You are on page 1of 6

How to Conduct Research on Burnout: Advantages and Disadvantages of a Unidimensional

Approach in Burnout Research

Author(s): V. Brenninkmeijer and N. Van Yperen
Source: Occupational and Environmental Medicine, Vol. 60, Supplement 1: Fatigue at Work (Jun.,
2003), pp. i16-i20
Published by: BMJ
Stable URL:
Accessed: 24-12-2015 20:26 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact

BMJ is collaborating with JSTOR to digitize, preserve and extend access to Occupational and Environmental Medicine.

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions

How to conduct research on burnout: advantages and

disadvantages of a unidimensional approach in burnout
V Brenninkmeijer, N VanYperen

Occup EnvironMed 2003;60(Suppl I):?16-?20

See end of article for

one should report results
authors' affiliations When conducting research on burnout, itmay be difficult to decide whether
for each burnout dimension or whether one should combine the dimensions. Although the
multidimensionality of the burnout concept iswidely acknowledged, for research purposes it is some
Correspondence to:
Dr V Brenninkmeijer, TNO times convenient to regard burnout as a unidimensional construct. This article deals with the question of
Work & Employment, Team whether and when itmay be appropriate to treat burnout as a unidimensional variable, and presents
UBS2 (Social Security: a decision rule to distinguish between people high and low in burnout. To develop a guideline for
Policy and Performance),
PO Box 718, 2130 AS obtaining a dichotomous measure of burnout, the scores on the Utrecht Burnout Scale (UBOS) of 44
well functioning individuals were compared with the scores of 29 individuals diagnosed as suffering "
from burnout. Based on these the authors recommend the "exhaustion + 1 criterion for research
v.brenninkmeijer@ data, in non-clinical populations. Following this criterion, individuals can be considered as burnt out when
they report, compared to a norm group, high emotional exhaustion, in combination with high deper
Accepted 12 September
2002 sonalisation or low personal accomplishment. The criterion may be used to estimate the percentage in
a sample of individuals in a state of burnout.

Burnout, a state of mental exhaustion resulting from the burnout syndrome, researchers generally report results
chronic stress in the working situation, is a "hot" issue. separately for each dimension.
In the mass in television When conducting research on burnout, itmay be difficult to
media, programmes, magazines,
and burnout receives a considerable decide whether one should report results separately for each
newspapers, currently
amount of attention. In scientific as the burnout dimension or whether one should combine the
journals well,
literature on burnout is abundant: in August 2002, dimensions. Although it is on conceptual grounds preferable
literature database of the to treat burnout as a multidimensional construct, for
the term "burnout" in psycINFO, the
researchers in the field of burnout it is sometimes more con
American Psychological Association, resulted in 3153 hits.
venient to treat burnout as a unidimensional variable. The
Burnout is commonly regarded as a syndrome comprising
purpose of this article is to offer guidelines about the presen
three dimensions.12 The first, most central aspect is emotional
tation of results for researchers in the field of burnout, in par
exhaustion.3 One feels drained and "worn out" or
ticular for those who areemploying the Maslach Burnout
The second symptom, depersonalisation, refers to a
Inventory. It deals with the question of whether and when it
attitude towards one's work or towards the
cynical, negative con
may be appropriate to treat burnout as a unidimensional
recipients of one's service. For instance, teachers may make
struct, rather than reporting results separately for the under
comments about their pupils or physicians about their
cynical attention is paid to the pros and
lying dimensions. Moreover,
patients. Finally, the third symptom is reduced of burnout
cons of burnout as a dichotomous versus a
a reduced sense of conceptualising
personal accomplishment, which denotes
continuous variable. Finally, we present an empirical study in
competency in comparison to one's past functioning. The
which we developed a decision rule for research purposes that
Maslach Burnout Inventory4 contains subscales for each
distinguishes between people high and low in burnout. Such
dimension of burnout and is the commonly used inventory to
a decision rule would be helpful in estimating the percentage
assess burnout. To respect the multidimensional structure of
of burnt out individuals in a specific sample.

