Teaching and Teacher Education 43 (2014) 99e109

Contents lists available at ScienceDirect

Teaching and Teacher Education

journal homepage: www.elsevier.com/locate/tate

Burnout among senior teachers: Investigating the role of workload

and interpersonal relationships at work
Filip Van Droogenbroeck a, *, Bram Spruyt a, Christophe Vanroelen b
a
b

TOR, Sociology Department, Vrije Universiteit Brussel, Belgium

Interface Demography, Sociology Department, Vrije Universiteit Brussel, Belgium

h i g h l i g h t s

Interpersonal relationships are differentially related to the burnout dimensions.

Relations with students are most strongly related to the three burnout dimensions.

Supervisors play a crucial role in giving teachers more autonomy.

More autonomy leads to less dissatisfaction with non-teaching-related workload.

Our data support the sequential process model of burnout.

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 9 October 2013
Received in revised form
7 July 2014
Accepted 10 July 2014
Available online 30 July 2014

According to the intensication thesis, external pressures from policymakers, supervisors, parents, and
experts fuel an ever-expanding teaching role and associated workload. Against that background, we
examined how four interpersonal relationships (students, colleagues, supervisors, and parents),
teaching-related and non-teaching-related workload (e.g., paperwork), and autonomy are related to
teacher burnout (emotional exhaustion, depersonalization, personal accomplishment). The participants
were 1878 Flemish teachers 45 years of age or older. The data were analyzed by means of structural
equation modelling. Interpersonal relationships were differentially related to burnout. Teaching-related
and non-teaching-related workload were both related to emotional exhaustion. Autonomy was most
strongly related to non-teaching-related workload.
2014 Elsevier Ltd. All rights reserved.

Keywords:
Teacher burnout
Teacher autonomy
Workload
Social support
Intensication

1. Introduction
Burnout is a persistent, negative, work-related state of psychological exhaustion that results from a mist between personal intentions and motivations on the one hand and actual on-the-job
experiences on the other (Schaufeli & Enzmann, 1998). It includes a
wide range of psychological (e.g., chronic fatigue, low self-esteem,
lack of self-condence, depression) and physiological (e.g., headaches, muscle pains, hypertension) symptoms. Burnout impedes
the attainment of professional goals (Maslach, 2003), depletes

* Corresponding author. Vrije Universiteit Brussel, Sociology Department, Pleinlaan 2, 1050 Brussel, Belgium. Tel.: 32 474481384; fax: 32 26148140.
E-mail addresses: Filip.Van.Droogenbroeck@vub.ac.be (F. Van Droogenbroeck),
Bram.Spruyt@vub.ac.be (B. Spruyt), Christophe.Vanroelen@vub.ac.be (C. Vanroelen).
http://dx.doi.org/10.1016/j.tate.2014.07.005
0742-051X/ 2014 Elsevier Ltd. All rights reserved.

coping resources (Park, Jacob, Wagner, & Baiden, 2013), and

consequently becomes self-perpetuating and hard to rehabilitate.
The prevalence of burnout has been investigated in a wide range
of occupational categories (Alarcon, 2011; Schaufeli & Enzmann,
1998). In particular, contact professions (e.g., nurses, physicians,
social workers, and teachers) turn out to be quite susceptible to
burnout (Maslach, 2003). Within this occupational category
teachers have repeatedly been shown to report the highest levels of
burnout (Heus & Diekstra, 1999; Schaufeli & Enzmann, 1998). Not
surprisingly, an extensive body of literature studies burnout among
teachers (Burke & Greenglass, 1995; Byrne, 1991; Dorman, 2003;
Grayson, Alvarez, & Grayson, 2008; Hakanen, Bakker, & Schaufeli,
2005; Kinman, Wray, & Strange, 2011; Kokkinos, 2007; Skaalvik
& Skaalvik, 2009, 2010, 2011; Van Der Doef & Maes, 2002). This
paper engages with this literature by studying the impact of (1) the
quality of the interpersonal relationships teachers maintain with

