Journal of Occupattonal PsyMogy (1990), 63, 193-210 Prtnted in Great Britain 193

© 1990 The British Psychological Society

The measurement of well-being and

other aspects of mental health

Peter Warr
Social and Applied Psychology Unit,
Umversity of Shefjudd, Shefjield Sl() 2TN. UK

New instruments are described for the measurement of both job-related and non-job
menta! health. These cover two axes of affective well-being, based upon dimensions of
pleasure and arousal, and also reponed competente, aspiration and negad ve job
carry-ovcr. Baseline data are presented from a sample of 1686 job-holders, and earlier
uses oí the well-being scalcs are summarized. The instruments appear to be psychomctri-
cally acceptable, and are associated with demographic and occupational features in
rxpected ways For example, older employees report greatcr job-related well-being;
occupational leve! is positively correlated with job depression -enthusiasm bur negatively
associated with job anxicty-contentment; depression-enchusiasm is more predictable
from iow-ro-medium opporcuniry for skiil use and task variety, whercas anxíety-
i onccntment is more a fuñerion oí workload or uncerrainry.

Many studies have examined the impact of work and careers on job-related and non-job
mental health. Nevertheless, there is st¿11 a shortage of instruments whose psychometric
properties have been determined through data from large samples of employees of both
genders and severa! occupational levels. In particular, there is a need for measures which
can provide infbrmation about affective well-being, subjecrive competence and aspir­
ation, through scores which can be com pared with known means and standard deviations
from appropríate demographic groups.
This paper aims to address those deficiencies, by describíng new instruments and
summarizing valúes obtained from a large sample of British job-holders. The approach to
affective well-being is through two principal axes (‘anxiety-contentment* and
‘depression-enchusiasm') which have emerged as important m non-occupacionaJ research,
and which will be examined through parallel measures in both job-related and non-job
settings. No other instruments are currentíy available which cover the full range of those
axes in both types of setting.
The approach taken is one which emphasizes practicality as well as psychometric
acceptability. Many occupational researchers are deterred by the length and cumbersome
language of sornc previous instruments, and are tempted to introduce the ir own
modifreatíons or creare new seales for one-off applicarion. Such developments prevent the
accumulation of' comparad ve data and cncourage an over-extensive tange of instruments
which all purport to rap the saíne construct. It is hoped that che straightforward na tu re of
the scales introduced in this paper will be of valué in many occupational settings.
194 Peter Warr
Affective well-being
A large number oí measures of job-related affective well-being has already been developed.
These cover specific facets of satisfaction, alienation from work, job attachment, job
tensión, depression, burnouc, involvement and job morale (e.g. Cook, Hepworth, Wall &
Warr, 1981). Context-free measures are available to tap life satisfaction, happiness,
pos i ti ve affect, negative affect, anxiety, depression, general dysphoria, self-esteern and
other types of féeling (e.g. Diener, 1984; Goldberg, 1972).
Restare h into these aspeets of well-being has been valuable and productive, but there is
merit also in drawing upon findings from investigarions into the strncture of emotions and
moods. Research has demonstrated the existente of two orthogonal dimensions, which
accounr fór the majority oí observed variante (e.g. Russell, 1979» ¡980; Watson &
Tellegen, 1985; Watson, Clark Tellegen, 1988; Zevon Tellegen, 1982). These are
summarized in Fig. 1, using the labels of ‘pleasure’ and arousal’, wirh illustrative
affective states ordered around the perimerer.
 The measurement of well-being 195
We may describe any form of affective well-being in terms of its location in relation to
those separare dimensions and its distance from the mid-point of the figure. A particular
level of pleasure may be accompanied by high or low levels of arousal, and a particular level
of arousal may be either pleasurable or unpleasurable. In devising primary measures of
well-being, decisions have thus to be taken about the location of key vectors in that
two-dimensionaí space.
Warr (1987) has suggested that three main axes should be considered. Two of these take
accounr of arousal as well as pleasure, by running diagonally between opposite quadrants
through the mid-point of Fig. 1. In addition, in view of the central importance of low or
high pleasure, ít is heípful to take measures aíong that horizontal dimensión alone,
wichout regard to variations in arousal.
That possibility is illustrated in Fig. 2. Pleasure and arousal are retained as the
horizontal and vertical dimensions, with two diagonal axes running between opposite
quadrants. The latter, and that representing pleasure alone, are labelled as the three key
indicators of affective well-being: (a) displeased-pleased, (2) anxious—contented, and
(3) dep ressed-enthusiastic. Principal types of affect may be located anywhere along
those axes. The arousal dimensión on its own is not considered to reflect well-being,
and its poles are therefore left un labelled.

