neuroticism is a greater
predictor of both happiness and life satisfaction. It is suggested that the reason for the importance of
neuroticism having been overlooked in the past, is the diculty of handling the idea that (positive) happi-
ness is related to the absence of a (negative) construct. This diculty could be resolved by the reversal of
neuroticism into an alternative and positive concept of ``emotional stability''. Happiness could then be
regarded as being associated with two positive qualities. With this change of emphasis, a short empirical
study has been made of the relationships between happiness as measured by the Oxford Happiness Inven-
tory (OHI) and extraversion and emotional stability. In bivariate and partial correlation, emotional stabi-
lity was more strongly associated with happiness than extraversion, and accounted for more of the total
variability in multiple regression. Emotional stability was also the greater correlate for a majority of the 29
items of the OHI, and the sole signi®cant predictor of the happiness of younger people. # 2001 Elsevier
Science Ltd. All rights reserved.
Keywords: Emotional stability; Extraversion; Satisfaction with life; Self-esteem; Psychological well being; Dimensions
of happiness

1. Introduction

Psychological well-being, or happiness, is a multidimensional construct comprising emotional


and cognitive elements. The basic framework was established by Bradburn (1969) who oper-
ationalised well-being in terms of separate positive and negative ``a€ects'', which are an amalgam
of the feelings, moods and emotional responses of individuals to the variety of pleasant and

* Corresponding author. Tel.: +44-1235-521-077; fax: +44-1235-520-067.


E-mail address: p_r_hills@hotmail.com (P. Hills).

0191-8869/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0191-8869(00)00229-4
1358 P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364

unpleasant events which make up normal life. According to Bradburn ``an individual will be high
in psychological well-being in the degree to which he has an excess of positive over negative a€ect
and will be low in well-being in the degree to which negative a€ect predominates over positive''
(p. 9). Subsequently Andrews and Withey (1976) showed that well-being could better be repre-
sented by the addition of a third, cognitive-evaluative element, life satisfaction, of which self-
esteem (Maslow, 1970), a sense of personal control (Rotter, 1966), optimism (Scheier & Carver,
1985) and goal ful®lment (Headey & Wearing, 1981) are some speci®c aspects.
The idea that the positive and negative moods associated with well-being are unrelated is
counter-intuitive; it is dicult to visualise being happy and unhappy at one and the same time.
Nevertheless, a range of studies has provided evidence in support of the independence of positive
and negative a€ect (for example Bradburn & Caplovitz, 1965; Diener & Emmons, 1985; Watson
& Clark, 1997), and this view has been accepted by the US National Advisory Mental Health
Council (1995)1. That happiness and its component parts are related di€erentially and indepen-
dently to the personality traits of extraversion and neuroticism has been extensively documented
since Costa and McCrae (1980) demonstrated that positive a€ect correlates (positively) with
extraversion and that negative a€ect correlates (negatively) with neuroticism. Similar relation-
ships have been reported for composite measures of well-being such as the Oxford Happiness
Inventory (Lu & Argyle, 1991) and cognitive factors such as self-esteem and optimism (Lucas,
Diener & Suh, 1996). Costa, McCrae and Norris (1981) found that extraversion and neuroticism
data antedated and predicted di€erences in happiness 17 years later. Similar results were reported
by Headey and Wearing (1989) over a period of 7 years for an Australian sample.
In parallel with the substantial literature on the relationships between extraversion, neuroticism
and well-being, the idea has emerged that extraversion is the predominant predictor of happiness.
This would appear to be self-evident because extraversion is associated with friendship and social
activity, which are major sources of joy, happiness and personal satisfaction in both private and
public life (Campbell, Converse & Rodgers, 1976). There are other plausible reasons for this
attribution. Most of the relevant work has been conducted in the West, particularly in the United
States, where sociability and the pursuit of personal happiness are legitimate and associated cul-
tural norms. Contemporary life in the West is also highly interactive, and the ability easily to
establish friendly relationships with many people is a valued and advantageous social skill
(Diener, Larsen & Emmons, 1984). Neuroticism, on the other hand, is associated with distress,
depression and despair, and these states are the antithesis of happiness. The construct of neuro-
ticism used in personality theory originated from a clinical measure of mental instability used for
the diagnosis of psychological disorder. In a society that is conditioned to ``accentuate the posi-
tive'', it is not intuitively easy to grasp the idea that happiness is related as much to the absence of
negative a€ect (neuroticism), commonly associated with mental illness, as to the presence of
positive a€ect (extraversion).
In order to resolve this diculty in studies of well-being, it is now suggested that the concept of
neuroticism should be reversed and termed ``emotional stability'', so that greater emotional sta-