Multidimensionality versus unidimensionality

to various researchers, the multidimensional struc
Main According
ture of the concept of burnout does not allow for combining
the different into a unidimensional variable.5 A
By discussing different conceptualisations of burnout, this
results first to a unidimensional approach is that the asso
article for the presentation of objection
provides guidelines
obtained with theMaslach Burnout Inventory. ciations between the dimensions themselves and with other
it is recommended that the choice for a
of burnout not on one's
conceptualisation only depends
main research interest, but also on tne
of the Policy implications
design and the distributional properties of the sample.
To obtain a dichotomous burnout score, the "exhaustion + When research in the field of burnout, or when
1" criterion is recommended for research in non-clinical interventions to combat burnout, it is important to
reflect carefully on the most appropriate
populations. conceptualisation
this criterion, individuals can be considered as of burnout.
burnt out when to a norm group, The choice for a specific conceptualisation of burnout may
they report, compared
emotional exhaustion, in combination with be determined by one's main research interest, the
high high
or low complexity of the design, and the distributional properties
depersonalisation personal accomplishment.
This criterion is useful for computing burnout percentages in of the sample.
and for investigating differences between individu To compute the burnout percentage in a specific sample,
samples "
?is low and in burn out. the "exhaustion + 1 criterion is recommended

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions
Burnout as a unidimensional research variable ?17

variables are
complex.5 Combining the dimensions would With regard to the practical reasons, a unidimensional
result in a considerable loss of information. A second objection approach would have the advantage of simplifying results
is that dimensions have been constructed in such a way that considerably. Reporting findings for the variable "burnout",
are maximally from one another.4 A third rather than separately for the underlying dimensions, may
they independent
is that the role of the dimensions in the burnout give a better understanding of the outcomes, especially when
in the different of the proc complex effects are studied. For instance, one may examine
process may possibly vary phases
ess of burning out.8-10 On the other hand, it has been the succ?ssfulness of two particular types of training in
that burnout is strongly dominated mental health, health, and commitment,
by emotional exhaustion improving physical
and that the additional weight of the other two dimensions is whereby one is interested in differential effects for individuals
restricted It is a common that low and high in burnout and in moderating effects of gender.
(for example, Shirom11). finding
The results of this complex, fictitious experiment may be more
variables such as job demands are more strongly associated
easily understood when the burnout dimensions are com
with emotional exhaustion than with the other two burnout
bined. In this way, one needs less effort to integrate the find
ings of the separate dimensions. To illustrate, what should one
Although the multidimensionality of burnout is widely
conclude when a significant effect is found for emotional
acknowledged, there are theoretical and practical reasons to
exhaustion, but a marginally significant effect for depersonali
consider burnout as a single construct. From a theoretical
sation, and a non-significant effect for reduced personal
viewpoint, it is important to note that burnout has been pro
accomplishment? By combining the burnout dimensions,
posed as a specific work related syndrome that consists of the
overall effects of burnout become more visible. And of course,
three burnout dimensions. As with other multidimensional
one may decide to conduct additional analyses to provide
the multifaceted nature of the burnout syndrome
syndromes, detailed information about the underlying dimensions for
does not imply that we should abandon the overall concept of
those who are interested.
burnout. In contrast, conducting research and theorising on
Whether the advantages of a unidimensional approach out
the overall concept of burnout may sometimes help us to
weigh the disadvantages, may be determined by the research
advance our knowledge in a more way than research er's focal
thorough interest and, somewhat relatedly, by the complexity
on the
separate, underlying dimensions. It should be noted of the research 1 presents a decision tree for
design. Figure
that a unidimensional approach is not uncommon in research between a three dimensional, continuous, or
on related multidimensional concepts in psychology. For dichotomous of burnout. As shown in this
example, depression researchers often focus on differences decision tree, researchers who are primarily interested in the
among depressed versus non-depressed individuals, but burnout dimensions results for each
underlying may report
ignore effects of the subdimensions
depression, of
such as dimension This would to for
separately. apply, example,
depressed affect or the absence of positive affect.13 Moreover, researchers on interactions between the different
with respect to the CES-D scale, a scale designed to measure of burnout, such as in the study of phase models
depressive symptomatology in the general population, it is of burnout Golembiewski and colleagues8), or to
(for example,
warned against undue emphasis on the subdimensions those who are interested in the differential effects of the
because of the high internal consistency of the total scale.14 burnout dimensions on other variables.15 For instance,

Interest in specific burnout dimensions Interest in overall burnout concept

Preference for dichotomous or continuous

Complexity design? conceptualisation?