The diagram is presented as an elongated (rather than circular) shapc to indícate that
pleasure is empirically accorded greater weight than arousal. Experienced pleasure may
differ substanriaíly across situacions, and these differences are more likely to be reflected ¿n
well-being than variations in arousal. Stores on axes two and three are thus likely to be
posicively intercorrelated in practice, rather than being independent, as suggested by the
ideal conceptual ization of Fig. I.
Within this framework, the precise location of the diagonal axes may be varied
according to research needs. For example, studies of cognitive performance in complex
tasks may be partí cularly concern ed with possi ble changes in arousal, to examine the
degrec to which job conditions lead to feelings of lethargy and fatigue. The third axis may
in rhose cases be defined towards the vertical dimensión. Measurements then would be
196 Peter Warr
more concerned with tircdness-vigour than with depression-enthusiasm. In some cases,
boch forms oí the third axis may be studied; this possibility ís considerad larer.
As outlined in the research literatura cited above, it is not possi ble to reduce all
cmotional experiences to these principal dimensions. The structure shown in Fig. 1 is
viewed as complementary to more di flfercntiated accounts, particular types oí which
mighr be more appropriate in certain settings. For example, detallad exanimación of
aspee es of well-being associated with specific pharmaceutical agenrs may benefit from
measures which cover a wider set of feelings. However, assessment of these three forms of
well-being can provide baste información, permití i ng addition of other measures when
that is des i red.
The pJeasure axis, shown horizontally in the figures, has often been measured through
scales of reported job or life sarisfaction. Many instruments are avaílable for job-specific or
context-free application (e.g. Cook el al.. 1981), and the present paper will focus on
well-being axes rwo and three. These deserve particular attenrion in occupational
research, to examine feelings of arousal as well as pleasurc.
Relevant 10-ítem scales have been devisad by Watson, Clark & Tellegen (1988), with
respondents ¿ndicacing thedegree to which rhcy are distressed, hostile, jittery, attentive,
inreresced, alerr, etc. Iinpressivc daca abone reliability and concurrent validity are
avaílable, and links with the pleasurc/arousal perspectiva llave been emphasized.
However, the scales are limitad in that all ítems fall into the top halfof Figs I and 2,
rather than covcring che íull length oí the axes. Furthermore, their focus is ¿nccntionally
vury broad, also asking, for example, whether respondents felt ashamed, guilry, proud
and strong. These context-free instruments do not lcnd rhumselves easily to occupational
research.
Axes rwo and three llave also been rapped through the context-free checklist described
by Mackay, Cox, Burrows & Lazzerini (1978); (see also Cox Mackay, 1985; and
Cruickshank, 1984). This contains 34 adjectives, 19 oí which fall on axis rwo as shown in
Fíg. 2. The others range from tiredness to vigour, covering a inore vertical form of axis
three than the one shown in the figure. A 20-ítem derivativa of this context-free measuru
has ticen examinad by Fischer Donacelli (1987), Fischer, Hansen & Zemorc ( 1988),
Kiiig, Burrows & Stanley (1985)» and King, Stanley Burrows (1987), in sruclics which
provide encouraging evidente about internal reliability and psychometric adcquacy.
'Hicsc and other authors have dernonstrared the robustness of che diagonal axes shown in
Fíg. 2. However, the emphasis has been on context-free mental health, and many ítems
are unsuitable for occupational application. The present study examines the two diagonal
axes through scales which use familiar adjectives, and cover equally the íour quadranes of
Fig. 2; axis three is representad by depression-enthusiasm rather than tiredness-vigour,
and the focus is upon job-relatad as well as non-job well-being.

Other aspeets of mental health

ln addition to affective well-being, high or low mental health is also exhibited through
behaviour in transactions with the environment. Such behaviours are conceptually quite
dístinct from the feelings involved in well-being, although associations bctwcen behav­
iour and well-being are likely to be observed in practica. Two inajor behavioural
com ponen ts are competente and aspiración (Warr, 1987).
 The measurement of well-being 197
Competence (e.g. Smírh, 1968) has been widely discussed in the psychological
literature. For example, Jahoda (1958) wrote in terms oí environmental mastery’,
Bradburn (1969) examined people’s ability to cope with and transcend their ‘difficulties in
living’, and Bandura (e.g. 1977) investigated beliefs about ’self-efficacy’ or expectations
of personal mastery’. A competent person is one who has adequate psychological resources
to deal with experienced difficulties. As with other aspects of mental health, a distinction
should be drawn between contexr-free competence and domain-specific competence, and
sepárate measurement of the two forms can be made. In both cases, it is necessary to
distinguish between subjective and independent assessments, with subjective competence
being measured through self-reports, and independent assessments being made by a
reíevant observer. The presenr study examines two scales of subjective competence,
covering separately job and non-job experiences.
A similar approach is taken to the measurement of aspiration. A mentally healthy
person is often viewed as having an ¿nterest in, and engaging with, the environment. He
or she establishes goals and makes active efforts to attain them, through motivated
behaviour, alertness to new opportunities, and efforts to meet challenges that are
personally significant. Conversely, low levels of aspiration are exhibited in reduced
involvement and activity, and in an acceptance of present conditions even when they are
unsatisfactory. This aspcct of mental health has been emphasized in occupational research
by Herzberg (1966), who examined the impact of job features on workers' ‘psychological
growth’. Maslow (e.g. 197>) developed a similar perspective in terms of ‘self-
actualization’. and parallel themes have been explored by Csikszentmihalyi (1975) and
Kornhauser (1965). Two measures of reported aspiration, job and non-job, will be
investigated herc.
There appear to be no instruments available to tap reported job-related competence and
aspiration in a manner appropriate for both shop-floor and managerial employees. Baseline
data about non-job as well as job-related components of those kinds would be useful for
comparison with subsequent invcstigations.
There has also been interest in the carry-over of job experiences into other activiries and
fcelings. For example, Evans & Bartolomé (1980), Piotrkowski (1978) and others have
documented negativo influentes on famíly and leisure life. Reports of such carry-over have
been examined in the presenr study, for comparison with other findings, and in rclation to
scores on the measures summarized abo ve.

Aims of this paper

This paper will describe new measures of these aspects of mental health, and present
baseline information from a largesample of male and female workers. Twoaxes ofaffective
well-being, and reported competence and aspiration, will be studied through parallel sets
of questionnaire ítems to assess both job-related and non-job mental health. The aims are
to characterize these instruments, and to make available comparative data from men and
women of different ages and occupational levels.
Scale validity will be examined in terms of demographic features, other aspects of
mental health, and through correlations with reported job characteristics. For example, it
is expected from research with other measures that older people will exhibit greater
198 Peter Warr
job-related well-being than younger people, and that employees in high-level jobs will
report more job enchusiasm and aspiration than those at Iower levels. Positivo correlations
with intrinsic job characteris ti es are expected, but the measures described here permit us
to go beyond most earlier research ¿n ¿nvestigating a differential pattern oí associations,
with varied relationships with job features expected according to which aspect oí mental
health is under investigation; this point will be developed later.
Evidente about axes two and three of job-related well-being will also be provided
through results from two sepárate samples, shop-floor workers in a manufacturing
company and professional staff attending a clinic for assistance in the management of
personal job stress.