100

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

pupils, colleagues, parents and principals, (2) teaching and nonteaching related workload, and (3) perceived autonomy on
burnout among senior teachers (aged 45e65) in Flanders (the
Dutch-speaking part of Belgium, N: 1878).
The focus on senior teachers is warranted, because the consequences of suffering from burnout for older employees in terms of
their (re)integration in the labour market prove to be dramatic.
Teachers suffering from burnout have a higher chance of leaving
the profession and retiring early (Leung & Lee, 2006; Martin, Sass,
& Schmitt, 2012; Rudow, 1999; Skaalvik & Skaalvik, 2011; Van
Droogenbroeck & Spruyt, 2014). This is especially problematic,
because in several countries, including Belgium, the teacher work
force is ageing (OECD, 2005, 2012). According to the OECD (2012)
this tendency will put increasing pressure on the teacher labour
market and may lead to increasing teacher shortages. Since policymakers attempt to motivate employees e and more specically
teachers e to work longer, it becomes crucial to understand the
determinants of burnout among senior teachers.
The remainder of this article is organized as follows. First, we
situate our research question against the background of changes
the teaching profession went through during the last two decades.
This section justies our focus on the impact of the quality of
different interpersonal relationships, types of workload, and autonomy. Next, we review the existing literature on burnout among
teachers and locate our own study against that background. This
section is followed by a description of our data and the methods we
used. Subsequently, we present the results of our study. In the
concluding section we elaborate on the implications of our ndings.

1.1. Intensication of the teaching profession: interpersonal

relationships and workload
Our investigation of burnout is inspired by the multitude of
changes the teaching profession has undergone during the past
decades, often referred to as intensication and deprofessionalization (Apple, 1988; Ballet, Kelchtermans, & Loughran, 2006;
Woods, 1999). The intensication thesis holds that teachers are
increasingly subjected to external pressures from policymakers,
supervisors, parents, and experts. This intensication results in an
ever-expanding teaching role, a signicant increase in nonteaching-related (largely administrative) workload, and less time
for social contact with colleagues and in private life.
The increasing demands and expectations towards education in
general and teachers in particular by policymakers and society at
large, results from the combination of on the one hand the way
education (has to) function(s) and the broadening of education's
objectives on the other hand.
First of all, already in the 1980s Apple (1988), pointed to the
increasing adoption of a more economic and management-oriented
perspective on education among policymakers. Educational policy
increasingly follows a market logic in which education is thought of
as a product produced by schools. Thus, efciency and effectiveness
become the central evaluative parameters together with maximal
freedom of choice for parents and pupils as consumers of education
(Ball, 2003; Ballet & Kelchtermans, 2009). In that view, quality
becomes a synonym for the effective and efcient achievement of
standards that are objectively measurable. This implies that highlevel tasks become routinized, the sphere of work narrows (e.g.,
pre-specied competencies), and high-stakes accountability demands and standardization, such as prescribed and pre-specied
curriculums, textbooks, methods, and standardized tests, increase
(Valli & Buese, 2007; Woods, 1999). These externally imposed demands reduce the autonomy and creativity in the classroom and
appear to lead towards deskilling and deprofessionalization.

