Downloaded from www.sjweh.

fi on April 20, 2019

Original article
Scand J Work Environ Health 2005;31(6):438-449

doi:10.5271/sjweh.948

The Copenhagen Psychosocial Questionnaire—a tool for the

assessment and improvement of the psychosocial work
environment
by Kristensen TS, Hannerz H, Høgh A, Borg V

Affiliation: National Institute of Occupational Health, Lerso Parkalle

105, DK-2100 Copenhagen O, Denmark. [email protected]

The following articles refer to this text: 2008;34(5):374-380;

2009;35(4):284-293; 2010;36(5):404-412; 2010;36(6):435-444;
2011;37(5):446-449; 2012;38(6):489-502; 2012;38(6):527-536;
2013;39(5):477-485; 2014;40(2):186-194; 2014;40(5):457-464;
2014;40(5):473-482; 2017;43(4):326-336; 2017;43(4):316-325;
2018;44(2):113-133; 2019;45(2):174-182; SJWEH Supplements
2008;(4):25-29; SJWEH Supplements 2008;(6):75-82; SJWEH
Supplements 2008;(6):83-90; SJWEH Supplements 2008;(6):150-162;
[online first; 28 December 2018]

Key terms: assessment; Copenhagen Psychosocial Questionnaire;

psychosocial factor; psychosocial work environment; stress; survey

This article in PubMed: www.ncbi.nlm.nih.gov/pubmed/16425585

This work is licensed under a Creative Commons Attribution 4.0 International License.

Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Copyright (c) Scandinavian Journal of Work, Environment & Health
 Original article
Scand J Work Environ Health 2005;31(6):438–449

The Copenhagen Psychosocial Questionnaire—a tool for the assessment

and improvement of the psychosocial work environment
by Tage S Kristensen, DrMedSc,1 Harald Hannerz, BSc,1 Annie Høgh, MSc,1 Vilhelm Borg, MSc 1

Kristensen TS, Hannerz H, Høgh A, Borg V. The Copenhagen Psychosocial Questionnaire—a tool for the
assessment and improvement of the psychosocial work environment. Scand J Work Environ Health
2005;31(6):438–449.

Objectives The aim of this article is to present the Copenhagen Psychosocial Questionnaire (COPSOQ), a
questionnaire developed in three different lengths for assessing psychosocial factors at work, stress, and the
well-being of employees and some personality factors. The purpose of the COPSOQ concept is to improve and
facilitate research, as well as practical interventions at workplaces.
Methods The development of the questionnaire was based on a survey of a representative sample of 1858
Danish employees aged 20–59 years. The response rate was 62%; 49% were women. Altogether 145 questions
from some international and Danish questionnaires and 20 self-developed questions were tested with factor
analyses, analyses of internal reliability, and analyses of response patterns.
Results The analyses resulted in a long research version of the questionnaire with 141 questions and 30
dimensions, a medium-length version for work environment professionals with 95 questions and 26 dimensions,
and a short version for workplaces with 44 questions and 8 dimensions. Most of the scales have good reliability,
and there seems to be very little overlap between the scales. A novel feature of the COPSOQ is the development
of five different scales on demands at work.
Conclusions The COPSOQ concept is a valid and reliable tool for workplace surveys, analytic research,
interventions, and international comparisons. The questionnaire seems to be comprehensive and to include most
of the relevant dimensions according to several important theories on psychosocial factors at work. The three
versions facilitate communication between researchers, work environment professionals, and workplaces.

Key terms psychosocial factors; stress; survey.

The psychosocial work environment is generally con- In research on psychosocial factors several differ-
sidered to be one of the most important work environ- ent methods can be used, such as standardized question-
ment issues in contemporary and future societies (1–5). naires, clinical examinations, observational methods, reg-
A large proportion of employees in the countries of the isters, secondary data, and qualitative interviews. Most re-
European Union (EU) reports being exposed to psycho- searchers agree that well-validated standardized question-
social stressors at work, and the consequences are be- naires are necessary tools in research, as well as in practi-
lieved to be very significant for workers, workplaces, cal prevention. In our article, such a tool is presented: the
and society. Among these consequences are muscu- Copenhagen Psychosocial Questionnaire (COPSOQ).
loskeletal disorders, cardiovascular diseases, mental dis-
orders, stress, burnout, reduced quality of life, sickness
absence, labor turnover, and decreased motivation and Background and aim
productivity. Recently, the EU member countries gave
psychosocial factors “top priority” among work envi- When the Psychosocial Department was established at
ronment factors (6), and the directors of the European the National Institute of Occupational Health in Den-
work environment institutes have estimated that psycho- mark in the middle of the 1990s, we were approached
social factors would be the most important research field by work environment professionals from occupational
in the future (7). health services, from the national Work Environment

1 The National Institute of Occupational Health, Copenhagen, Denmark.

Correspondence to: Dr TS Kristensen, National Institute of Occupational Health, Lerso Parkalle 105, DK-2100 Copenhagen
O, Denmark. [E-mail: [email protected]]