Mcthod

Responderos and data-galhering

Data were obtained from 1686 employed men and women, all of whorn were in jobs for more than 50 hours a
week. A sample was drawn with approximacely equal numbers of men and women (839 and 847), in which
gender was balanceó within occupacional leve! and age.
Three occupational levels were specificd, in rerms defined by che Registrar-General as grades A and B, C1
and €2, and D. Grades A and B inelude professional and managerial workers in relatively sénior posirions;
grades C1 and C2 cover Iower professional and supervisory jobs and skilled non-supervisory posirions; and
grade 1) jobs are those manual posirions requinng littlc or no skill. Within diese three levels, che numbers of
malc* and female respondones respeccively were: AB, 269 and 24?; C1C2, 507 and 542; D, 265 and 258.
Three age levels were specificd in drawing up the sample: 18 ro 54 years, 55 to 49, and 50 years or above.
Within vach agegroup, the numbers oí men and women were: 18 to 54, 279 and 299; 55 ro49, 287 and 289;
50 and above. 27 5 and 259.
Respondones were incerviewed ar homo by survey research company staff in 75 locations throughout the
United Kingdom. Inicial and final groups of questions were asked oraily, covering a range of job and
personal i ty is.sues, and respondones rhemsclves completed rwo quescionnaires during the middle part of the
meering. The íirst oí diese was headed ‘Questions about your job', and instructions asked fot responsos in
cernís of how things have bcen going in the pasr tew weeks’. The second questionnaire was labe 11 ed ‘Questions
about your lile outsidc your job’, and asked respondones ro turn your artcntion away from your job and think
about other pares of your lile, your family and spare-rime activiries’. As before, the focus was on the pasr few
weeks’.
Both questionnaires contained Ítems in the sequence oí well-being axes two and three íbllowed by reponed
< ompetence and aspiración. Within the job-related questionnaire, subsequent Ítems, not reporccd in derail
here. toncerncd perccptions of job features such as worldoad and opporrunity for personal control.

The measures
Axes two and three of affective well-being were exaniined in the terms described earlier, through six-itern
scales derived from pre-tests carried out in a range of settings. The job-related ítems were preceded by the
quesdon, ’Thinking of the pasr few weeks, how much of the time has your job made you feel each of the
following?’ Responses were: never, occasíonally, some of che time, much of the time, most of the time, all of
che time; and answers were scored from 1 to 6 respeccively. Items covering the two axes were interminglcd in
the questionnaire.
Axis two, job-related anxicty-concentmenc, was assessed through the adjcctives tense, uneasy, worried,
calm. contcnted and relaxed. Responses to the firsr three ítems were re verse-scored, so that high stores
índicaced positive well-being, and coefíicient alpha was found ro be .76.
Axis three, job-related depression-enchusiasm, was capped by depressed, gloomy, miserable, cheeríui,
enthusiasric, optimistic. The lirst three ítems were again reverse-scored, and coefficienr alpha was .80.
 The measurement of well-being 199
For non-}ob well-being these two axes were measured in the same way, but with the question, 'In the past
fcw weeks, how much of the time in your life oucside your job have you felt each of the following? Alpha
coefficients were .81 for each of che axes.
Reponed job competente and job aspiration were each measured through responses scored I to 5 aJong the
continuum strongly disagree, disagree, netther disagree ñor agree, agree, strongly agree. Six statements were
used in each case, interspersed in che hrst questionnaire, and coefficients alpha were .68 and .62 rcspectively.
These scalcs, and those for non-job competente and non-job aspiration (inrermixed in the second ques­
tionnaire), are presented in Appendix I. Coefficients alpha for the non-job mensures were .71 and .64
respectively. Boch sets of mensures were based upon concepts and inore wide-ranging questionnaires
developed by Pearlin, Menaghao, Lieberrnan Mullan (1981). Rotter (1966) and Wagner Morsc (1975).
Note that they covcr reported compcrence or aspiration which is relativo to a person's own circuinstances. For
example, the job competente ítems cap ability to cope with a current job rather than absoluto levéis of
occupat tonal ability
Within ihe first questionnaire were four ítems to cover negative job carry-ovcr. Responses were again in
terrns oí the five-point agree--disagree seale, and coefficient alpha was found to be .78. These ítems are also in
Appendix 1.

Results

Affect i ve well-being: Principal componente analyses

The 12 adjectives to tap well-being were chosen on the basís of the earlier studies described
above, which have pointed to the existente of two bipolar axes runing diagonaliy through
the midpoint of Figs 1 and 2. Can these axes be ¿dentified empirically, through principal
components analyses of the present data? Previous analyses in other settings (for example,
of the Múltiple Affect Adjective Check List and the Profile of Mood States) have pointed to
the need to control for generalized response set. It is typically found that, without such
control, the two major components in this kind of data are sepárate groups of positive and
negative items, to the right and left of the vertical axis in Figs I and 2.
Such a separación appears likely to reflect the operar ¿on of acquiescent response set.
Gotlib & Meyer (1986) and ochers have pointed out that this feature may be of differential
impact, for example, between respondents, or with variations in acquiescence greater for
negative than for positive adjectives. These effeets would reduce correlations between
negative and positive seale Ítems and increase correlations within negative and positive
groups, giving rise to factors made up entirely of either negative or positive items.
Several invesrigators have therefore controlled for acquiescent response set in examining
the factor structure of adjectives ni context-free self-report scales. For example, Lorr, Shi &
Youniss (1989) found ¿n a 60-item list the two bipolar components which are expeettd
here. Response set was estimated as the total store across ail items, without any
reverse-scoring, and this was controlled in partial correlations between each ítem; factor
analysis was carried out on the matnx of these partial correlations. Applying this
procedure to the 12 adjectives in the present well-being mensure revealed the structure
shown in Table 1. For both job-related and non-job affect two factors were present
(cigenvalues greater than 1.00), and it can be seen that these correspond to axes two and
three of well-being as defined above.
There is another way to examine the structure of the well-being items without possible
interference from differential acquiescence response set fot positive and negative ítems.
This involves sepárate analyses for the six positive and the six negative items making up
200 Peter Warr
Table I . Varimax-rotated factor loadings greater than .40, from parcial correlations
between affective well-being ítems, controiling for response set (N — 1686)