1
This view is not universally accepted. Several workers, notably Russell and Caroll (1999), have argued that the
observed independence of positive and negative a€ects is an artefact created by inherent measurement error, and have
reported alternative evidence which supports theoretical predictions based on bipolarity.
P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364 1359

bility would be regarded as a positive aspect of personality. Emotionally stable people could be
expected to be calm, imperturbable and to complain little about their personal worries and
anxieties. With this modi®cation, happiness could be regarded as a relationship with two positive
constructs. Which is the more important?
The idea that extraversion is the predominant predictor of happiness, is not fully consistent
with the ideas of those classical philosophers such as Aristotle and Epicurus, who have given
most attention to human happiness. Both recommended living a quiet, contemplative existence in
relative solitude. The same can be said of the ways of life commended by some religious systems
that involve a degree of enclosure, and medieval hermits and anchorites believed that total isola-
tion from other people was the way to salvation and eternal happiness. It has recently been
reported (Hills & Argyle, 2001) that a substantial minority of ordinary people can be classi®ed as
happy introverts, that their reported levels of happiness are only marginally less than happy
extraverts, and that the social behaviours of happy introverts and extraverts are virtually indis-
tinguishable. In that experiment, it was also noted that emotional stability was a greater predictor
of happiness in regression than extraversion and explained more of the variability in the data
(38%) than did extraversion (23%). In a meta-analysis of the relationships between 137 distinct
personality constructs and subjective well-being, DeNeve and Cooper (1998) found that emo-
tional stability was the strongest predictor of both happiness and life satisfaction. The average
weighted correlations with overall subjective well-being were 0.22 for emotional stability (74
independent samples) and 0.17 for extraversion (82 independent samples). The corresponding
®gures for life satisfaction were 0.24 and 0.17.
The present communication describes a short and speci®c study of the relative magnitudes of
the relationships between happiness as measured by the Oxford Happiness Inventory (OH), and
extraversion and emotional stability.

2. Method

2.1. Participants

Two-hundred and forty-four residents of Oxfordshire and their friends and acquaintances (101
men and 143 women) were recruited by personal contact with a variety of leisure groups, Oxford
Brookes and Oxford Universities, and by advertisements in public places. Ages ranged from 18 to
85 (M=45.9, S.D.=17.3) years. The respondents were mainly professional; 73% were graduates
or undergraduates, 47% were employed or self-employed and 21% retired. Most (64%) were
living with a partner.

2.2. Measures

Participants completed a revised version of the OHI (Argyle, Martin & Lu, 1995); a broad 29-
item measure designed to re¯ect the major dimensions of subjective well-being. Factor analysis
(Hills & Argyle, 1998b) identi®ed these dimensions, in decreasing order of magnitude, as satis-
faction with life, personal ecacy, sociability/empathy, a positive outlook, physical well-being,
cheerfulness and self-esteem. Respondents are invited to select from one of four possible graded
1360 P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364