Continuous Dichotomous
Simple design Complex design

Statistical assumptions ANOVA violated?

variance or group size)

Yes No

results separately for each Treat burnout as a continuous Treat burnout as a dichotomous
burnout dimension variable variable

1 Decision tree for choosing between a three dimensional, continuous, or dichotomous of burnout.
Figure conceptualisation

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions
Brenninkmeijer, VanYperen

researchers may want to know which dimension is most pre colleagues,8 the three components of burnout can be
dictive of turnover. They may also want to examine whether combined into or eight Individuals
eight phases, gradations.
the antecedents of burnout, as personality
such characteris would be in the severest state of burnout when they score on
tics, work related attitudes, and work and organisational all three dimensions above a "universal norm". This universal
characteristics, are differentially related to the three burnout norm is the median in a norm of employees
group consisting
dimensions. Finally, researchers evaluating interventions to from a large federal agency. However, Schaufeli and
reduce burnout may wish to know precisely which burnout Enzmann12 view the criteria used Golembiewski et al as
dimensions have (and have not) improved as a result of the quite liberal and arbitrary, and plead for the development of
intervention (for example, Cooley and Yovanoff16). clinically validated, local criteria.
Researchers interested in the
overall concept of burnout, We conducted a study to provide a decision rule for
rather than in the underlying dimensions, may decide to com a dichotomous burnout score of
computing by comparing
bine the burnout dimensions into a single score. This may be group of "healthy individuals" with a group of individuals
useful, for example, for researchers who wish to estimate the who were as burnt out.19 The purpose of
clinically diagnosed
prevalence of burnout in a sample (for example, Schaufeli and this study was to yield a decision rule with large distinguish
Enzmann12). In addition, researchers may sometimes want to for individuals in a non-clinical context. This deci
ing power
examine differences between "healthy" individuals and sion rule would not be meant as a tool for individual diagnos
individuals in a state of burnout. In particular, this may tics, but as an instrument for research is, for
apply purposes?that
to researchers working with complicated research For conclusions about non-clinical Therefore, we investi
designs. samples.
reasons a complex
of parsimony, a design how the scores on the three dimensions of the Utrecht
design (for example, gated
with three
interacting factors and several dependent vari Burnout Scale (UBOS)20 could be validly combined into a
ables )may urge researchers to focus on the overall concept of qualification or "low" in burnout, we
"high" whereby prima
burnout and to pay less attention to the underlying rily aimed to keep the chance of a type 1 error (that is, a false
dimensions. positive: someone low in burnout is falsely qualified as high in
All in all, the specific research question, as well as the com burnout) around the conventional level of 5%. The qualifica
plexity of the research question, may be taken into considera tion "high in burnout" that individuals are in a state
tion when choosing between a multi- or unidimensional of burnout or are at a very high risk of getting into a state of
approach. A unidimensional approach implies a substantial burnout. Thus, these individuals would for a therapeu
loss of information, though it may increase the clarity of the tic treatment to combat or prevent burnout.
presented results. We not examined the predictive power of the UBOS
dimensions by means of logistic regression, we also looked for
Dichotomous versus continuous conceptualisation a practically useful categorisation based on the norm scores of
Once opted for a unidimensional approach, it is important to the UBOS,20 a categorisation that has the advantage of being
decide between a dichotomous or continuous of the current Because emotional
conceptualisa independent sample.
tion of burnout (see fig 1 ).Although this issue is applicable to exhaustion is considered to be
symptom the core
of burnout,
almost any psychological variable, including the separate we examined only in which emotional
categorisations high
burnout dimensions, and although it is widely covered in the exhaustion was a precondition for the qualification burnout.
methodological literature, we believe that it may be useful to The decisive of the two remaining was var
weight symptoms
discuss it briefly with respect to the measurement of burnout. ied.
For computing burnout percentages, it is obvious that burnout
should be treated as a dichotomous variable. However, for METHODS
studying effects of burnout, or differences between individuals Procedure and participants
low and high in burnout, the answer is more complicated. The data of
the study from a study by Wagenvoort
Clearly, the researcher's ideas about the distribution of and in which 73 individuals Analo
colleagues,21 participated.
burnout, in general or in a specific sample, come into play: is to the study et al, we both a
gous by Wagenvoort employed
burnout found in different or is it rather a matter clinical and a control
gradations, group group. The clinical group consisted
of "all or nothing"? issues may
Moreover, methodological of 29 participants (25 men and four women) who were suffer
influence the choice for a dichotomous or a continuous were
ing from burnout and who selected via an institute for
variable. Opting for burnout as a dichotomous work related and psychotherapy. The mean in
independent problems age
variable permits the use of t tests, ANOVAs, or MANOVAs (that this group was 45 years (SD 6.46). clients who were
is, univariate or multivariate of treated for burnout for the criteria of burnout),
analyses variance). (see below the
Nevertheless, it is sometimes difficult to meet the criteria for controlled with a checklist to what extent the clini
these statistical analyses. An important precondition is homo cal picture to the criteria of the
corresponded diagnosis
geneity of variance; in particular, the largest variance should neurasthenia. clients who met the criteria
Only diagnostic
not be associated with the smallest group.17 Moreover, small were their to fill in a questionnaire and
requested by therapist
group sizes generally reduce the power of these tests. In burn to return the filled in questionnaire.
out research, it is often the case that the group with burnout The control group consisted of 44 individuals (26men and
is small and that the variance in this group is large. In this 18 women) who were not treated for burnout and who were
situation, itmay be wiser to conduct with in various The average
regression analyses working professions. age in this group
burnout as a continuous variable. was 43 years were
Although regression analy (SD 9.89). Individuals asked if they wanted
sis also assumes constant variance, the use of continuous to participate in a study about the relation between personal
variables eliminates the of small size. ity characteristics and the way
problem group people cope with problems. The
Nonetheless, it should be noted that heterogeneity of variance could be filled in at home and could subse
can itself be the subject of investigation.18 For example, itmay be returned. The in this
quently response percentage group
be informative to know that individuals suffering from burn was 95%.
out vary largely in their emotional reaction to certain threats. In the study and the clinical
by Wagenvoort colleagues,21
group appeared to diverge from the non-clinical group in sev
How to combine the burnout dimensions into a eral respects. For instance, the clinical group scored higher on
dichotomous score exhaustion and distance, and lower on In
There are several to combine the burnout dimensions
ways addition, individuals in the clinical group were lower in self
into a unidimensional score. According to Golembiewski and esteem, tended to somatise more, and were more neurotic.