A second e and related e issue concerns the scope of education's tasks within larger society which continued to broaden.
Today, education is expected to contribute to the solution of
numerous social problems (e.g., drug abuse prevention, civic education, health education, etc.). This so called educationalization of
society implies that schools are expected to take over tasks that
were once carried out by parents (Smeyers & Depaepe, 2008). Some
of these are experienced by teachers as improper to their assumed
job description (Kelchtermans, 1999). Moreover, the broadening of
education's scope also increased the number of other professionals
(e.g., equal opportunities teachers, student counsellors, special
needs teachers, education inspectors) that work with pupils and
teachers. Teachers thus nd themselves in a contradictory situation
in which they face greater responsibility for their work but have
less control over the way they wish to work (Ball, 2003).
Although the literature often emphasizes the negative impact of
intensication, different authors stress that it may also have positive effects. Being confronted with intensication, for example,
some teachers look for further training or collaborative lesson
planning as coping strategies which may result in re-skilling and
re-professionalization (e.g., action research, teacher-as-researcher
movements) (Ballet & Kelchtermans, 2009; Woods, 1999). More
generally, intensication does not directly ow from educational
policy into the classroom. Intensication is a nuanced, mediated
and not completely deterministic process (Ballet et al., 2006). For
example, teachers lter imposed policy changes through their
personal ideologies and perspectives and have a certain freedom in
how they implement them in the classroom. Ballet et al. (2006)
distinguished several working conditions in schools that mediate
the impact of intensication. First, allowing teachers to co-decide
how work is organized can induce job satisfaction and fosters
feelings of collegiality and a sense of community. Participation in
decision-making is crucial to mitigate the impact of intensication
by enabling teachers to identify problematic external pressures and
tackling them collectively. At the same time, however, participating
in school policy-decisions can also increase the feeling of being
overburdened especially when a lot of time has to be spent on
making decisions without direct benet for the pupils. Secondly,
supervisors play a central role in translating policy arrangements
and demands from the local community or parents to the specic
school context. Supervisors, for example, can act as a buffer by
postponing the implementation of a new curriculum if they feel
that this would cause too much strain on the teachers. This implies
that supervisors also face intensication which makes their buffering role more difcult. They have to deal with more (administrative) tasks and have to focus on the more managerial aspects of
their job. Thirdly, positive interpersonal relationships at school can
also mediate the impact of intensication. Positive relations with
colleagues are very important in building shared values and norms
and create a collective goal orientation. In this regard, Hargreaves
(1994) noticed signicant changes in teacher culture. He points to
a shift from individualism to collaboration, from hierarchies to
teams, from supervision to mentoring, from in-service training to
professional development and from authority towards parents to a
contract with parents. Indeed, teachers have to interact with and
are made accountable to an increasing numbers of actors. Not only
do they have to maintain relationships with students, but also with
colleagues, supervisors, parents, and experts. On the one hand,
positive interpersonal relationships can function as a buffer against
the impact of intensication. On the other hand, Hargreaves identied contrived collegiality, which entails enforced collaboration
resulting from a drive towards managerialism and accountability,
rather than the interests of the students. This creates stress and
contributes to the alienation of individuals who are opposed to
such developments.

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

It is important to note that the issue we discuss above, does not

concern policy changes or accountability themselves. It is, after all,
reasonable that teachers are accountable for the work they perform
with students as part of their responsibility as a professional. In
addition, policy changes and innovation are necessary to keep up
with research on teaching, student performance, and inequality.
Research also shows that teachers are ready to accept changes,
which have a positive impact on the learning process and wellbeing of pupils (Smylie, 1994). What matters here concerns the
type of policy context teachers are working in and the way that
accountability is dened (for example high-stakes accountability
with a focus on summative assessments versus formative assessments which are more focused on feedback to modify the teaching
and learning activities).
The preceding arguments lead to two insights. First, the discussion about the ultimate or general consequences of the intensication of the teaching profession has not been settled yet.
Secondly, all processes associated with intensication point towards an increased importance of (a) the quality of (an increased
multitude of) interpersonal relationships, (b) different types of
workload and (c) experienced autonomy. In this paper we do not
investigate the intensication process itself, but focus on some of
the main consequences of intensication on the prevalence of
burnout among senior teachers.
The antecedents of burnout have often been investigated by
discriminating between job demands and job resources as latent
variables comprising a wide array of job conditions (e.g., Alarcon,
2011; Elovainio et al., 2005; Schaufeli & Bakker, 2004). The renements of the original intensication thesis, however, argue
against that practice. Indeed, as the experience of intensication is
mediated, it follows that it is important to distinguish different
aspects of the working conditions and interpersonal relationships
with relevant others for there is no reason to assume that their
impact is identical. In their everyday work, teachers interact with
students, parents, colleagues, and supervisors. It is the quality of
these relationships that determines if the relationship creates
stress e job demand e or instead is experienced as positive and
serves as a buffer e job resource e against stress. In addition, the
experience of intensication is mediated by the participation in
school policy-decisions and how administrative duties are dealt
with. Instead of grouping variables in latent factors, such as job
demands or resources, we investigate several aspects of the
working conditions separately. Regrouping variables into latent
constructs that only reect a distinction between demands and
resources, may lead to a loss of information, certainly since previous research has indicated that burnout is differentially related to
different working conditions (e.g., Lee & Ashforth, 1996; Skaalvik &
Skaalvik, 2009).
1.2. Teacher burnout
Burnout is a multidimensional construct that is often considered
the end stage in a chain of reactions to chronic occupational stress.
Maslach and Jackson (1986, p. 1) dene burnout as a syndrome of
emotional exhaustion, depersonalization, and reduced personal
accomplishment that can occur among individuals who do people
work of some kind. Emotional exhaustion refers to feelings of fatigue that develop as one's emotional energies become drained.
Cynical depersonalization points to the development of indifferent
and negative attitudes towards others; it refers to the impersonal
and dehumanized perception of (service) recipients. Finally,
reduced personal accomplishment is the negative evaluation of
personal accomplishments in working with other people and
dissatisfaction with one's achievements, often accompanied by
feelings of insufciency and poor professional self-esteem. Initially,