Job-related Non-job
¡iffecr affecr

Tense .70 .71

Uneasy .43 .52 .58
Worried .66 .66
Calm .78 .76
Conrented .41 .62 .65
Relaxed .79 .81

Dep ressed .68 .47 .67

Gloomy .75 .44 .61
Miserable .71 .47 .63
Cheerfui .56 .47
Enthusiastic .79 .72
Optimiscic .65 • 79

Note. Scoring of the six negative ítems has been re verseó.

che two scales. Four principal component analyses of those kinds have been carried out
separately for positive and negative ítems in the job-related and the non-job scales. Each
varimax-rotated two-factor sol ut ion contained distinct three-item components repre-
senting quadrants above and below the horizontal axis of Figs 1 and 2, supporting the
structure assumed in the pleasure/arousal model.

Lnterwrrelations hetween meas uves

The correlations observed between these measures of affective well-being and other aspeets
of mental health are shown in Table 2. Recall that higher scores indícate greater
contentment or enthusiasm (and thus lower anxiety or depression), as well as greater
competente, aspiration and negative carry-over.
Consistent with the location of axes in Fig. 2 (at less than 90 degrees to each other), the
two well-being scores in each domain (jc>b and non-job) are strongly intercorrelated (.66
and .73)- Correlations of job-related depression-enthusiasm are sigmficantly greater with
job aspiration than with job competence (.46 and . 26, p < .001). This appears to reflect
the impact of the arousal dimensión, with reports of enthusiasm and aspiration both
deriving from high arousal as well as high pleasure (the top right-hand quadrant in the
well-being figures).
Conversely, negative carry-over from the job is significantly more strongly correlated
with job anxiety-contentment than with depression~enthusiasm ( — .54 and —.30), and
inverse carry-over correlations are significan tly greater with reported competence than
reported aspiration (both job and non-job). Anxiety-contentment is also sigmficantly
more predictable from reported competence than from aspiración. (The/? < .00 1 criterion
has been used in all cases.)
 The measurement of well-being 201
Table 2. Intercorrelations between aspects of mental health (N — 1686)

2 3 4 5 6 7 8 9

Affective well-being
1. Job anxiety-contentment .66 .34 .48 .36 .24 -.54 .28 • 13
2. Job depress ion-enthusiasm .27 .58 .26 .46 -.30 .27 .21

3. Non-job anxiety-contentment ■ 73 .30 .17 -.46 .48 .25

4. Non-job dcpression-enthusiasm .28 .30 -.32 .49 .40

Other aspects of mental health

5. Reported job comperence .36 -.37 .44 .21
6. Reported job aspiration -.10 .29 .38

7. Negative carry-over '.30 - -.13

8. Reponed non-job comperence -49

9- Reponed non-job aspiration

Note. As described ín the cext, high stores on variables I and 3 rcflect concent me nt, and high stores on
variables 2 and 4 reflecr enthusiasm. Valúes greater than .07 are significanc ac che p < .001 level.

Mean stores and demographic vaviations

Mean stores are set out ín 'Iabie 3, with the overall pattern surnmarized in the first
column. Because of the large number of cases, these valúes may be used in comparison
with subsequently gathered data, but it should be noted that the sample was not intended
to be entirely represe ntat i ve of the population. In order to permit reliable analyses of
sub-group data, the design sought disproportíonately more people in occupational levels
either higher or lower than the average. In the present sample, 31, 38 and 3 1 per cent of
respondents are in occupational levels AB, C1C2 and D respectively; comparable figures
for the employed population as a whole are around 17, 58 and 25 per cent.
Comparisons between male and female responses (in columns two and three) reflect job
and othcr environmcntal variatíons as well as possi ble gender effects on their own. In the
present case, women exhibir significantly greater job enthusiasm than men, and report
significantly greater difficulty in coping with their paid work (in terms of the job
comperence scale).
Occupational-level stores present a coherent pattern ofdifferences, with the direction of
trends varying between specific aspects of mental health. As wouid be expected, people in
high-level jobs report significantly more job-related enthusiasm; but their scores are al so
signíhcantly lower on job anxiety-contentment (that is, they report more anxiety).
Higher-level employment is thus associated with greater arousal levels of both kinds
idenrified by axes two and thtee.
Reported job aspiration and negative carry-over are also greater at higher job levels, but
the reverse pattern is found for reported job comperence. This latter finding reflects the
fact that the subjective job comperence scale taps reported ability to cope with one’s own
 202
Table 3. Mean mental health stores (standard deviations in parentheses)

Gender Occupationai levcl Age

Full
sample Men Wornen AB C1C2 D 18 to 54 55 to 49 50 and above

Affatwe weli-being
1. Job anxiety-contentment 4.17(.81) 4.12 (.82) 4.22 (.80) 3.96*(.77) 4.20*(.82) 4.55*(.8O) 4.06*(.80) 4.12*(.82) 4.34*( 82)
2. Job depression-enchusiasm 4.55C.79) 4.48*(81) 4.62*(.76) 4.61 *(.70) 4.58*(.82) 4.46*1.83) 4.48*(.78) 4.51*(.8l) 4.68*(.76)