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions
Burnout as a unidimensional research variable i 19

Moreover, individuals in this group were shy and not very

Table 1 Definition of burnout and resulting burnout
extraverted in comparison to the non-clinical group.
We used to examine how well the UBOS percentage
logistic regression
dimensions predicted burnout. In addition, we investigated Burnout %
three possible categorisations that were based on the UBOS
Definition of burnout Control group Clinical group
norm scores. In these categorisations, individuals were classi
fied as high in burnout when they were characterised by: EE^2.67* 9.186.2
EE high 15.991.3
High exhaustion
EE high, and D high or C low 6.8 69.0
exhaustion, distance or low EE high, D high, and C low 0.041.4
High accompanied by high
*EE, emotional exhaustion; D, distance, C, competence.
High exhaustion, accompanied by high distance and low

means in the 75th or exhaustion; following this categorisation, persons with a score
"High" "scoring percentile higher",
whereas "low" refers to "scoring in the 25th or similar or higher than 2.67 were categorised as burnt out.
lower".20 In addition to the logistic regression analyses, the accuracy
of three categorisations based on the UBOS norm scores were
Measures examined.20 The categorisation in which high exhaustion, and
Burnout diagnosis high distance or low competence were conditions for burnout
The diagnosis burnout was made when the patient met the ("exhaustion +1") seemed to be the most effective categori
criteria for the diagnosis of neurasthenia following the ICD-10 sation to distinguish between individuals high and low in
classification (International Classification of Diseases of the burnout. This resulted in a fairly small chance (6.8%) of an
World Health Organisation22), and when the
picture clinical inaccurate qualification of burnout (that is, a false positive),
resulted from a long lasting process of relative overload. The although the burnout percentage in the clinical group (69.0%)
diagnostic criteria of neurasthenia following the ICD-10 are as was somewhat conservative (see table 1 ).When high exhaus
follows: persistent and distressing complaints of either tion, distance, and low competence were all three condi
increased fatigue after mental effort, or bodily weakness and tions for burnout, this reduced the chance of a false positive to
exhaustion after minimal effort. Furthermore, the patient zero, but the burnout percentage in the clinical group was very
experiences at least two of
the following complaints: muscular low (41.4%). A categorisation in which only high exhaustion
aches and pains, dizziness, tension headaches, sleeping was a condition for the qualification burnout yielded a burn
disturbance, inability to relax, irritability, and dyspepsia out percentage of 15.9% in the control group and 93.1% in the