choices for each item, and the aggregate score forms the measure of overall happiness. Higher
scores indicate greater happiness. In the work now described, the OHI demonstrated an accep-
table scale reliability with a Cronbach of 0.90 (n=240). The items that comprise the scale are
indicated in Table 2. This measure was supplemented by the extraversion and neuroticism sub-
scales of the short-scale form of the revised Eysenck Personality Questionnaire (Eysenck, Eysenck
& Barrett, 1985), Rosenberg's Self-esteem Scale (Rosenberg, 1989) and the Satisfaction with Life
scale (Diener, Emmons, Larsen & Grin, 1985). The Cronbach s recorded for the Self-esteem
and Satisfaction with Life scales in this study were respectively 0.86(243) and 0.89(242). With the
exception of the OHI, items in the original scales were re-worded where necessary as single
statements to which participants could respond on a 6-point scale ranging from ``strongly agree''
to ``strongly disagree''. Before administration, the individual items of these scales were combined
and rearranged in random order to minimise contextual answering (Hills & Argyle, 1998a). The
measures were incorporated into a postal questionnaire which recipients were invited to complete
and return along with some demographic information.

3. Results and discussion

Bivariate correlation coecients were ®rst calculated between each of the principal variables
and extraversion (E) and emotional stability (ES). The correlations of the total OHI scores with E
and ES were, respectively, 0.49 and 0.53. The corresponding ®gures for satisfaction with life were
0.29 and 0.47 and for self-esteem 0.34 and 0.63. These values were signi®cant at the P<0.001
level in every instance. The correlations between E and ES and satisfaction with life are consistent
with previously reported values obtained using di€erent personality scales. Herringer (1998)
reported a correlation of 0.38 between life satisfaction and extraversion as measured by the
revised version of the NEO Personality Inventory (Costa & McCrae, 1992), and Arrindell, Hee-
sink and Feij (1999) a correlation of 0.30 between life satisfaction and neuroticism as measured
by items selected from the neuroticism subscale of the 3-Dimensional Personality Test (van
Kampen, 1996).
Linear multiple regressions were carried out both by simultaneous and stepwise entry for each
of the three principal variables as dependent variables with E and ES entered as independent

Table 1
Extraversion and emotional stability as predictors of overall Oxford Happiness Inventory (OHI) scores, satisfaction
with life and self-esteem

Dependent variable Independent variable R2 R2 change a


F

OHI Emotional stability 0.28 0.28 0.47***


Extraversion 0.44 0.16 0.40*** 92.09***
Satisfaction with life Emotional stability 0.23 0.23 0.22***
Extraversion 0.27 0.04 0.44*** 44.42***
Self-esteem Emotional stability 0.40 0.40 0.59***
Extraversion 0.45 0.05 0.24*** 97.97***
a
-Values are the standardised coecients appearing in the ®nal regression equation.
***P<0.001.
P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364 1361

variables. The regression coecients produced by both methods were identical. Table 1 presents
the results obtained by the stepwise method because this allows the relative importance of each
independent variable to be assessed from the corresponding changes in R2. All the regressions and
standardised regression coecients ( s) were signi®cant at the P<0.001 level. In each instance ES
entered the equation before E and was the larger predictor of each independent variable. The
proportions of the total variability explained by the ®nal equations accounted for by ES were 80,
91 and 94% for the total OHI scores, satisfaction with life and self-esteem, respectively.

Table 2
Correlations of the Oxford Happiness Inventory (OHI) items with extraversion (E) and emotional stability (ES)
Item Correlations Partial correlations