438 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 438 19.12.2005, 13:50

 Kristensen et al

Authority, and from the occupational health clinics with sample of 1858 working Danes between 20 and 60 years
queries about a suitable standardized and validated ques- of age (49% women, response rate 62%). Names and
tionnaire in the field of the psychosocial work environ- addresses were provided by the Central Person Regis-
ment. We investigated a large number of Danish and ter. Two-thirds of the sample were allocated to receive
international questionnaires but could not recommend a postal questionnaire, while we tried to reach the last
any of them for standard use in Denmark. After some third by telephone. (The differences between the an-
hesitation, we decided to develop our own questionnaire, swers of the two groups were, in most cases, minor, and
or to be more precise three versions of a questionnaire: there was no systematic information bias.) Two remind-
a long version for research use, a medium-length ver- ers were sent to nonrespondents, the second with a new
sion to be used by work environment professionals, and questionnaire in case the first had disappeared. The 1858
a short version for workplaces. By developing such a respondents were classified according to job title using
“three-level instrument”, we hoped to achieve the fol- the International Standard Classification of Occupations
lowing objectives: (i) to develop valid and relevant in- (ISCO). In this paper, we report the results on job titles
struments for the assessment of psychosocial factors at using 32 sufficiently homogeneous jobs with 20 re-
work, (ii) to make national and international compari- spondents or more (N=1220; altogether 638 persons had
sons possible, (iii) to improve evaluations of interven- jobs with less than 20 respondents). The results from the
tions, (iv) to facilitate surveillance and benchmarking, study on socioeconomic status and the psychosocial work
(v) to improve the communication between workplac- environment have been published earlier (8).
es, work environment professionals, and researchers,
and (vi) to make it easier for the users to understand
difficult concepts and theories. Selection of scales
The three versions of the COPSOQ questionnaire As indicated, we wanted the COPSOQ scales to cover
were developed on the basis of the following basic prin- some of the main components of relevance for research
ciples and theoretical considerations: (i) the question- and prevention. Table 1 gives an overview of the struc-
naire should be theory-based, but not attached to one ture of the questionnaire.
specific theory (eg, the Job Content Questionnaire), (ii) The first five scales in table 1 are “demand” scales
the questionnaire should consist of dimensions related [1–5]. (During the rest of the article the dimensions of
to different levels of analysis (organization, department, the COPSOQ are referred to with numbers in italics.)
job, person–work interface, and individual), (iii) the These scales are related to the type of production and
questionnaire should include dimensions related to worktasks at the workplace. The next five scales are re-
worktasks, the organization of work, interpersonal rela- lated to the organization of work and to job content [6–
tions at work, cooperation, and leadership, (iv) the ques- 10]. Then follow some scales of relevance for interper-
tionnaire should cover potential work stressors, as well as sonal relations and for leadership [11–18]. The next two
resources such as support, feedback, commitment, and scales are at the person–work interface level [19, 20].
good health, (v) the questionnaire should be comprehen- The health and well-being of the employees are eluci-
sive (ie, there should not be any significant “white spots” dated through the following six scales [21–26], and the
in the picture painted), (vi) the questionnaire should be last four scales measure personality traits [27–30]. Ta-
generic, meaning that it should be applicable in all sec- ble 1 shows the scales included in the final research
tors of the labor market (not only industry, but also the questionnaire. As indicated in the following discussion,
service sector, human service work, and communication), a few changes were made during the analysis phase, but
(vii) the medium-length and short versions should be “user these minor changes did not affect the overall structure
friendly” with regard to work environment professionals of the dimensions of the COPSOQ (See the formation
and respondents (employees). of the scales for the three levels.)
In the following, we present the development of the
three versions of the COPSOQ questionnaire, which was
based on results and analyses of the Danish Psychoso- Selection of questions for the test questionnaire
cial Work Environment Study. Before we started to construct our test questionnaire, we
collected and reviewed some international and Danish
questionnaires in the field. Among the main question-
Study population and methods naires were the Setterlind Stress Profile (9), the White-
hall II questionnaire (10), the Short Form-36 (SF-36)
questionnaire (11), the Dutch QEAW questionnaire
Study population
(12), the QPS-Nordic (13), the Finnish OSQ (14), the
Our analyses were based on responses to a standardized Job Content Questionnaire (15), and some Danish ques-
questionnaire from an age-stratified representative tionnaires previously used by members of our group.

Scand J Work Environ Health 2005, vol 31, no 6 439

kristens.pmd 439 19.12.2005, 13:50

 The Copenhagen psychosocial questionnaire

Table 1. Main scheme for the development of the three versions of the Copenhagen Psychosocial Questionnaire (COPSOQ).

Context and level Scales Sample questions

Workplace
Type of production and tasks 1. Quantitative demands Do you have enough time for your worktasks?
2. Cognitive demands Does your work require that you remember a lot of things?
3. Emotional demands Is your work emotionally demanding?
4. Demands for hiding emotions Does your work require that you hide your feelings?
5. Sensory demands Does your work require that you have very clear and precise eyesight?
Work organization and job content 6. Influence at work Do you have a large degree of influence concerning your work?
7. Possibilities for development Do you have the possibility of learning new things through your work?
8. Degree of freedom at work Can you decide when to take a break?
9. Meaning of work Do you feel that the work you do is important?
10. Commitment to the workplace Do you enjoy telling others about your place of work?
Interpersonal relations and leadership 11. Predictability At your place of work, are you informed well in advance about, for
example, important decisions, changes, or plans for the future?
12. Role clarity Do you know exactly how much say you have at work?
13. Role conflicts Are contradictory demands placed on you at work?
14. Quality of leadership To what extent would you say that your immediate superior is good at
work planning?
15. Social support How often do you get help and support from your colleagues?
16. Feedback at work How often does your superior talk with you about how well you carry out
your work?
17. Social relations Do you work isolated from your colleagues?
18. Sense of community Is there a good atmosphere between you and your colleagues?
Work–individual interface
19. Insecurity at work Are you worried about becoming unemployed?
20. Job satisfaction Regarding your work in general, how pleased are you with the people you
work with?
Individual
Health and well-being 21. General health In general, would you say your health is excellent, very good, good, fair or
poor?
22. Mental health How much of the time during the past 4 weeks have you been a very
nervous person?
23. Vitality How much of the time during the past 4 weeks did you feel worn out?
24. Behavioral stress During the past 4 weeks, I have not had the time to relax or enjoy myself.
25. Somatic stress How much of the time during the past 4 weeks have you had a stomach-
ache or stomach problems?
26. Cognitive stress How much of the time during the past 4 weeks have you found it difficult
to think clearly?
Personality 27. Sense of coherence I believe I can cope with most situations in life.
What do you usually do when problems arise at work?
28. Problem-focused coping - Do you try to find out what you can do to solve the problem?
29. Selective coping - Do you try to think of something else or do something you like?
30. Resignation coping - Do you accept the situation because there is nothing to do about it anyway?

Among the reviewed questionnaires, only the QPS-Nor- of the “statement type”. Along the same line of thought,
dic was close to meeting our needs. This is a modern, we avoided response categories such as “agree” or
comprehensive, and very well validated questionnaire, “strongly disagree”, since such response categories as-
but we missed some dimensions, such as emotional and sume that the questions are about attitudes. Furthermore,
cognitive demands, the meaning of work, job insecurity, we avoided questions in which the stressor and the
job satisfaction, stress, and health. Furthermore, this ques- stress-response were mixed together, such as: “Do you
tionnaire does not have national reference values, and it feel stressed by your work?” or “Is your workpace very
does not exist at three levels. With regard to other widely demanding?” In most cases we preferred five response
used questionnaires, such as the Job Content Question- options to increase precision and reliability and to pro-
naire (15) and the questionnaire on effort–reward imbal- vide a reasonable range of choices for the respondents.
ance (16), they were too limited in scope for our purpose. For most of the questions, we used either intensity (from
With respect to the individual questions, we pre- “to a very large extent” to “to a very small extent”) or
ferred them to be straightforward and simple and for- frequency (from “always” to “never/hardly ever”), pre-
mulated as real questions—not as statements such as “I ferring the format that fit the question. We ended with
get sufficient support from my superior”. Our qualitative 165 questions in the test questionnaire (including 20 new
pilot studies with regard to some of the questionnaires questions formed by us) intended to capture 31 psy-
have indicated that most respondents dislike “questions” chosocial dimensions. [See www.ami.dk/apss for the

