The Oxford Happiness Questionnaire: a compact scale
for the measurement of psychological well-being
Peter Hills*, Michael Argyle
The Oxford Happiness Project, School of Psychology, Oxford Brookes University, Headington Campus,
Gipsy Lane, Oxford OX3 0BP, UK
Received 2 May 2001; received in revised form 11 November 2001; accepted 9 December 2001
An improved instrument, the Oxford Happiness Questionnaire (OHQ), has been derived from the
Oxford Happiness Inventory, (OHI). The OHI comprises 29 items, each involving the selection of one of
four options that are different for each item. The OHQ includes similar items to those of the OHI, each pre-
sented as a single statement which can be endorsed on a uniform six-point Likert scale. The revised instrument
iscompact,easy to administer and allows endorsements over anextended range. When tested against the OHI,
the validity of the OHQ was satisfactory and the associations between the scales and a battery of personality
variables known to be associated with well-being, were stronger for the OHQ than for the OHI. Although
parallel factor analyses of OHI and the OHQ produced virtually identical statistical results, the solution for
the OHQ could not be interpreted. The previously reported factorisability of the OHI may owe more to the
way the items are formatted and presented, than to the nature of the items themselves. Sequential ortho-
gonal factor analyses of the OHQ identified a single higher order factor, which suggests that the construct
of well-being it measures is uni-dimensional. Discriminant analysis has been employed to produce a short-
form version of the OHQ with eight items. # 2002 Published by Elsevier Science Ltd.
Keywords: Subjective well-being; Oxford Happiness Inventory; Oxford Happiness Questionnaire; Measurement of
The Oxford Happiness Inventory (OHI, Argyle, Martin, & Crossland, 1989) was devised as a
broad measure of personal happiness, mainly for in-house use in the Department of Experimental
Psychology of the University of Oxford in the late 1980s. The development of the scale and some
of its properties were reviewed by Argyle, Martin, and Lu (1995). The scale has been found to
behave consistently, and other workers have reported its use both in the UK (Furnham & Brewin,
1990, Joseph & Lewis, 1998), in Spain (Sanchez, 1994) and the USA (Valiant, 1993). The OHI has
0191-8869/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
Personality and Individual Differences 33 (2002) 1073–1082
* Corresponding author. Tel.: +44-1235-521-077; fax: +44-1235-520-067.
E-mail address: email@example.com (P. Hills).
also been used cross-culturally to compare students in Australia, Canada, the UK and USA
(Francis, Brown, Lester, & Philipchalk, 1998). A Hebrew translation has been applied in Israel
(Francis & Katz, 2000) and it forms the basis of the Chinese Happiness Inventory (CHI) which
has been used in Taiwan (Lu & Shih, 1997).
The OHI follows the design and format of the Beck Depression Inventory (BDI, Beck, Ward,
Mendelson, Hock, & Erbaugh, 1961) which provided, when reversed, a set of 20 multiple-choice
items relevant to subjective well-being. Further items were added to cover aspects of happiness
which were not otherwise included and 29 items were retained in the final scale. Each item was
presented in four incremental levels, numbered from 0 to 3, for example:
I am not particularly optimistic about the future1
I feel optimistic about the future.
I feel I have so much to look forward to.
I feel that the future is overflowing with hope and promise.
The BDI was designed for clinical application with the purpose of diagnosing manic and
depressive states of mind. In non-clinical populations, few are manic or depressive and the
extremes of the corresponding OHI item alternatives are little used. In practice, ‘‘normal’’ parti-
cipants mainly endorse one or other of the two central items. For a substantial minority of items
the mean scores are less than, or do not comfortably exceed, their corresponding standard
deviations. This suggests that answers to these items may be uniformly, rather than normally
distributed, and might not be making their full contribution to the measurement of happiness.
The statistical properties of the individual items would be improved if respondents could select
answers from a wider range. The multiple-choice format also necessitates a bulky scale that can
only be presented as a stand-alone instrument.
An alternative scale, the Oxford Happiness Questionnaire (OHQ) has been devised which con-
sists of single items that can be answered on a six-point Likert scale (Appendix). These items may
easily be incorporated into larger questionnaires in random order, and the opportunity has also
been taken of reversing about half of the items. These changes should reduce the probability of
contextual and compliant answering (Hills & Argyle, 1998a). The purpose of this paper is to
describe the improved scale and its psychometric properties and, by placing it in the public
domain, to allow its wider use and further examination by others.
One hundred and seventy-two undergraduate students of Oxford Brookes University and their
friends and relations (66 men, 99 women, seven unspecified) took part in the study. Ages ranged
from 13 to 68 (M=30.9, SD=12.9) years.
1It could be argued that respondents might be averse to endorsing a multiple choice item with an apparent score of
zero. In versions of the OHI used at, and distributed from, Oxford Brookes University since 1998, the items in the OHI
have been identified as a, b, c, and d, and scored on a 1–4 scale.
1074 P. Hills, M. Argyle/Personality and Individual Differences 33 (2002) 1073–1082
Respondents were invited to complete and return a self-report questionnaire constructed from
the OHI, the OHQ, and a number of published scales that are known to correlate with well-being.
These were the Extraversion, Neuroticism and Psychoticism sub-scales of the short form Eysenck
Personality Questionnaire (Eysenck, Eysenck, & Barrett, 1985), Rosenberg’s Self-esteem Scale
(Rosenberg, 1989), the Life Orientation Test (Scheier & Carver, 1985)—a measure of disposi-
tional optimism, the Life Regard Index (Battista & Almond, 1973)—a measure of both purpose
in life as represented by the existence of a set of life goals and the extent to which an individual
feels that he has fulfilled them, and the Depression–Happiness Scale (Joseph & Lewis, 1998). To
ensure uniformity of presentation, items in the original scales were reworded where necessary as
single statements to which participants could respond on a uniform six-point Likert scale ranging
from ‘‘strongly agree’’ to ‘‘strongly disagree’’. Before administration, the individual items of all
the scales with the exception of the OHI, were combined and rearranged in random order. It was
not, therefore, considered necessary to retain any filler items included in the original scales.