(stomach and intestinal complaints). Moreover, the clinical clinical group.

picture may not correspond to a more specific ICD-10 disorder

(for example, an anxiety or depressive disorder). When the DISCUSSION

were work related and when professional qualities When conducting research on burnout, it may be difficult to
had declined, this was interpreted as additional support for decide whether one should report results separately for each
the diagnosis of burnout.2? burnout dimension or whether one should combine the
dimensions. In research on burnout, results are often
Burnout questionnaire presented for each dimension separately. This approach
Burnout was assessed with the Utrecht. Burnout Scale the multidimensionality of the concept of burnout
(UBOS), which consists of 16 items.20 This scale has been and is the appropriate for researchers interested in
developed to also measure burnout outside the human the underlying dimensions of burnout. However, when one is
services and consists of the following three subscales: exhaus
primarily interested in differences between individuals high
tion (U scale), distance (D scale), and competence (C scale). and low in burnout, it seems more convenient to combine the
The three factor structure fits reasonably well to empirical dimensions. This the understandability and clarifies
findings. The internal consistency is good and the stability is the results, when research such
especially complex questions,
reasonably good. Items could be responded to on a seven point as three way interactions, are studied. Moreover, from a theo
scale varying from 0 ("never") to 6 ("always, daily"). retical viewpoint, it is important to note that researchers have

proposed burnout as a specific syndrome, thereby underlining

RESULTS the importance of an overall conceptualisation of burnout. In
To determine the predictive value of the UBOS dimensions, we 1, a decision tree is depicted for choosing between a three
regression analyses with
(coded burnout as dimensional, continuous, or dichotomous conceptualisation of
performed logistic
0 for "burnout" and 1 for "no burnout") as dependent burnout.
variable. The model, consisting of the three UBOS dimensions, We conducted an empirical study to develop a decision rule
had more predictive power for burnout compared to the con for research purposes that distinguishes between people high
stant (model %2(3) = 65.26, p < 0.001 ) and yielded a burnout and low in burnout. This criterion would be useful for

percentage of 9.1% in the non-clinical group and 86.2% in the computing burnout percentages in samples and for investigat
clinical group. Hence, the chance of a type 1 error was 9.1%. ing differences between individuals low and high in burnout.
However, only exhaustion appeared, to be a significant predic We found that a categorisation in which both high exhaustion
tor (B = -2.10, =
SD 0.54, p < 0.001); distance (B 0.07, SD and high distance or low competence were conditions for
and competence = SD 0.55, NS) were not burnout + in a fairly small chance
0.51, NS) (B 0.51, ("exhaustion 1"), resulted

significant predictors, which may be attributed to the high (6.8%) of an inaccurate qualification of burnout (that is, a
= of an accurate of
intercorrelations between the three dimensions (rtcd 0.67, false positive). The chance qualification
= =
rcc-c -0.46, r.(l -0.34, p < 0.01). To determine the optimal burnout with this decision rule was 69.0% (see table 1). The
critical score for exhaustion, we performed a logistic of this categorisation is that it is based on existing
regression with only exhaustion as
a predictor. This analysis norm scores that are independent of the current sample.
resulted in the same burnout percentages as the first analysis However, regression revealed that an alternative
(see table 1 ).This analysis yielded also the following formula: categorisation, in which we applied a critical value of 2.67 on
= - x
burnout 6.16 2.12 exhaustion. Filling in the critical emotional exhaustion, was on average most accurate. Al
value of burnout (0.5), resulted in a critical score of 2.67 for this decision rule resulted in a larger type 1 error