440 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 440 19.12.2005, 13:50

 Kristensen et al

complete list of questions.] In addition, we included 13 comparisons with the work of other researchers. More-
questions on background, 10 questions on conflicts, bul- over, our analyses resulted in the abandonment of scales
lying, sexual harassment, and violence, 8 questions on on “demands for responsibility at work” and “repetitive
family-work issues, and 1 on tobacco. work”, and in the formation of a new and unintended
scale on “demands for hiding emotions”. In this way,
we ended up with 30 scales based on 141 items (table 2).
Formation of scales for the three levels
As shown in table 2, the long questionnaire includes
The focus of the COPSOQ is the scales measuring psy- 18 work environment dimensions covering three main
chosocial factors at work, individual health and well-
being, and personality factors such as coping style and
sense of coherence (SOC). The questions on violence, Table 2. Scales, number of questions, and Cronbach’s alphas for
harassment, and family–work interface are optional in the three versions of the Copenhagen Psychosocial Questionnaire
(COPSOQ). (N=1603–1850 for the different scales)
the long and medium-length versions and will not be dis-
cussed in this paper. All the scales in the long and me- Scale Long Medium-length Short
questionnaire questionnaire questionnaire
dium-length questionnaires go from 0 to 100 with high
values representing a high level of the concept being Ques- Cron- Ques- Cron- Ques- Cron-
measured. Therefore a high value on, for example, the tions bach’s tions bach’s tions bach’s
(N) α (N) αa (N) αa
“mental health” scale [22] means good mental health,
while a high value on the “somatic stress” scale [25] 1. Quantitative demands 7 0.80 4 0.65 3⎫
2. Cognitive demands 8 0.86 4 0.78
indicates a high stress level. All of the scales were con-
3. Emotional demands 3 0.87 3 (0.87) 2
structed as simple averages with equal weights for the ⎬ 0.68
4. Demands for hiding 2 0.59 2 (0.59) 1
items and equal intervals between the response options. emotions
A person was considered missing on a scale if fewer 5. Sensory demands 5 0.70 4 0.66 ⎭
than half of the questions in the scale were answered. 6. Influence at work 10 0.83 4 0.73 3⎫
7. Possibilities for 7 0.82 4 0.75 2
development
The long questionnaire 8. Degree of freedom 4 0.68 4 (0.68) 1 ⎬ 0.78
at work
For the construction of the scales for the long question- 9. Meaning of work 3 0.77 3 (0.77) 2
naire, we used principal-component factor analyses and 10. Commitment to the 4 0.74 4 (0.74) 2⎭
workplace b
analyses of internal reliability. In the factor analyses we
11. Predictability 2 0.78 2 (0.78) 2⎫
looked for weak loadings and cross-loadings on other 12. Role clarity b 4 0.77 4 (0.77)
scales. Items with cross-loadings above 0.40 or with 13. Role conflicts b 4 0.72 4 (0.72)
loadings of less than 0.40 were excluded. With regard 14. Quality of leadership b 8 0.93 4 0.87 2
to the internal reliability, we analyzed inter-item corre- 15. Social support 4 0.74 4 (0.74) 2⎬ 0.81

lations and item correlations with the whole scale. (The 16. Feedback at work 2 0.64 2 (0.64) 2
17. Social relations 2 0.65 2 (0.65)
“rules of thumb” were that inter-item correlations should 18. Sense of community 3 0.80 3 (0.80) 2⎭
be between 0.20 and 0.70 and that correlations with the
19. Insecurity at work 4 0.61 4 (0.61) 4 (0.61)
total scale should be above 0.40.) Furthermore, we in-
20. Job satisfaction 7 0.84 4 0.75 4 (0.75)
spected the response distribution of the individual items.
Our guiding principle was to avoid floor or ceiling ef- 21. General health b 5 0.75 5 (0.75) 1 –
22. Mental health b 5 0.80 5 (0.80) 5 (0.80)
fects and also to avoid items with more than 5% miss-
23. Vitality b 4 0.80 4 (0.80) 4 (0.80)
ing values. We started out with 19 work environment
24. Behavioral stress b 8 0.79 4 0.65
scales, 2 scales for the work–individual interface [19, 25. Somatic stress b 7 0.76 4 0.62
20], 6 scales on health/well-being/stress [21–26], and 4 26. Cognitive stress b 4 0.85 4 (0.85)
on personality/coping [21–30]. These scales were based 27. Sense of coherence b 9 0.71
on our theoretical expectations. The factor analyses and
28. Problem-focused 2 0.75
our inspection of the items forced us, however, to coping
change most of the scales. In fact, only four of our orig- 29. Selective coping 2 0.61
inal work environment scales “survived” in the intend- 30. Resignation coping 2 0.66
ed form [10, 12–14]. The two SF-36 scales on mental Number of questions 141 95 44
health and vitality [22, 23] could not be confirmed as Number of scales 30 26 8
separate dimensions in our factor analyses. However, we a For the medium-length and short questionnaires alphas are in
decided to keep the three SF-36 scales [21–23], the three parentheses for the scales that are identical to the scales in the long
stress profile scales [24–26], and the SOC scale [27] un- questionnaire.
b
These scales remained in the final COPSOQ as intended before the
changed in our long questionnaire in order to facilitate statistical analyses.