Questionnaires were administered in alternative versions; in one, the OHI was presented first and
in the other, last.
2. Results and discussion
2.1. Scale reliabilities
Both the OHI and the OHQ demonstrated high scale reliabilities with values a(167)=0.92 and
a(168)=0.91 respectively. The inter-item correlations for the OHI ranged from?0.03 to 0.58,
mean 0.28, and the corresponding values for the OHQ were ?0.04 to 0.65, mean 0.28. These
virtually identical results show that the multiple-choice items of the OHI can be replaced with the
more compact single choice items of the OHQ without detriment. The observation that the
maximum inter-item correlations within the two scales, r=0.65 and r=0.58, suggests that no two
items are so alike that they are measuring the same facet of happiness; in other words, no items
are semantically redundant. The questionnaires used in this study were administered in two ver-
sions in which the OHI was completed either first or last. A comparison of means (independent
samples t-tests) showed that the order of presentation resulted in no significant differences
between versions for either of the scales. The OHI and OHQ scores aggregated over all items were
strongly and significantly related, r(163)=0.80, P<0.001, which shows that both scales provide
very similar results.
2.2. Internal consistencies
The collected data were split into high and low groups above and below the mean aggregated
values for the OHI and the OHQ, respectively. The difference between the means of individual
item scores were then compared (independent samples t-tests) with respect to the two groups.
There were significant differences between the high and low groups for every item of the OHI and
the OHQ. Most were highly significant, P<0.001, and all differences were in the same direction as
P. Hills, M. Argyle/Personality and Individual Differences 33 (2002) 1073–10821075
the partitioned total scores. This indicates that all the items of both the OHI and the OHQ are
making a valid contribution to the measurement of overall happiness.
2.3. Between scales consistencies
Table 1 presents the correlations between corresponding items of the OHI and OHQ of which
all were significant at the P<0.001 level. The mean value was r=0.50, SD=0.11 with individual
values ranging from 0.69 down to 0.26. The table also identifies the 14 items that were adminis-
tered in reverse form in the OHQ. Since these items are more or less evenly distributed when the
correlations are arranged in descending order of magnitude, it would appear that item reversal
has not significantly affected the nature of the measure. This conclusion was supported by the
Correlations of corresponding OHI and OHQ items
28. look attractive (?)
12. wake up rested (?)
10. make decisions easily
16. feel healthy
15. mentally alert
21. can organise time
06. pleased with self (?)
04. in control
23. cheerful effect on others
05. life is rewarding (?)
13. feel energetic (?)
09. interested in others (?)
25. committed and involved
18. happy memories (?)
08. life is good
29. find things amusing
27. laugh a lot (?)
24. life has meaning and purpose (?)
20. done things wanted
26. world is good (?)
14. find beauty in things
02. optimistic (?)
19. joy and elation (?)
17. warmth for others
01. feel happy
03. satisfied with life
22. have fun with others (?)
07. good influence (?)
11. can do most things
OHI, Oxford Happiness Inventory; OHQ, Oxford Happiness Questionnaire.
a(?) items reversed in scoring.
bAll correlations significant at the P<0.001 level.
1076P. Hills, M. Argyle/Personality and Individual Differences 33 (2002) 1073–1082
observation that the sums of the positive and negative OHQ items both had high and virtually
equal correlations with the whole OHQ scale, r(168)=0.92, P<0.001 and r(168)=0.94, P<0.001
respectively, and that the positive and negative items scores were also highly correlated, r=0.73,
P<0.001. However, while the significance and magnitude of the associations between corre-
sponding items are satisfactory for the large majority of items, the correlation coefficients vary
over a considerable range. This implies that participants’ endorsements of some individual items
differ between the OHI and the OHQ and this is particularly marked for, say, the six items with
inter-correlations of <0.40.
2.4. Construct validities
Past research has established relationships between the OHI and a variety of trait and cognitive
variables that are associated with psychological well-being. Argyle and Lu (1990) found a strong
positive association with extraversion, which was confirmed by Furnham and Brewin (1990), who
also identified a strong negative asssociation with neuroticism. Substantial positive associations
have also been reported between the OHI and self-esteem, the life regard Index and the life
orientation test (Hills & Argyle, 2001a), and satisfaction with life (Hills & Argyle, 2001b). Joseph
and Lewis (1998) found a strong positive association between the OHI and the depression–hap-
Table 2 reports the correlations between the earlier variables and the OHI and the OHQ. Psy-
choticism excepted, all correlations are substantial and equally and highly significant. There is no
difference in the strength of associations with extraversion, but in all other instances, the per-
sonality variables correlate more strongly with the OHQ than with the OHI. It has previously
been reported (Argyle & Hills, 2000) that the association between psychoticism and the OHI does
not achieve significance. The present results confirm this observation, but there is a weak negative
relationship between psychoticism and happiness as measured by the OHQ. The overall rela-
tionships with the personality variables suggest that the construct validity exhibited by the OHI
Correlations between trait and cognitive variables for theOHI and OHQ
Satisfaction with life
Life orientation test
Life regard index
Depression–happiness (DH) Scale
DH positive items
DH negative items
OHI, Oxford Happiness Inventory; OHQ, Oxford Happiness Questionnaire.
P. Hills, M. Argyle/Personality and Individual Differences 33 (2002) 1073–10821077