3. Non-job anxiety-cüntentment 4.4U.76) 4.43 (-75) 4.39 i.78) 4.31*C74) 4.43*(.77) -i.48*í.77) 4. 56*(.75) 4.36*(.79) 4.52*(.73)
4.62C.72) 4.56 (.75)

Peter Warr
4. Non-job depression-enthusiasm •1.63 (-70) 4.62 (.73) 4.66 (.69) 4.65 (.70) í 60 (.67) 4.57 (.76) 4.71 (.71)

Other aspects oj mental health

5. Reponed jcb competente 3.88C54) 3.94*04) 3.82*653) 3.78*1.56) 3.93*(.53) 3-93*(.5O) 3-77*(.53) 5.91*(.53) 3.98»(.54)
6. Reported job aspiration i.O6(.50) 4.03 (.53) 4.09 (.48) 4.22*(.42) 4.()7*(.50) 3.9O*(.53) 4.04 (.50) 4.07 (.52) 4.09 (.49)

7, Negarive carry-over 2.73(.9D 2.68 (.91) 2.77 (.92) 3.O3*(.95) 2.65*(.9O) 2.53*(.82) 2-75 (.88) 2.77 (.91) 2.65 (.95)

8. Reported non-job competen ce 3.76C55) 3 ’9 (.50) 3.74 (.56) 3.78 (.54) 5.79 (.53) 5.72 (.53) 3-75 (.52) 3.76 (.55) 3.79 (.53)
9. Reported non-job aspiration 3 92(.48) 3.90 (.48) 3.95 (.50) 3.99V.47) 3.93*(.48) 3.84*(.50) 3-97 (.46) 5.91 (.51) 3.88 (.49)

Number oí cases 1686 839 847 516 649 521 578 576 532

Signiíicant differences between means in each group have Leen confirmed in more control led analyses through múltiple regrcssions of each mental healch variable on gender,
occupationai ievel and age togerher.
* p < .001, in compansons berween groups.
 The measurement of well-being 203
job; it is not an absolute índex of competence of the kind that would be expected to
correlate positively with job level.
Significant age dífferences are found in respect of job-related anxiety-contentment,
depression—enthusiasm and reported job competence, with older people exhibitíng higher
scores. This pattern ¿s wideíy found (e.g, Rhodes, 1983), and is líkely in part to reflect
variations in job content. Observed greater non-job contentment ín older respondents is
also likely to derive from environmental dífferences associated with increasing age within
chis range; note, however, that reported non-job competence and aspiración remaín
constant across the age groups.
Two-way analyses of variance were carried out on all combinations of gender,
occupational level and age; no interactíons were statistically significant. Additional
examination was made in order to identify the pattern of associations with the number of
weekly hours worked (mean = 44.67; SD — 10.83). Correlations with all the mental
health variables were insignificant (p > .001), except for negative carry-over (r — . 19).
Controííing for gender, occupatfonal level, age, marital status and educational quaíifi-
cations left that correlation unchanged (partía! r — . 19).

Asfocta/tons with job content

It is of ínterest to examine the associations of mental health scores with variations in
perceived job characteristics. Table 4 summarizes correlations with reports on three-item

Table 4. Correlations betwecn perceived job characteristics and mne indices of mental
health (N = 1686)

Low-co-medium levels

Personal
Skill use control Work-load

Affective well-being
1. Job anxiety-contentment .01 .25 -.10
2. Job depression- cnrhusiasm .26 .37 . 17

3. Non-job anxiety-contentment .04 . 17 -.01

4. Non-job depression-enthusiasm .14 .22 .11

Other aspeas of mental health

5. Reported job competence .02 . 13 .00
6. Reported job aspiración .45 .40 .39

7. Negative carry-over . 10 -.14 .23

8. Reported non-job competence .09 .17 .13

9. Reported non-job aspiración .13 .20 . 18

Noce. Valúes grcater than .07 are significanr ar the p < .001 level.
204 Peter Warr
scales of perceived skill use, personal control and workload. These scales were designed to
cover low-to-medium valúes of each characteristic, with ítems worded consistently in
terms of very iittle skill use, control or workload (for further de tai Is, see Warr, 1990).
Low levels of ¿ntrinsic job characteristics, such as skill use and personal control, might
be expected to be primarily associated with reduced arousal, reflected in low depression-
enthusiasm and low job aspiration, rather than in terms of anxiety-contentment and
ability to cope as tapped by the job competente scale. That pattern is found in the table.
For skill use, correlations are significantly greater with job and non-job depression-
enthusiasm (.26 and . 14) than with anxiety-contentment (.01 and .04) (p < .001), and
differences between correlations with job aspiration and job competente mirror that
contrast (.45 and .02 respectively; p < .001). These differences are also significant for
personal control (p < .001 for the two axes of job well-being and for job competence and
aspiration). In general, variations in rhose two job characteristics are accompanied by
variations in low-arousal pleasure, in the terms of Fig. 2.
Conversely, perceived workload is significantly more negatively associated with
anxiety-contentment than with depression-enrhusiasm ( —. 10 vs. . 17, and — .0 1 vs. . 1 1
for job and non-job affcct respectively; p < .00!), as expected from the fací that raised
workload is Iikely to be associated with both greater arousal and negative feelings: the
top-left qu adran r of the figures. The scales for well-being dimensions two and three are in
these ways differentially associated with job characteristics in a manner which parailels the
common belief, m clinical psychology, that depressed feelings are more associated with
loss or deprivation whereas anxious feelings are more Iikely to reflect a responso to r brear or
danger.
The correlations in Table 4 remain almost unchanged after controls are introduced for
occupaúonal lovel, educar ional qualiheations, age, gender and marital status. However,
the exacr valúes are of less concern here than the finding that associations with job fea tu res
vary berween the sepárate instruments under examination.