Scand J Work Environ Health 2005, vol 31, no 6 441

kristens.pmd 441 19.12.2005, 13:51

 The Copenhagen psychosocial questionnaire

domains of the psychosocial work environment: de- forth. In the graphics of the software presentation pro-
mands at work [1–5], work organization and job con- gram, all of the bars between 40 and 60 are yellow, bars
tent [6–10], and interpersonal relations and leadership above 60 are green, and bars below 40 are red. For the
[11–18]. The distinction between five different kinds of nine scales for which the high values are “bad” (“quan-
demands is a specific feature of the COPSOQ. The work– titative demands” [1], “emotional demands” [3], “de-
individual interface dimensions [19, 20] and the health and mands for hiding emotions” [4], “sensory demands” [5],
well-being dimensions [21–26] are normally treated as end “role conflicts” [13], “insecurity at work” [19], and the
points, while the personality dimensions [27–30] are of- three stress scales [24–26]), the red and green colors are
ten analyzed as modifying factors in the associations be- reversed. This program makes it simple for the profes-
tween work characteristics and (health) effects. sionals to spot the “problem areas” (red) and the “re-
source areas” (green). With groups of 20 persons or
more, deviations of 10 points from the median value of
The medium-length questionnaire
50 is normally statistically significant. Therefore all the
The next step was to create the medium-length version comparisons with this software program are with the
of the questionnaire. The reduction in the number of Danish national average level. If other normative levels
items was accomplished in two ways. First, the four in- are wanted, the program has to be changed accordingly.
dividual-level scales on coping and SOC were exclud-
ed since the medium-length questionnaire is to be used
for the evaluation of workplaces and jobs and not per- The short questionnaire
sons. Second, we reduced the length of all the longer In order to construct the short questionnaire, we entered
scales to a maximum of four items (with the exceptions all of the 18 work environment scales of table 2 into a
of the two 5-item scales from the SF-36 [21, 22]). In new factor analysis. This time, the following three main
this reduction process we tried to keep the new scales as clusters of scales emerged: (i) demands at work, (ii)
broad as the original scales and not per se to get as high work organization and content, and (iii) interpersonal
an internal reliability as possible. Our analyses showed relations and leadership. We decided to pick items from
that the reduced scales explained 85–96% of the varia- the middle-length questionnaire that represent these
tion of the longer scales. The medium-length questionnaire three main dimensions of work. This selection of items
consists of 95 questions forming 26 scales (table 2). was done by two of us (TSK and VB) on a theoretical
Our next step was to develop a user-friendly soft- basis, and it was not guided by statistical procedures.
ware system for the medium-length questionnaire to be Furthermore, we chose to reduce the number of items
used by work environment professionals. In this system, by not including the three stress scales [24–26] and by
all the scales have been transformed into adjusted scales using just one item to assess general health [21]. In this
with a median value of 50 points in our representative way, the short questionnaire ended up consisting of 44
database. Furthermore, 60 points corresponds to the questions forming 8 scales (table 2).
60% percentile, 90 points to the 90% percentile, and so In the user-friendly version of the short question-
naire, we developed a simple scoring system, which
makes it possible to calculate the average scores of de-
Table 3. Origin of the questions included in the test question-
naire and in the three final versions of the Copenhagen Psycho-
partments or workplaces for the eight dimensions. In the
social Questionnaire (COPSOQ). (SF-36 = Short Form-36, QEAW instruction pamphlet, we explain how to interpret and
= questionnaire on the experience and assessment of work, QPS evaluate these scores.
Nordic = General Nordic Questionnaire, OSQ = Occupational
Stress Questionnaire)
Origins of the questions used at the three levels
Original questionnaire COPSOQ questionnaire
In table 3 we show the source of the questions used for
Test Long Medium Short
the three versions of the questionnaire. The stress pro-
Stress profile (9) 33 32 16 3 file items include the three stress scales [24–26] and the
Whitehall II (10) 30 28 16 12 scale on SOC [27] (9). The Whitehall II questions in-
SF-36 (11) 14 14 14 10 clude the scale on job satisfaction [20] and some items
QEAW (Dutch) (12) 11 11 10 2
on support, influence, and job demands (10). The SF-
QPS Nordic (13) 9 6 4 1
OSQ (Finnish) (14) 2 2 2 1 36 questions are the three scales on “general health”,
Job Content Questionnaire (15) 1 1 0 0 “mental health”, and “vitality” [21–23] (11). The Dutch
Danish questionnaires 45 34 24 9 QEAW contributed with items on emotional demands,
New questions 20 13 9 6 cognitive demands, and other work-related factors (12).
Total 165 141 95 44 All of the other questionnaires are represented with less
than 10 questions in our long questionnaire.

442 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 442 19.12.2005, 13:51

 Kristensen et al

Results impression that many respondents felt that they could

have used more than these two options.
In this section we (i) present the basic structure of the
COPSOQ questionnaires, (ii) bring coefficients of in- Validity of the questionnaire and scales
ternal reliability, average scores, standard deviations and The basic question with regard to validity is “Do we meas-
proportions of missing values of the scales, (iii) analyze ure what we intend to measure?” Since the COPSOQ
intercorrelations between the scales, and (iv) present two comprises 141 questions forming 30 scales of different
psychosocial job-exposure matrices in order to elucidate lengths, it is impossible within the limits of one article
the validity. to go into any depth about the individual scales. Many
of the scales were, in fact, analyzed very carefully by
the researchers who developed them or used them in
Psychometrical characteristics their research (9–11). Instead, in this article, we con-
The Cronbach’s alpha for internal reliability was lower centrate on the validity of the questionnaire by analyz-
than the conventional 0.70 for seven of the research ing the correlations between the scales and constructing
scales, but most of these scales are short (2–3 items),
and therefore the probability of low alphas was in-
Table 4. Average scores and standard deviations of the Copen-
creased (table 2). The item correlations with the total
hagen Psychosocial Questionnaire (COPSOQ) scales. (N=1603–
scale were higher than 0.40 for five of the cases with 1850 for the different scales)
low alphas. The two scales with low item correlations
with the total scale are “degree of freedom at work” [8] Scale Long Medium-length Short
questionnaire questionnaire questionnaire
and “insecurity at work” [19]. As has already been men-
tioned, high internal reliability was not a goal in the de- Mean SD Mean SD Mean SD
velopment of the two other versions of the questionnaire. 1. Quantitative demands 44.3 17.9 46.8 18.6 ⎫
In table 4, the average scores and the standard devi- 2. Cognitive demands 62.8 19.3 62.9 21.5
ations are shown for all of the scales. It follows directly 3. Emotional demands 37.8 25.5 37.8 25.5
from the preceding remarks that we tried to avoid scales 4. Demands for hiding 30.3 22.6 30.3 22.6 ⎬ 42.0 16.8
emotions
with very high or low average scores. The table shows 5. Sensory demands 63.9 21.0 62.4 23.2
⎭
that we succeeded with the work environment scales, 6. Influence at work 62.8 18.8 55.4 24.2 ⎫
since 15 of the 18 long work environment scales had 7. Possibilities for 68.5 18.4 72.4 19.1
average levels between 25 and 75. The three exceptions development
were “meaning of work” [9], “role clarity” [12], and 8. Degree of freedom 64.8 25.3 64.8 25.3 ⎬ 65.3 16.6
at work
“sense of community” [18] with high values reflecting 9. Meaning of work 77.7 16.4 77.7 16.4
positive assessments by the Danish respondents. The SF- 10. Commitment to the 57.1 21.4 57.1 21.4
workplace ⎭
36 scales [21–23] also had relatively high average val-
ues (indicating good levels of health status), and the Set- 11. Predictability 60.4 23.3 60.4 23.3 ⎫
terlind stress scales [24–26] had low average values (in- 12. Role clarity 75.7 14.9 75.7 14.9
13. Role conflicts 37.3 18.5 37.3 18.5
dicating relatively few of the stress symptoms measured).
14. Quality of leadership 56.7 20.8 55.5 21.4
In table 5 we show the proportion of missing values 15. Social support 68.3 20.4 68.3 20.4 ⎬ 58.9 15.9
for the individual questions and the scales. For 24 of 16. Feedback at work 39.2 23.3 39.2 23.3
the 30 scales in the long questionnaire, the missing val- 17. Social relations 67.7 30.1 67.7 30.1
ues made up less than 5%. Of the six scales with higher 18. Sense of community 82.0 17.4 82.0 17.4 ⎭
levels of missing values, five were about interpersonal 19. Insecurity at work 16.9 25.1 16.9 25.1 16.9 25.1
relations and leadership. Closer inspection showed that 20. Job satisfaction 69.3 15.6 68.2 16.7 68.2 16.7
almost all of the items with many nonrespondents dealt 21. General health 80.9 16.9 80.9 16.9 75.9 19.8
with leadership and supervisors. Other questions as- 22. Mental health 79.0 15.1 79.0 15.1 79.0 15.1
sumed the presence of colleagues in the daily work of 23. Vitality 64.1 18.8 64.1 18.8 64.1 18.8
the respondent. It seems obvious that many respondents 24. Behavioral stress 15.5 15.0 17.1 17.4
did not answer these questions because they found them 25. Somatic stress 16.0 14.2 19.1 16.1
26. Cognitive stress 20.0 18.5 20.0 18.5
irrelevant for their daily work situation. We did have a
response category of “irrelevant” for most of these 27. Sense of coherence 81.9 13.3