E ES Ea ESb

04. I feel I am in control most of the time 0.28*** 0.45*** 0.20** 0.44***
12. I usually wake up feeling rested nsc 0.40*** ns 0.41***
03. I am satis®ed with many things in my life 0.17** 0.43*** ns 0.41***
01. I am very happy 0.18** 0.41*** ns 0.39***
13. I feel very energetic 0.21*** 0.33*** ns 0.35***
06. I am very pleased with the way I am 0.29*** 0.40*** 0.24*** 0.35***
08. Life is very good 0.27*** 0.36*** 0.22*** 0.34***
05. I feel that life is very rewarding 0.22*** 0.35*** 0.15** 0.33***
18. Most past events seem to have been happy 0.24*** 0.35*** 0.20** 0.33***
15. I feel very mentally alert 0.21*** 0.32*** 0.15** 0.31***
10. I ®nd it easy to make most decisions 0.26*** 0.34*** 0.23*** 0.31***
02. I feel I have so much to look forward to 0.29*** 0.30*** 0.26*** 0.29***
20. I have done many of the things I wanted 0.28*** 0.34*** 0.23*** 0.29***
16. I feel very healthy ns 0.27*** ns 0.28***
07. I often have a good in¯uence on events 0.25*** 0.31*** 0.21*** 0.26***
24. I have a great sense of meaning and purpose 0.19** 0.28*** 0.14** 0.25***
23. I often have a cheerful e€ect on others 0.47*** 0.29*** 0.47*** 0.25***
14. I ®nd beauty in most things 0.21*** 0.27*** 00.16** 0.23***
19. I often experience joy and elation 0.34*** 0.26*** 0.29*** 0.21***
11. I ®nd it easy to do things 0.23*** 0.23*** 0.23*** 0.19**
21. I organise my time very well 0.15** 0.20** 0.15** 0.18**
25. I often become committed and involved 0.18** 0.20*** 0.15** 0.16**
26. I think the world is a very good place 0.33*** 0.23*** 0.31*** 0.16**
22. I often have fun with other people 0.51*** 0.21** 0.50*** 0.14**
29. I ®nd most things amusing 0.30*** ns 0.28*** ns
28 I think I look attractive 0.30*** 0.19** 0.29*** ns
17. I have very warm feelings towards others 0.32*** ns 0.29*** ns
09. I am very interested in other people 0.39*** ns 0.39*** ns
27. I laugh a lot 0.40*** 0.14** 0.37*** ns

*P<0.05.
**P<0.01.
***P<0.001.
a
Controlling for ES.
b
Controlling for E.
c
ns=not signi®cant.
1362 P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364

Because the OHI is a broadly based measure of happiness, it is possible to examine the asso-
ciations with E and ES in detail. Table 2 summarises the bivariate correlations of each item with
E and ES, and the corresponding partial correlations controlling for ES and E in turn. There is
not a great deal of di€erence between the corresponding individual bivariate and partial correla-
tions, which suggests that there was little covariation between E and ES. ES correlates more
strongly than E with some two-thirds, 19 out of 29, of the items. It is of special interest that the
item that deals most directly with subjective happiness, ``I am very happy'', correlates only with ES;
the bivariate association with E becomes insigni®cant when the e€ect of ES is partialled away.
In partial correlation, there is a signi®cant association with ES, and not E, for ®ve items. All
these items are associated with a sense of individual well being: feeling healthy, energetic, rested,
in control and being pleased with self. An equal number of items are associated with E and not
ES and these include some but not all of the exclusively social items, for example ``I am interested
in other people'' and ``I have warm feelings towards others''. Although the two most clearly
social items, ``I often have fun with other people'' and ``I often have a cheerful e€ect on others'',
are strongly associated with E, they are still associated with ES to a smaller degree. Overall, these
observations do not fully support the thesis that extraversion is the principal predictor of happiness.
The participants in this study were largely mature professional adults; their average age was 46,
some two-thirds were employed or retired and a similar proportion said they were living in a
stable relationship. Such people might be particularly likely to exhibit and endorse stability and
this could be the reason for the associations reported above being greater for ES than for E.
Extraversion and happiness might be more strongly associated for a younger sample. It has been
pointed out elsewhere (Hills & Argyle, 2000) that the items used to measure extraversion in the
Eysenck Personality Questionnaire mostly assess gregariousness. Younger people, who may not
have yet formed stable relationships, are more likely to take part in highly social and gregarious
activities. Younger people may also not yet have achieved fully adult stability; Costa and McCrae
(1994) consider that personality development continues until the age of 25 or 30, or even later.
Extraversion could still be the dominant predictor of happiness for a younger sample.
Table 3 compares the simultaneous entry regression data for participants aged 30 and above
with those aged less than 30. The results were completely unexpected. For the older sample, ES
was the greater predictor of happiness by a narrow margin. But for the younger sample, ES was a
stronger predictor of happiness and E did not achieve signi®cance!