Áddit tonal data

Earlier versions of the scales to cap axes two and three of job-related affective well-being
have been used in research with several other groups. Two examples will be cited here:
from 246 shop-floor assembly workers m a micro-electronics factory (both malo and
female), and 96 professional and managerial workers (male and female) who were taking
part in psychotherapy to reduce their high levels of job srraín. Data were gathered by
Wall, Clegg, Davies, Kemp & Mueller (19S7) and Shapiro, Barkham, Hardy & Morrison
(1990) respectively.
Well-being axes two and three were each examined through six adjectives. Seven of the
12 were the same as in the study described above; and anxious, frustrated, comfortable,
keen and lively were used in place of the later selección of worried, uneasy, contentcd,
cheerful and optirnistic respectively. Controlling for acquiescent responso set again
revealed the expected rwo-factor structure in both sets of data.
The two forms of job-related affective well-being were significantly intercorrelated (.54
and .58 in the two studies), but their validity is again supported through a difierentiated
pattern of associations with other variables. For example, Table 5 shows that job
depression-enthusiasm is significantly more el osel y related to i ntrinsic job satisfacción
 The measurement of well-being 205
Table 5. Affective well-being dimensions two and rhree among shop-floor workers in a
micro-electronics assembly dcpartment: Correlations with other variables (N = 248)

Job Job
anxiety- depression—
contentment cnthusiasm

Intrinsic job satísfactíon .21* .40*

Extrinsic job satísfactíon .31 .33
Reported job motivación .20* .40*

Reported skill use . 15* .46*

Reported job complexity . 18* .47*
Reported attentional control .12* .31*
Reported task repetition .05* -.22*
Reported work overload -.40* — .09*

Context-free di stress (GHQ) -.46 -.39

*/» < .001 between a pair of correlations.

than is anxiety—contentment (r — .40 and .21 respectively, measuring satísfactíon

through the sea le described by Warr, Cook & Wall, 1979). This significant difference ís
also present in relation to reported job motivación, measured through ítems similar to
those in the scale of job aspiration described above.
In terms of job characteristics, the pattern is similar to that in Table 4. Strong
associations are found between depression-enthusiasm and aspeets of intrinsic job
fcarures. reported skill use, job complexity, a tren ti onal control and the absence of
repetition; correlations of chese variables with the anxiety-contentment axis are signifi-
cantly lower. Conversely, work overload ¿s associated with greater job anxiety (r = — .40),
but the re is no such association with job enthusiasm—contentment (r = —.09).
Final! y, ir can be seen in Table 5 that both aspeets of job-related well-being are
significantly correlated with stores on the 12-¿tem General Health Questionnaire
(Goldberg, 1972). 'That questionnaire is an indicaror of context-free well-being along the
íeft-hand side of the horizontal axis of Fig. 2. The new scales described here provide
addirional data to those derived frorn the GHQ, in that they cover separare diagonal axes
and permit measurement of both job-related and non-job affect.
A chird set of daca is íHustratcd in Table 6, which presen ts intercorrelacions between
variables measured beforc the commencement of therapy. The job characteristic measures
were different from those described above, but the scale of job satísfactíon was the same as
in the manufacturing study, The findings provide similar supporr for rhe new well-being
scales.
As expected from rhe previous results, job depression-enthusiasm is significantly more
dosel y associated with intrinsic job satísfactíon and reported skill use than is anxiety-
contentment. On the other hand, anxiety-contentment is more predictable from job
clarity (about standards and fu ture prospeets) and (negatively) from reported overload than
is the other dimensión. Table 6 also reports associations with two context-free measures of
206 Peter Warr
lable 6. Affective well-being dimensions two and three among professional and
rnanagerial workers attending a job stress clinic: Correlations with other variables befare
commencement of therapy (N = 96)

Job Job
anxicty- depression-
contentment enchusiasm

Intrinsic job satisfaction .34* .52*

Extrinsic job satisfaction .40 .45
Negativo carry-over ”.46* -.30*

Reported skill use .30* .51*

Reponed control opportunity .29 .34
Reported task variety .02 . 19
Reported social support .31* .48*
Reported job clarity .32 . 19
Reported work overload -.39** -.10**

Con tex t-free depress ion -.09 -.24

Self-esteem . 33 .30

*p < .05; **/> = .01 berween a pair of correlations.

mental health, the Beck Depression Inventory (BDI) (Beck, Ward, Mendeison, Mock &
Erbaugh, 1961) and an eighr-item self-esteem measure based on that devised by
Rosenberg (1965). As would be predicted, job-related depression--enthiisiasm is more
related than anxiety-contentment ro the (context-free) BDI ( — .24 vs. —.09), buc no
differenccs are found for the more general self-esteem measure. However, far this size of
sample the difference between correlations with the BDI is not significant.
This study also permitted measurements of changes as a result of psychotherapy.
Significant ¿mprovements were record ed in severa! standard clínica! indicators, assessing
context-free mental health. Since the focus of the therapy was on job-related as well as
more general problems, scores on borh axes two and three of job-related well-being would
also be expected to improve significantly. This was in fact observed (/> < .001 in borh
cases).