items, but many respondents did not tick the appropri- 28. Problem-focused 83.4 16.7
ate box. The scale for “insecurity at work” [19] also had coping
29. Selective coping 34.9 23.1
many missing values. This was the only scale based on 30. Resignation coping 29.0 22.5
questions with “yes-no” response options. It is our

Scand J Work Environ Health 2005, vol 31, no 6 443

kristens.pmd 443 19.12.2005, 13:51

 The Copenhagen psychosocial questionnaire

Table 5. Proportion of missing values for the individual ques- “job-exposure matrices” (17, 18) with some of the newly
tions and the scales of the long version of the Copenhagen Psy- developed scales.
chosocial Questionnaire (COPSOQ).
A sign of low validity would be that the scales meant
Scale Range of Values miss- to measure different theoretical constructs were highly
values miss- ing for the
ing for the scale whole scale
correlated. Normally, a correlation above 0.70 between
items (%) (%) two scales is considered “too high” (19). In table 6 we
show the correlations between the 18 work environment
1. Quantitative demands 1.6–4.1 3.6
2. Cognitive demands 3.7–4.7 3.7 scales [1–18]. The highest correlation between the de-
3. Emotional demands 3.1–3.9 3.0 mand scales [1–5] is between “emotional demands” [3]
4. Demands for hiding emotions 3.3–4.0 2.5 and “demands for hiding emotions” [4] (r=0.46). Among
5. Sensory demands 3.8–4.8 3.8
the five scales on work organization and content [6–10]
6. Influence at work 3.6–4.1 3.5 the highest correlation was between “possibilities for
7. Possibilities for development 3.0–4.3 2.8
8. Degree of freedom at work 3.7–3.9 3.4
development” [7] and “meaning of work” [9] (r=0.54).
9. Meaning of work 2.9–3.2 3.0 Finally, the highest correlation between the scales on
10. Commitment to the workplace 3.0–3.3 2.9 interpersonal relations and leadership [11–18] occurred
11. Predictability 2.2–4.3 2.2 between “predictability” [11] and “quality of leader-
12. Role clarity 4.3–4.7 4.3 ship” [14] (r=0.59). It is noteworthy that the highest
13. Role conflicts 4.2–4.7 4.3 correlation between all of the 18 work environment
14. Quality of leadership 9.9–10.5 13.6
15. Social support 3.4–6.2 6.1
scales is between “cognitive demands” [2] and “possi-
16. Feedback at work 3.6–6.1 7.1 bilities for development” [7] (r=0.63).
17. Social relations 5.7–5.9 5.6 Table 7 shows the correlations between the 12 re-
18. Sense of community 5.8–5.9 8.9 maining scales [19–30]. The highest correlation among
19. Insecurity at work 5.7–6.2 5.8 all of the scales in the questionnaire was that between
20. Job satisfaction 4.5–4.8 5.2 the two SF-36 scales “mental health” [22] and “vitali-
21. General health 0.4–3.3 1.0 ty” [23] (r=0.73). This result fits well with the factor
22. Mental health 1.7–3.3 1.5 analysis in which these two scales could not be identi-
23. Vitality 1.0–2.6 0.8 fied as two separate dimensions. The second highest
24. Behavioral stress 0.9–2.5 1.1 correlation was between “mental health” [22] and “be-
25. Somatic stress 1.7–2.6 1.3
havioral stress” [24] (r=–0.64). As has already been
26. Cognitive stress 2.2–2.7 1.3
mentioned, we chose to keep these scales as originally
27. Sense of coherence 1.0–3.6 1.6
formulated to facilitate international comparisons.
28. Problem focused coping 3.2–3.6 3.1 As indicated in the introduction, one of our objec-
29. Selective coping 3.9–4.7 3.3
30. Resigning coping 3.4–3.7 3.3 tives was to create a questionnaire that would be able
to reflect important aspects of industrial work, as well

Table 6. Correlations between the 18 work environment scales in the long version of the Copenhagen Psychosocial Questionnaire
(COPSOQ).

Scale Spearman correlation coefficients

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1. Quantitative demands +
2. Cognitive demands .34 +
3. Emotional demands .25 .45 +
4. Demands for hiding emotions .21 .23 .46 +
5. Sensory demands .12 .31 .12 .14 +
6. Influence at work –.04 .43 .07 –.16 –.01 +
7. Possibilities for development .13 .63 .19 –.02 .11 .48 +
8. Degree of freedom at work –.02 .14 –.21 –.23 –.08 .49 .27 +
9. Meaning of work –.03 .39 .10 –.10 .18 .43 .54 .19 +
10. Commitment to the workplace .04 .37 .36 –.01 .10 .45 .41 .21 .51 +
11. Predictability –.12 .20 –.05 –.16 –.01 .44 .31 .27 .35 .42 +
12. Role clarity –.12 .23 –.07 –.12 .16 .43 .28 .21 .46 .39 .49 +
13. Role conflicts .30 .18 .28 .31 .05 –.18 –.05 –.16 –.19 –.15 –.31 –.34 +
14. Quality of leadership –.23 .09 –.06 –.19 .02 .34 .28 .21 .37 .38 .59 .41 –.33 +
15. Social support –.19 .04 –.09 –.20 .04 .19 .18 .18 .25 .20 .36 .28 –.22 .54 +
16. Feedback at work –.04 .14 .03 –.09 .03 .23 .19 .17 .16 .20 .25 .16 –.07 .33 .34 +
17. Social relations –.04 –.04 –.16 –.16 –.06 –.01 .05 .19 .01 –.02 .05 –.01 –.06 .10 .25 .09 +
18. Sense of community –.16 .08 –.13 –.20 .09 .28 .21 .22 .34 .32 .33 .36 –.27 .45 .45 .19 .26 +

444 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 444 19.12.2005, 13:51

 Kristensen et al

Table 7. Correlations between the scales on health, well-being, satisfaction, insecurity, and coping in the long version of the Copenha-
gen Psychosocial Questionnaire (COPSOQ).