Table 3
Oxford Happiness Inventory Regression data for older and younger participants

Independent variable F

Participants aged >29 (n=186)


Emotional stability 0.455***
Extraversion 0.449*** 84.75***
Participants aged <30 (n=50)
Emotional stability 0.503***
Extraversion 0.164 nsa 11.65***
a
ns, not signi®cant.
***P<0.001.
P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364 1363

4. Summary and conclusions

That both extraversion and neuroticism are correlates of happiness has long been known and
extraversion is usually considered as the predominant predictor of happiness. Nevertheless, a
recent meta-study of publications over the last 30 years (DeNeve & Cooper, 1998) has produced
evidence that neuroticism is the greater (negative) predictor of both happiness and life satisfac-
tion. The reason for the in¯uence of neuroticism having been overlooked could be found in the
relative conceptual ease of accounting for the associations of extraversion and neuroticism with
happiness. It seems instinctively natural to associate the positive state of happiness with extra-
version which is also regarded positively, particularly in Western society, but less natural to
associate happiness with the absence of a negative state which is more commonly associated with
depression and mental illness. We have suggested that were the concept of neuroticism reversed
and termed ``emotional stability'', it would become easier to handle; happiness could then be
understood in terms of a relationship with two positive and socially desirable constructs.
With this change of emphasis, a short study has been made of the relative importance of
extraversion and emotional stability, using the OHI to measure happiness. The conclusions from
this study are:
1. Emotional stability correlates more strongly with overall happiness, satisfaction with life and
self-esteem than does extraversion.
2. In multiple regression, emotional stability is a stronger predictor of overall happiness, satis-
faction with life and self-esteem than extraversion, and explains more of the variability in
each regression than does extraversion.
3. Emotional stability is a greater correlate and partial correlate than extraversion of a major-
ity of the individual items of the OHI. For ®ve items that relate to personal and internal
sources of happiness, emotional stability is a strong and signi®cant correlate whereas the
e€ect of extraversion does not reach signi®cance.
4. For younger participants aged <30, emotional stability is a substantial and signi®cant pre-
dictor of overall happiness and extraversion is not.
It is hoped that these results will lead to a broader understanding of the nature of well-being,
by suggesting that emotional stability is at least as important as extraversion in the consideration
of human happiness.

References

Andrews, F., & Withey, S. (1976). Social indicators of well-being. New York: Plenum Press.
Argyle, M., Martin, M., & Lu, L. (1995). Testing for stress and happiness: The role of social and cognitive factors.
In C. Spielberger, & I. Sarason, Stress and Emotion, 15 (pp. 173±187). Washington, DC: Taylor & Francis.
Arrindel, W., Heesink, J., & Feij, J. (1999). The satisfaction with life scale (SWLS): Appraisal with 1700 healthy young
adults in The Nederlands. Personality and Individual Di€erences, 26, 815±826.
Bradburn, N. (1969). The structure of psychological well-being. Chicago: Aldine.
Bradburn, N., & Caplovitz, D. (1965). Reports on happiness: A pilot study. Chicago: Aldine.
Campbell, A., Converse, P., & Rodgers, W. (1976). The quality of American life. New York: Sage.
Costa, P., & McCrae, R. (1980). In¯uence of extraversion and neuroticism on subjective well-being: Happy and
unhappy people. Journal of Personality and Social Psychology, 38, 668±678.
1364 P. Hills, M. Argyle / Personality and Individual Di€erences 31 (2001) 1357±1364