www.occenvmed .com

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions
?20 Brenninkmeijer, VanYperen

the chance of an accurate diagnosis of burnout was 2 Maslach C, Schaufeli WB, LeiterMP. Job burnout. Annu Rev Psychol
considerably larger (86.2%). The drawback of this decision 3 Schaufeli WB, Maslach C, Marek T, eds. Professional burnout: recent
rule is that it is based only on a small sample. in theory and research. London: Taylor and Francis, 1993.
Recently, the "exhaustion +1" rule is referred to in the 4 Maslach C, Jackson SE. TheMaslach Burnout Inventory. Manual edition.
manual of the Utrecht Burnout Scale.24 In addition, in a study Palo Alto, CA: Consulting Psychologists Press, 1986.
5 Maslach C. Burnout: a multidimensional perspective. In:Schaufeli WB,
by Brenninkmeijer and colleagues,25 the "exhaustion +1" cri
Maslach C, Marek T, eds. Professional burnout: recent developments in
terion has been successfully applied to examine differences DC: Taylor and Francis, 1993:19-32.
theory and research. Washington,
between individuals low and high in burnout. The purpose of 6 Schaufeli WB, Van Dierendonck D. Burnout, een concept gemeten: De
Nederlandse versie van de Maslach Burnout Inventory (MBI-NL) [Burnout,
that study was to examine whether teachers high and low in
the measurement of a concept: the Dutch version of theMaslach Burnout
burnout differ in the perception of being superior to others.
Inventory (MBI-NL)].GeaVag en Gezondheid 1994;22:153-72.
Participants were asked to generate information about 7 Koeske GF, Koeske RD. Construct validity of theMaslach Burnout
inferior and superior others. Perceived superiority was Inventory: a critical review and reconceptualization. J Appl Behav Sei
assessed by response latencies and the quality of the 1989;25:131-44.
As those in
8 Golembiewski RT, Boudreau RA, Munzenrider RF, et al. Global
information generated. expected, among high burnout: a world-wide pandemic explored by the phase model.
burnout (following the "exhaustion +1" criterion), positive in organizational behavior and industrial relations, Vol. 21.
superiority (that is, feeling better than others) was reduced, Greenwich, CT: JAI Press, 1996.
whereas the perception of negative 9 Leiter MP. Burnout as a developmental process: consideration of models.
superiority (that is, feeling
In:Schaufeli WB, Maslach C, Marek T, eds. Professional burnout: recent
less bad than others) was intact. Thus, the "exhaustion + 1"
developments in theory and research. Washington, DC: Taylor and
rule seems to be an effective for mapping
categorisation Francis, 1993:237-50.
differences between individuals high and low in burnout. 10 Van Dierendonck D, Schaufeli WB, Buunk BP. Toward a process model
of burnout: results from a secondary analysis. European Journal of Work
It should be noted that the results in this study were
and Organizational Psychology 2001 ;10:41-52.
obtained in a relatively small sample and with a preliminary
11 Shirom A. Burnout inwork organizations. In:Cooper CL, Robertson I,
version of the UBOS. In 2000 a final version of the UBOS was eds. International review of industrial and organizational psychology.
published, with slightly different items and revised norm New York:Wiley & Sons, 1989:25-48.
Future in a larger 12 Schaufeli WB, Enzmann D. The burnout companion to study and
scores.24 research, preferably sample, would
be to examine the of the developed practice: a critical analysis. London: Taylor & Francis, 1998.
necessary accuracy 13 Ahrens AH, Alloy LB. Social comparison processes in depression. In:
criterion for researchers using the revised version of the Buunk BP, Gibbons FX, eds. Health, coping, and well-being: perspectives
UBOS. Nevertheless, the current criterion is formulated in from social comparison theory. Hillsdale, NJ: Erlbaum, 1997:389-410.
terms of scores or higher) on 14 Rad off
I LS. The CES-D scale: a self-report depression scale for research
"high" (75th percentile in the general population. Applied Psychological Measurement
emotional exhaustion and distance and "low" scores (25th 1977;1 :385-401.
percentile or lower) on competence, and may therefore be 15 Lee RT, Ashforth BE. A meta-analytic examination of the correlates of the