Scale Spearman correlation coefficients

19 20 21 22 23 24 25 26 27 28 29 30

19. Insecurity at work +

20. Job satisfaction –.17 +
21. General health –.17 .28 +
22. Mental health –.15 .36 .43 +
23. Vitality –.12 .35 .45 .73 +
24. Behavioral stress .14 –.26 –.31 –.64 –.61 +
25. Somatic stress* .17 –.30 –.46 –.50 –.52 .46 +
26. Cognitive stress .18 –.31 –.36 –.54 –.53 .49 .54 +
27. Sense of coherence –.31 .36 .32 .46 .42 –.39 –.37 –.43 +
28. Problem focused coping –.09 .30 .11 .23 .21 –.17 –.14 –.22 .28 +
29. Selective coping .16 –.11 –.07 –.13 –.12 .11 .11 .13 –.22 –.26 +
30. Resigning coping .16 –.25 –.11 –.17 –.17 .17 .17 .19 –.31 –.35 .33 +

Points

Emotional
demands Work with
70 clients
Elementary school teachers
Home helpers
Nurses Nurses’ aids

60
Kindergarten teachers

50
High school teachers

Managers
40
Foremen
Work with Drivers
things or Salesmen
30 symbols
Mechanics

Accountants Work with

Shop assistants
Metal workers customers or
colleagues
20 Construction workers System planners

Food indtustry workers

10
Demands for
hiding emotions
Points
15 20 25 30 35 40 45 50 55
Figure 1. Distribution of the 32 jobs on the scales for emotional demands and demands for hiding emotions.

as information work and human service work. The two matrix” showed a very clear picture. Jobs with client
scales on “emotional demands” [3] and “demands for work (patients, students, children, etc) scored high on
hiding emotions” [4] were particularly relevant in con- both dimensions, while customer work and work with
nection with human service work. In figure 1 we show colleagues scored high on hiding emotions and medium
the distribution of jobs on these two “emotion scales”. high on emotional demands. Finally, jobs implying
The distribution of jobs in this “emotional job-exposure work with physical objects or symbols (computers,

Scand J Work Environ Health 2005, vol 31, no 6 445

kristens.pmd 445 19.12.2005, 13:51

 The Copenhagen psychosocial questionnaire

communication) scored low on both dimensions. This on the same analyses and basic theoretical assumptions,
distribution of jobs in a matrix based on the two very (ii) the COPSOQ is a concept aiming at describing a
brief “emotion scales” is theoretically meaningful and large number of relevant factors within the field of psy-
supports the construct validity of the two scales. chosocial work environment, health, well-being, and
In the final example, we illustrate the construct va- personality (iii) the COPSOQ includes five different
lidity of two of the scales from the short version of the demand dimensions, including emotional and cognitive
questionnaire, “work organization and content” and “in- demands, (iv) the COPSOQ includes a medium-length
terpersonal relations and leadership”. In figure 2 we and a short questionnaire, each with its own user-friendly
show the distribution of the 32 jobs in our sample in a scoring system, and (v) the COPSOQ is based on com-
matrix consisting of these two dimensions. The figure parisons with national average scores.
also shows jobs with high (top 10), medium (middle 12), The analyses presented in this article show that most
and low (bottom 10) levels of job satisfaction. Jobs with of the COPSOQ scales have good internal reliability and
high values on both dimensions were kindergarten that the correlations between the scales are moderate to
teachers and managers, while jobs with low values are low, and these levels of correlation indicate that the
food-industry workers, cleaners, and drivers. The fig- scales measure different aspects of the work environ-
ure also demonstrates a very close association between ment or the well-being of workers. As has already been
position in the matrix and job satisfaction. The “satis- mentioned, the main exception is the two SF-36 scales
fying jobs” are in the upper right part, while the “unsat- on “mental health” [22] and “vitality” [23]. The scales
isfying jobs” are in the lower left part of the figure. This are also able to differentiate between groups (such as
positioning indicates a high correlation between the two jobs or socioeconomic groups) within the labor market.
dimensions of psychosocial work environment and the (For more details see www.ami.dk/apss.) The issue of
level of job satisfaction. validity is complicated and difficult to analyze once and
for all. Some of our analyses, such as the results con-
cerning job satisfaction in figure 2, seem to indicate that
Discussion our scales really capture what we intended to measure,
but we have also found signs of low validity for some
The purpose of our paper is to present the COPSOQ and of our items or scales. In the years to come, we will be
to discuss its special features, its weaknesses and strengths. using the COPSOQ scales in many prospective studies
To our knowledge COPSOQ is unique in the following being carried out the Danish National Institute of Oc-
ways: (i) the COPSOQ consists of questionnaires at three cupational Health, and these studies will enable us to
levels with different lengths and complexity, but based be more definite about the validity issue and also to shed

10 jobs with highest job satisfaction

12 jobs with medium job satisfaction

Points
Î 10 jobs with lowest job satisfaction
Work organization
and content Architects Managers
75 System planners Kindergarten teachers
High school teachers Foremen Vocational Salesmen
school teachers
70
Warehouse assistants Î Bank clerks Kindergarten assistants
65 Technicians Î Nurses’ aids Î Î
Cooks
60 Constuction workersÎ

Î Office clerks
Cleaners Î
55 Drivers Î Î
Shop assistants

50

45 Î Food industry workers Interpersonal relations

and leadership
Points
55 60 65 70
Figure 2. Distribution of jobs on the two dimensions of “work organization and content” and “interpersonal relations and leadership” in the short
version of the questionnaire. Typical jobs are shown for each of the three types of work. For both scales high-quality work environment conditions
are indicated with high scores. Dotted lines indicate average scores.