Discussion
These questionnaire measures of affective well-being and other aspects of mental health
appear to be psychometrically acceptable, easy for job-holders at all levels to complete,
and associated with demographic and occupational features in predicted ways. The
approach to well-being is based upon an externally supported two-dimensional
framework, and, although the well-being instruments yield scores which are (as expected)
intercorrelated, they exhibit different associations with other factors which provide
evidence of their validity.
For example, higher occupational levei was found to be positively associated with job
depression-enthusiasm but negatively associated with anxiety—contentment, reflecting
 The measurement oj well-being 207
raised arousal of both kinds in higher-level jobs. Age differences paralleled those found in
earlier research, but provided additional información through the comparison of job-
rclated competente and aspiración; the former was significantly positively correlated with
age, but the latter was not. The length of a person’s working week was unrelated to all
aspects of mental health examined here, except fot negative carry-over (r = . 19).
Intrinsic job satisfacción tended to covary more with depression-enthusiasm than with
anxiety-contentment, but no difference was observed for extrinsic satisfacción; this
partero is consisrent with the meanings of the constructs. So is the observed overlap
between negative carry-over, job anxiety-contentment and reported competence. Across
the studies, with different measures of job f’eatures, depression-enthusiasm was more
predictable from variables such as low-to-medium opportunity for skill use and task
variery, whereas anxiety-contentment was more a function of high workload or
unccrtainty. The former characteristics can be viewed as illustraring aspects of depr i va­
cien, whereas the latter reflect possible threats; the partero is consistent with non-
occupational thinking about depression and anxiety.
Conrcxt-frec constructs of those kinds have been investigated in clinical and com-
rnunity research. Measures of anxiety and depression are consistently found to be highly
intercorrelated (Dobson, 1985, reports an average correlation of .61), but there is usuaily
agreed to be merit in retaining both conceprs when examining context-free mental health
and íll-health. By the same token, the present domain-specific measures of axes two and
three of affective well-being are empirically associated (between .54 and .73 with the
samples used above), but they are differentially correlated with other factors. It is
irnportant to consí der them both.
The third axis oí well-being was identified in Fig. 2 as running from depression to
enthusiasm, and Ítems to tap those aspects of the opposed quadrants of che figures have
been described here. However, it was suggested in the introducción to this paper that in
some invesrigations there might be particular interese in a more vertical axis within those
quadrants, closer to the arousal dimensión and ranging from tiredness to vigour. Such an
axis has also been examined in the present study, but for clarity of presentation full results
have not been ineluded in this paper. Summary details are provided in Appendix 2.
The new scales of reported competence and aspiration in job and non-job setrings were
found to yield coherent results along plausible lines. Significant associations with
occupational level, age, job characteristics and affective well-being have been described
above. However, those measures are so far primarily supported by other forms of
self-report data. This also applies to the instruments for measuring axes two and three of
affective well-being, and there is nowr a general need to test the vahdity and usefulness of
all these scales against bchavioural and other objectíve criteria.

Acknowledgement
I am very grateful to Jan Jackson, for extensive help with data-processing.

References
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84,
19 1215.
Bcck, A. T., Ward, C. H., Mendelson, M., Mock, J. & Erbaugh,J. (196t). An inventory for measuring
depression. Archives of General Psychiatry, 4, 561-571.
208 Peter Warr
Bradburn, N. M. (1969). The Structure of Psychological Well-being. Chicago: Aldine.
Cook,JL D., Hepworth, S. J.» Wall, T. D. & Warr, P. B. (1981). / he Experience of Work. London: Academic
Press.
C.ox, T. & Mackay, C. (1985). The measurement of scjf-reported stress and arousal. British Journal of
Psychology. 76, 183—186.
Cruickshank, P. J. (1984). A stress and arousal mood scale for low vocabulary subjects: A reworking of
Mackay el al. (1978). British Journa/ of Psychology. 75, 89~94.
Csikszentrnihalyi, M. (1975). Beyond Boredum and Anxieiy. San Francisco: Jossey Bass.
Diener, E. (1984). Subjccrive well-being. Psychological Bullelin. 95, 542-575.
Dobson, K. S. (1985). The rclationship betwcen anxicry and depression. Clinical Psychology Ret’iew, 5,
307-324.
Evans, P. & Bartolomé, F. (1980). Must Success Cosí So Much? London: Grant Mclntyre.
Fischcr, E). G. & Donatelli, M. J. (1987). A meas u re oí stress and arousal: Factor structure of the Stress
Adjective Checklist. Falucational and Psychological Measurement. 47, 425-435.
Fischcr, D. G., Hansen, R. J. & Zemore, R. W. (1988). Factor structure of the Stress Adjective Checklist:
Replicated. Educa!ional and Psychülogical Measurement. 48, 127-136.
Goldberg, D. P. (1972). The Detection tf Psychiatric lllness by Questionnaire. Oxford: Oxford University Press.
Gotlib, 1. H. &Meyer,J. P. (1986). Factor analysis of the Múltiple Affcct Adjective Check Lisc: A separación
ol positive and negativo affect. Journal of Personality and Social Psychology. 50, l lól-I 165.
Herzbcrg, F. (1966). Work and the Nalure of Man. Chicago; World Publishing Company.
Jahoda. M. (1958). Current C.oncepts of Positivo Mental Health. New York: Basíc Books.
Ring, M. C., Burrows, G. D. ¿icStanlcy, G. V. (1985). Measurement of stress and arousal: Validación of the
stress/aroysal adjective checklist. British Journal of Psychology, 74, 473-479.
King, M. C.» Sean ley, G. V. & Burrows, G. D. (1987). Stress: Theory and Practico. Sydney: Gruñe and
S trac con.
Kornhauser, A. W. (1965). Mental Health of the Industrial Worker. New York: Wiley.
Lorr, M.,Shi, A. Q. & Youniss, R. P. (1989). A bipolar muí tí factor concepción of mood scates. Personality and
Individual Difjerences, 10, 155—159.
Mackay, C., Cox, T., Burrows, G. & Lazzerini, T. (1978). An inventory for the measurement of scjf-reported
stress and arousal. British Journal of Social and Clinical Psychology. 17, 28 3 -284.
Maslow, A. H. (197 3). The Parlher Reaches of Human Nature. London: Penguin.
Pearlin, L. L, Menaghan, E. G., Licberman, M. A. &. Mullan, J. T. (1981). The stress pruccss. Journal of
Health and Social Behavior, 22, 337—356.
Piotrküwski, C. S. (1978). Work and the Pamily System. New York: Free Press.
Rhodes, S. R. (1983). Age-reiated difiere nces tn work actitudes and behavior. Psychological Bidletin, 93,
328-367.
Rosenberg, M. (1965). Society and the Adolescent Self-image. Princeton: Princeton University Press.
Rotter, J. B. (1966). Generalized expectancies for infernal versus external control of reinforccment.
Psychological Monographs. 80, 1-28.
Russell, J. A. (1979). Affective space is bipolar. Journal of Personality and Social Psychology. 37, 345-356.
Russell, J. A. (1980). A circumplex model of affcct. Journal of Personality and Social Psychology. 39,
1161-1178.
Shapiro, D. A., Barkharn, M. J., Hardy, G. E. & Morrison, L. A. (1990). The Second Sheffield
Psychotherapy Projcct: Raciónale, design and preltminary outeome data. British Journal of Medical
Psychology, 63, 97-108.
Smith, M. B. (1968). Compecence and 'mental health’. InS. B. Sells(Ed.), The Definí t ion and Measurement of
Mental Health. Washington: Department of Health, Educación and Welfare.
Wagner, F R. & Morsc, J. J. (1975). A measure of individual sense of competence. PsychologicalReports. 36,
451-459.
WalI.T. 1)., Clegg, C. W., Davies, R. T., Kemp, N. J. &Mueller, W. S. (1987) Advanced manufacturing
technology and work simplificación; An empirical study. Jwz/ra?/ of Occupational Behaviour. 8, 23.3—250.
Warr, P. B. (1987). Work. Dnemployment, and Mental Health. Oxford: Oxford University Press.
Warr, P. B. (1990). Decisión laticude, job demaods and employee well-being. In preparation.
Warr, P. B., Cook, J. & Wall, T. D. (1979). Scalesfor the measurement of some work attitudes «and aspcccs of
psychological well-being. Jíwrw/ of Occupational Psychology. 52, 129-148.
 The measurement of well-being 209
Watson. D. & Tellegen, A (1985). Toward a consensúa/ scrucrurc of mood. PsyMogical Bulletin. 98,
219-2.35.
Watson, D., Clark, L. A. & Tellegen, A. (1988). Development and validación of brief measures of positive
and negativo affecr: The PANAS scaies. Jtwrwa/ of Personality and Social Psychology. 54. 1063-1070.
Zevon, M. A. & Tellegen, A. (1982). The scructure of mood chango: An ídiographic/nomochetic analysis.
J ourna l of Personality and Social Psychology. 43, 111-122.