Costa, P., & McCrae, R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO Five-factor inventory
(NEO-FFI) manual. Odessa, FL: Psychological Assessment Services.
Costa, P., & McCrae, R. (1994). Set like plaster? Evidence for the stability of adult personality. In T. Heatherton, & G.
Weinberger, Can personality change? Washington DC: American Psychological Association.
Costa, P., McCrae, R., & Norris, A. (1981). Personal adjustment to aging: Longitudinal prediction from neuroticism
and extraversion. Journal of Gerontology, 36, 78±85.
DeNeve, K., & Cooper, H. (1998). The happy personality: A meta-analysis of 137 personality traits and subjective well-
being. Psychological Bulletin, 124, 197±229.
Diener, E., & Emmons, R. (1985). The independence of positive and negative a€ect. Journal of Personality and Social
Psychology, 47, 1105±1117.
Diener, E., Emmons, R., Larsen, R., & Grin, S. (1985). The satisfaction with life scale. Journal of Personality
Assessment, 49, 71±75.
Diener, E., Larsen, R., & Emmons, R. (1984). Personsituation interactions: Choice of situations and congruence
response models. Journal of Personality and Social Psychology, 47, 580±592.
Eysenck, S., Eysenck, H., & Barrett, P. (1985). A revised version of the psychoticism scale. Personality and Individual
Di€erences, 6, 21±29.
Headey, B., Wearing, A. (1981), Australians' priorities, satisfactions and well-being. Monograph in Public Policy Stu-
dies, No. 4, Melbourne: University of Melbourne.
Headey, B., & Wearing, A. (1989). Personality, life events, and subjective well-being: Toward a dynamic equilibrium
model. Journal of Personality and Social Psychology, 57, 731±739.
Herringer, L. (1998). Facets of extraversion related to life satisfaction. Personality and Individual Di€erences, 24, 731±
733.
Hills, P., & Argyle, M. (1998a). Musical and religious experiences and their relationship to happiness. Personality and
Individual Di€erences, 25, 91±102.
Hills, P., & Argyle, M. (1998b). Positive moods derived from leisure and their relationship to happiness and person-
ality. Personality and Individual Di€erences, 25, 523±535.
Hills, P., & Argyle, M. (2001). Happiness, introversion-extraversion and happy introverts. Personality and Individual
Di€erences, (in press)..
Lu, L., & Argyle, M. (1991). Happiness and cooperation. Personality and Individual Di€erences, 12, 1019±1030.
Lucas, R., Diener, E., & Suh, E. (1996). Discriminant validity of well-being measures. Journal of Personality and Social
Psychology, 71, 616±628.
Maslow, A. (1970). Motivation and personality. New York: Harper & Row.
National Advisory Mental Health Council (1995). Basic behavioral science research for mental health, a national
investment: Emotion and motivation. American Psychologist, 50, 838±845.
Rosenberg, M. (1989). Society and the adolescent self-image. Middletown CT: Wesleyan University Press.
Rotter, J. (1966). Generalised expectancies for internal versus external control of reinforcement. Psychological Mono-
graphs, 80, 1 Whole No. 609.
Russell, J., & Caroll, J. (1999). On the bipolarity of positive and negative a€ect. Psychological Bulletin, 125, 3±30.
Scheier, M., & Carver, C. (1985). Optimism, coping and health: Assessment and implications of generalised outcome
expectancies. Health Psychology, 4, 219±247.
van Kampen, D. (1996). Basic personality factors from a clinical/theoretical perspective. Doctoral Dissertation, Vrije
Universitet, Amsterdam.
Watson, D., & Clark, L. (1997). Measurement and mismeasurement of mood: Recurrent and emergent issues. Journal
of Personality Assessment, 67, 267±296.

You might also like