446 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 446 19.12.2005, 13:51

 Kristensen et al

light on the predictive validity of the scales. In particu- dimensions related to modern worklife. Among these
lar, the use of COPSOQ scales in the Danish Work En- dimensions are cognitive demands, emotional demands,
vironment Cohort Study (20) will be one of the corner- demands for hiding emotions, predictability, role clari-
stones of this work. ty, and role conflicts. We find these dimensions impor-
As stated in the introduction, we wanted COPSOQ tant for describing and analyzing (human) service work,
to be theory-based without being based on one specific flexible work, and group work. Furthermore, these di-
theory. This apparent self-contradiction deserves some mensions seem to play an important role in analyses of
further comments. In a recent overview of the theories outsourcing, downsizing, and other organizational trends
on psychosocial factors at work, Kompier analyzed the in today’s globalized labor market. A striking example
following seven influential theories (5): (i) the job char- of the shortcomings of the “old questionnaires” is found
acteristics model, (ii) the Michigan organizational stress in the burnout literature. Burnout is usually conceptual-
model, (iii) the demand–control–(support) model, (iv) ized as a reaction to demands in connection with doing
the sociotechnical approach, (v) the action–theoretical “people work”. But, as pointed out by Zapf (22), hardly
approach, (vi) the effort–reward–imbalance model, and any studies on burnout have tried to measure the spe-
(vii) the vitamin model. One of Kompier’s main points cific factors related to this type of work, such as emo-
is that these seven theories, or models, include several tional demands, demands for hiding emotions, and role
of the same psychosocial factors at work. Skill variety, conflicts. By restricting themselves to the “well validat-
autonomy and demands play a part in six of the theo- ed” questionnaires, almost all burnout researchers have
ries, social support is a factor in four, and three of the missed the very specific dimensions of interest in the
theories include feedback, job future ambiguity and task development of burnout.
identity. In COPSOQ the first six of these concepts cor- As stated in the introduction, we also wanted the
respond closely to possibilities for development, influ- questionnaire to cover different levels of analysis. The
ence, quantitative demands, social support, feedback, final questionnaire includes questions at the following
and job insecurity. The only missing factor seems to be levels: (i) individual (health, coping [21–30]), (ii) indi-
task identity (the degree to which a job requires com- vidual–work interface (satisfaction, insecurity [19, 20]),
pletion of a whole and identifiable piece of work). This (iii) job (eg, influence, demands, possibilities for devel-
concept, however, comes close to the COPSOQ concept opment [1–7]), (iv) department (eg, quality of leader-
of meaning of work. Therefore, the COPSOQ includes ship, predictability, feedback [11, 14, 16]), and (v)
substantial parts of the main dimensions of the sev- workplace or organization (commitment to workplace,
en major theories in occupational health psycholo- sense of community [10, 18]). It was also an aim to in-
gy. clude both appreciative and problem-oriented dimen-
From a theoretical point of view the major shortcom- sions. Some of the appreciative dimensions are commit-
ings of COPSOQ seem to be related to the effort–re- ment to the workplace, social support, and sense of com-
ward–imbalance model. This shortcoming was actually munity [10, 15, 18], while some of the more problem
pointed out to us by kindergarten teachers, who did not oriented are quantitative demands, emotional demands,
understand how we could claim that they have the best and role-conflicts [1, 3, 13]. In practice many of the di-
psychosocial work environment in Denmark. They felt mensions may be perceived as both appreciative and
that they have a poor psychosocial work environment, problem-oriented, depending on the results of a specif-
and our “solution” to this paradox is that these workers ic study.
receive rather low rewards (salary, esteem, and career While the theoretical validity is a central issue, the
possibilities). This explanation is in accordance with the “tests of real life” should also be taken into considera-
Siegrist effort–reward–imbalance model (21). COPSOQ tion. We have now been using the questionnaires for a
does include aspects of rewards, such as feedback, job few years, and the concept has been used by many re-
(in)security, and social support, but esteem, career pos- searchers, a large number of work environment profes-
sibilities, and monetary rewards are missing in the ques- sionals, hundreds of workplaces, and thousands of em-
tionnaire. Other aspects of work that may be missing in ployees in Denmark. Moreover, the questionnaires have
the present version of COPSOQ have to do with values been or are in the process of being translated into seven
at work, such as justice, trust, and discrimination. languages. In quantitative terms our success has been
Thus it is our conclusion that COPSOQ includes much bigger than anticipated, which is one of the rea-
most of the important dimensions in the main theories sons for this rather late presentation of the concept in
of the psychosocial work environment. In this sense, we the international literature.
find it justified to call the questionnaire “theory-based”. The wide usage of COPSOQ has enabled us to look
It should, however, be stressed that our ambition has not at a very important aspect of the validity of a work en-
only been to cover some of the main factors of main- vironment questionnaire, the face validity: Do the em-
stream models from the last 50 years, but also to include ployees and employers who use the questionnaire feel

Scand J Work Environ Health 2005, vol 31, no 6 447

kristens.pmd 447 19.12.2005, 13:51

 The Copenhagen psychosocial questionnaire

that the dimensions are important and relevant? And can wording of a few items. These and other problems will
they recognize themselves and their own workplace be addressed in a future revision of the concept. In or-
when they see the results? Thus far, we have received der to avoid confusion due to many consecutive ver-
feedback from a large number of Danish users and also sions, we have decided to make major revisions at in-
from our colleagues in Spain, who are developing the tervals of at least 5 years.
Spanish version. The main impression is that the peo- Two more general issues have often been raised in
ple at workplaces find the dimensions relevant and im- connection with COPSOQ. One has to do with the use
portant, but also that they have to “get used to” many of generic questionnaires, the other with the use of na-
of the new concepts before they are able to understand tional average normative values. The use of generic
the results. This will necessarily be a learning process questionnaires, in which the same questions are used in
for all the parties involved. a shipyard and in a research institution, has been criti-
Our own analyses and our contacts with the users cized by many researchers, as well as by work environ-
have made us aware of several problems. One of them ment professionals (24, 25). It has been pointed out that
has to do with the scale for quantitative demands [1]. the questionnaire creates an artificial world in which
We have discovered that this scale hides important fea- very different things can be measured with the same
tures of quantitative demands by mixing the two funda- yardstick. This characteristic also means that the re-
mental aspects of these demands: number of workhours spondents may be forced to answer questions with very
and workpace (extensive and intensive demands). Since little relevance to themselves. In the final interpretation
extensive and intensive quantitative demands character- the people at the workplace have difficulties recogniz-
ize different job groups and since these two aspects of ing their own work situation, and they may feel that the
quantitative demands may have very different conse- “problems” displayed in the report are not their prob-
quences, this seems to be a major problem with this lems. The contrast to the generic questionnaire is the
scale. A paper on this topic has been published (23). specific context-based questionnaire grounded in the
Some of the individual questions have also been crit- individual workplace and developed together with the
icized. An example is the question “Would you like to people who work there. Such questionnaires can be
stay at your current place of work for the rest of your much closer to the everyday problems of the workplace
worklife?” (in the scale on “commitment to the work- and include very specific and local questions.
place” [10]). This item has a strong age bias, as any stu- The problem with the very specific and local ques-
dent could have told us. Apparently, we were too old to tionnaires is that people often ask the researcher: “Is this
see the problem. Two other examples are the questions good or bad? How are we doing compared with other
on “feedback at work” [16]. These are phrased as fol- workplaces?” Such questions can only be answered by
lows: “How often do you talk with your superior (col- using generic questionnaires with normative compara-
leagues) about how well you carry out your work?” tive values. With COPSOQ we have chosen the nation-
Many feel that a more appropriate wording would be: al average as the basis for comparison. This use too has
“How often does your superior (colleagues) talk with been criticized by many users. The two main arguments
you about how well you carry out your work?” Finally, have been (i) “Who says that the national average is
we will have to address the issue of supervisors and good? Shouldn’t we use a more fundamental criterion
leadership. Many people today work in loose networks, such as the risk of ill health?”, and (ii) “Why are we
in changing groups with changing leadership, have sev- being compared with the national average? We would
eral leaders at the same time, or have no leader. This like to be compared with a more relevant comparison
type of situation makes it increasingly difficult to an- group such as our own industry.”
swer questions assuming a “normal” hierarchy at the Basically, all this criticism is relevant and important.
workplace. The high level of missing values on several With regard to the issue of context, we have chosen to
of the questions dealing with leadership shown in table construct the medium-length version of the question-
5 demonstrates a problem that will only get bigger in naire in a way that allows the workplaces to add specif-
the years to come. ic and local questions at the end of the questionnaire.
All in all, we find it reasonable to conclude that we For these questions there are no national comparison
have achieved the goals described in the introduction. values, but the inclusion of local questions seems to be
The questionnaire is theory-based without being at- a very good supplement to the generic part. Many work-
tached to a specific model, it covers most of the rele- places have used this option. Regarding the issue of na-
vant areas in the psychosocial work environment, it in- tional values, we have decided to develop industry-spe-
cludes dimensions at different levels that are apprecia- cific normative values for future versions of COPSOQ.
tive as well as problem-oriented. We have identified These values will enable workplaces to choose between
shortcomings with regard to missing dimensions, low three levels of comparisons: their own workplace, their
levels of internal reliability on a few scales, and the own industry, or the country.