useful for other burnout inventories as well, including the new three burnout dimensions of job burnout. J Appl Psychol
version of the UBOS. 1996;81:123-33.
16 Cooley E, Yovanoff P. Supporting professionals at risk: evaluating
When the group with burnout is too small or too
interventions to reduce burnout and improve retention of special
heterogeneous for reliable analyses, which may often be the educators. Exceptional Children 1996;62:336-55.
case, it seems better to employ a continuous burnout scale. 17 Stevens J. Applied multivariate statistics for the social sciences, 2nd edn.
Little is known, however, about the way in which the dimen Hillsdale, NJ: Lawrence Erlbaum Associates, 1992.
18 Tacq J. Multivariate analysis techniques in social science research.
sions can be most accurately combined into a continuous
London: Sage Publications, 1997.
score. Our study19 would that one might use
empirical suggest 19 Brenninkmeijer V, VanYperen NW. Diagnostiek van burnout door
emotional exhaustion as a sole, continuous because middel van de Utrechtse Burnout Schaal [Diagnosing burnout by means
this was the only dimension that reached in the of the Utrecht Burnout Scale]. Nederlands Tijdschrift voor de Psychologie
logistic regression analysis. We feel, however, that abandoning
20 Schaufeli WB. Utrechtse Burnout Schaal {UBOS): Voorlopige
the other two burnout dimensions be a somewhat drastic
may handleiding [Utrecht Burnout Scale (UBOS): Preliminary manual]. Internal
conclusion from our study, especially as our sample was not report, section Psychology ofWork, Health and Organization, University
a few studies of Utrecht, Netherlands, 1995.
very large. Moreover, have presented their results
21 Wagenvoort MA, VanYperen NW, Hoogduin CAL, et al.
with a combined, continuous measure of burnout with high Persoonskenmerken en burnout: De medi?rende rol van copingsti?l
internal In a study by Buunk and all
consistency. colleagues,26 [Personality characteristics and burnout: The mediator role of coping
items of the Maslach Burnout Inventory were summed up, style]. Nederlands Tijdschrift voor de Psychologie 1998;53:128-43.
22 WHO. The ICD-10 classification of mental and behavioural diseases:
which resulted in a Cronbach's alpha of 0.87. In two other
clinical description and diagnostic guidelines. Geneva: WHO, 1992.
studies, this procedure resulted consistent in a highly
burnout 23 Hoogduin CAL, Schaap CPDR, Methorst GJ. Burnout: Klinisch beeld en
scale > future research In:CAL
(Cronbach's alphas 0.80).27 Hopefully, diagnostiek. [Burnout: clinical picture and diagnostics].
will teach us about the preferable procedure for creating a sin Hoogduin, CPDR Schaap, Kladler AJ, Hoogduin WA, eds.
continuous burnout variable. Behandelingsstrategie?n bij burnout [Treatment strategies of burnout].
gle, Netherlands: Bohn Stafleu Van Logum, 1996:31-41.
24 Schaufeli WB, Van Dierendonck D. UBOS: Utrechtse Burnout Schaal.
Handleiding [UBOS: Utrecht Burnout Scale. Manual]. Lisse, Netherlands:
Swets and Zeitlinger, 2000.
Authors' affiliations
V Brenninkmeijer, N VanYperen, of Groningen,
25 Brenninkmeijer V, VanYperen NW, Buunk BP. Iam not a better
Netherlands teacher, but others are doing worse: burnout and perceptions of
superiority among teachers. Social Psychology of Education
This study was conducted as part of the Netherlands concerted research 2001;4:259-74.
26 Buunk BP, Ybema JF,Gibbons FX, et al. The affective consequences of
action on "Fatigue atWork" granted by the Netherlands Organisation of
social comparison as related to professional burnout and social
Scientific Research (NWO: grant no. 580-02.230)
comparison orientation. European Journal of Social Psychology
REFERENCES 27 Brenninkmeijer V, VanYperen NW, Buunk BP. Burnout, social
1 Maslach C, Jackson SE. The measurement of experienced burnout. J comparison orientation and the affective consequences of social
Occup Behav 1981;2:99-113. comparison information. Manuscript submitted for publication, 2000.

www.occenvmed .com

This content downloaded from on Thu, 24 Dec 2015 20:26:13 UTC
All use subject to JSTOR Terms and Conditions