Received 10 Marcb 1989: revised versión received 8 January 1990

Appendix 1
The ítems used to measure reported competente, aspiración and negativo jobcarry-over aregtven below. Those
marked '(R)’ were reverse-scorcd.

Reported job competería

I can do my job wcll
I sometí mes think l am ñor very competen t at my job (R)
I can deal with just about any problem in my job
I hnd my job quite difficult (R)
I feel 1 am better than most people at tackling job difficukics
In my job I oír en ha ve temible coping (R)

Reported job aspiration

In my job I like to set myself challenging targets
I am not very interested in my job (R)
í enjoy domg new chings in my job
i pretor to avoid difficult activitics ¿n my job (R)
In my job. 1 make a special effort to keep tryíng when rhings seem difficult
1 am not very concerned how things turo out in my job (R)

Negative job carry-over

After 1 lea ve my work, I keep worrying about job probJcms
I find ir difficult to unwind at che end of a work-day
I feel used up at the end of rhe work-day
My job makes me feel quite exhausted by che end of a work-day

Reported non-job competería

I can deal with just about any problem in my non-job lifc
I sometimos rhink 1 am not very comperent in my non-job life (R)
Most rhings I do, I do well
I find my non-job lifc quite difficult (R)
I feel 1 am better than mosr people at tackling difficulties
I often have rroublc coping m my non-job life (R)

Reported non-job aspira ti un

i enjoy domg new chings in my non-job life
I am not very interested in che world a round me (R)
I likc ro set myself challenging targets in my non-job lifc
1 pretor to avmd difficult activitics in my non-job life (R)
I make a special effort to keep trymg when things seem difficult
I am not very concerned how things tura out in my non-job life (R)
210 Peter Warr
Appendix 2
As indicated in the Discussion sección, this study also obtained información about an alternativo indicaror oí
the chird axis of affective well-being, ranging ffom tiredness to vigour. The ítems used to measure chis set of
feelings in both the job and non-job questionnaircs were: fatigued, lifeless, tired, aJcrc, full ofenergy and
lively; and as with the other scales the first three were reverse-scored. These ítems were additionally presentad
with the 12 ítems oi the two prunary scales, described above.
Principal component analyses of these axis three Ítems with the previously described axis two ítems
revealed che sepárate factors specified by che rnodel. Using the parcial correlación procedurc described c-arlicr,
the expecred two-componenc stnicture was apparent in boch job and non-job rcsponses after varimax rotación.
in order to assist comparison with fu ture studics using che tiredness vigour scale, the lblk) wing mean
valúes and standard deviación* (in paren rh eses) are ciced. Job-relared rcsponses: full samplc, 4.20 (.76); men,
4.18(.75), women 4.22 (.77); levcl AB, 4.16(.69), levcl C1C2, 4.23 (-75), levcl D, 4.22 (.82);age 18to34,
4.14 (.75), age 1.5 to 49, 4.17 (.77), age 50 and above, 4.31 (-74) (/> < .001). For non-job rcsponses,
corrcsponding valúes were: 4.22 (.74), 4.22 (.72) and 4.22 (.75); 4.12 (.7 l), 4.24 (.72) and 4.29 (.77)(g <
.001); 4.24 (.72), 4. 16 (.75.) and 4.26 (.74). Samplcs sizes are shown in Tablc .3.
Copyright of Journal of Occupational Psychology is the property of British
Psychological Society and its contení may not be copied or emailed to múltiple sites
or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.