448 Scand J Work Environ Health 2005, vol 31, no 6

kristens.pmd 448 19.12.2005, 13:51

 Kristensen et al

International collaboration al study. Scand J Public Health 2002;30:41–8.

9. Setterlind S, Larson G. The stress profile: a psychosocial
Groups working with COPSOQ have now been estab- approach to measuring stress. Stress Med 1995;11:85–92.
lished in six other countries. These groups have, in all 10. Marmot MG, Smith GD, Stansfeld S, Patel C, North F, Head
cases, been initiated by local researchers or profession- J, et al. Health inequalities among British civil servants: the
als who have contacted us for collaboration. We have Whitehall II study. Lancet 1991;337:1387–93.
11. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 health
tried to provide the necessary pieces of information and
survey: manual and interpretation guide. Boston (MA): The
help for the translation and validation procedures. We Health Institute, New England Medical Center; 1993.
have also tried to establish a situation in which there is 12. van Veldhoven M, Meijman TF. Questionnaire on the experi-
one—and only one—validated COPSOQ translation in ence and assessment of work: VBBA—English version. Am-
each of the countries involved. The national contact per- sterdam: The Foundation for Quality in Occupational Health
sons can be found at the website www.ami.dk/apss. It Care; 1994.
13. Dallner M, Elo A-L, Gamberale F, Hottinen V, Knardahl S,
is our hope that this collaboration will grow into an in-
Lindström K, et al. Validation of the General Nordic Ques-
ternational network in the years to come, not only tionnaire (QPSNordic) for psychological and social factors at
among researchers, but also between researchers and work. Copenhagen: Nordic Council of Ministers, 2000. Nord
practitioners from the occupational health system and 12.
from workplaces. 14. Elo A-L, Leppänen A, Lindström K, Ropponen T. OSQ,
Occupational Stress Questionnaire: user’s instructions. Hel-
sinki: Finnish Institute of Occupational Health; 1992.
15. Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P,
Amick B. The Job Content Questionnaire (JCQ): an instru-
Acknowledgments
ment for international comparative assessments of psychoso-
cial job characteristics. J Occup Health Psychol 1998;3:322–
The study was financially supported by the Service 55.
Center of the Danish Work Environment Council. 16. Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Nied-
hammer I, et al. The measurement of effort-reward imbalance
at work: European comparisons. Soc Sci Med 2004;58:1483–
99.
17. Johnson JV, Stewart WF. Measuring work organization expo-
References
sure over the life course with a job-exposure matrix. Scand J
Work Environ Health 1993;19:21–8.
1. European Commission. Guidance on work-related stress: spice 18. Schwartz JE, Pieper CF, Karasek RA. A procedure for linking
of life or kiss of death? Luxembourg: Office for Official psychosocial job characteristics data to health surveys. Am J
Publications of the European Communities; 2000. Public Health 1988;78:904–9.
2. Cox T, Griffiths A, Rial-González E. Research on work- 19. Nunnally JC, Bernstein IH. Psychometric theory. New
related stress. Luxembourg: Office for Official Publications York(NY): McGraw-Hill, Inc; 1994.
of the European Communities; 2000. 20. Burr H, Bjorner JB, Kristensen TS, Tüchsen F, Bach E. Trends
3. Kompier M. The psychosocial work environment and health— in the Danish work environment in 1990–2000 and their asso-
what do we know and where should we go? [editorial]. Scand ciations with labor-force changes. Scand J Work Environ
J Work Environ Health 2002;28(1):1–4. Health 2003;29(4):270–9.
4. Paoli P, Merllié D. Third European survey on working condi- 21. Siegrist J. Adverse health effects of high-effort/low-reward
tions 2000. Dublin: European Foundation for the Improve- conditions. J Occup Health Psychol 1996;1:27–41.
ment of Living and Working Conditions; 2001. 22. Zapf D, Vogt C, Seifert C, Mertini H, Isic A. Emotion work as
5. Kompier M. Job design and well-being. In: Schabracq MJ, a source of stress: the concept and development of an instru-
Winnubst JAM, Cooper CL, editors. The handbook of work ment. Eur J Work Organ Psychol 1999;8:371–400.
and health psychology. Chichester (UK): John Wiley & Sons; 23. Kristensen TS, Bjorner JB, Christensen KB, Borg V. The
2003. p 429–54. distinction between work pace and working hours in the meas-
6. European Agency for Safety and Health at Work. Future urement of quantitative demands at work. Work Stress
occupational safety and health research needs and priorities in 2004;18:305–22.
the member states of the European Union. Luxembourg: Of- 24. Cox T, Griffiths AJ, Barlow CA, Randall RJ, Thomson LE,
fice for Official Publications of the European Communities; Rial-Gonzáles E. Organisational interventions for work stress:
2000. a risk management approach. Sudbury(): HSE Books, 2000.
7. Marklund S. Questionnaire survey on research priorities. In: 25. Hane M, Wennberg B-Å. Frågeformulärens kris: varför vi
Westerholm P, Marklund S, editors. Strategies for occupa- övergett frågeformulär som hjälpmedel i arbetsmiljöarbetet
tional health research in a changing Europe. Stockholm: Na- [The crisis of the questionnaire: why we have given up using
tional Institute for Working Life; 2000. p 31–3. Arbete och the questionnaire as a tool in occupational health]. Göteborg
Hälsa, vol 12. (Sweden): MediaGraphic; 2002.
8. Kristensen TS, Borg V, Hannerz H. Socioeconomic status and
psychosocial work environment: results from a Danish nation- Received for publication: 14 April 2004

Scand J Work Environ Health 2005, vol 31, no 6 449

kristens.pmd 449 19.12.2005, 13:51