101

burnout was restricted to those individuals working with recipients

(e.g., student, clients, patients), but the concept was later broadened to include all occupations (Maslach, Schaufeli, & Leiter, 2001).
Burnout is distinguished from other syndromes, such as depression
and chronic fatigue syndrome (CFS), because burnout is considered
to be mainly job-related, whereas depression and CFS are more
context-free (Schaufeli & Enzmann, 1998). Unlike burnout, CFS is
pervasive and includes mainly physical symptoms, next to psychological symptoms. By contrast, burnout symptoms are primarily
psychological, although accompanying physical symptoms are not
uncommon. Although burnout and depression share similar workrelated antecedents, research has concluded that they are conceptually and empirically distinct (Schaufeli & Enzmann, 1998; Toker &
Biron, 2012).
Studies have conrmed the multidimensional nature of burnout
and indicated that the three dimensions cannot be added up to a
single measure (Byrne, 1994; Langballe, 2006; Schaufeli, Daamen, &
Van Mierlo, 1994). The multidimensional nature, however, does not
imply that there are no relationships between the burnout dimensions. Several models have been proposed to explain the associations among the three dimensions. One of the best known
models assumes that the three dimensions follow a sequential
process starting with emotional exhaustion resulting from the
emotional demands of dealing with recipients (Leiter & Maslach,
1988). In an inappropriate attempt to cope with high levels of
exhaustion, workers then tend to withdraw psychologically from
others, which leads to dehumanized and depersonalized contact.
This dysfunctional coping strategy can further deteriorate the
relationship with recipients. Finally, this process leads to the
development of feelings of diminished personal accomplishment,
as more failures are experienced and workers become more
dissatised with their achievements.
Empirical evidence about the causal relations between the
burnout dimensions is scarce. One exception is a study from Taris,
Le Blanc, Schaufeli, and Schreurs (2005), which reviewed the evidence of several models and tested these models in a new longitudinal study. Their study supported the Leiter and Maslach (1988)
model but also found that in their teacher sample higher levels of
depersonalization were associated with higher levels of exhaustion
across time. Although in our cross-sectional study it is impossible to
investigate the causal nature of the burnout dimensions, we expect
to nd the same sequential process as proposed by Leiter and
Maslach (1988).
1.2.1. Interpersonal relationships
Interpersonal relationships play a central role in the work of
teachers. Not surprisingly, then, previous research has revealed that
the support teachers receive from their supervisor and the positive
relationships they maintain with colleagues, parents, and students
are negatively related with emotional exhaustion and cynical
depersonalization and are positively related with personal
~ oz, & Carrascoaccomplishment (e.g., Cano-Garca, Padilla-Mun
Ortiz, 2005; Dorman, 2003; Friedman, 1995, 2003; Gavish &
Friedman, 2010; Grayson et al., 2008; Greenglass, Burke, & Konarski, 1997; Leung & Lee, 2006; Russell, Altmaier, & Vanvelzen, 1987;
Skaalvik & Skaalvik, 2009, 2011). Our data allow us to investigate
four interpersonal relationships: students, colleagues, supervisors,
and parents. Based on previous research we expect direct paths
from all the interpersonal relationships to the three burnout dimensions (see Fig. 1). However, there is little research that effectively takes into account all these components separately and at the
same time. Often only one or two burnout dimensions or only one
or two interpersonal relationships are investigated (e.g., CanoGarca et al., 2005; Kinman et al., 2011; Skaalvik & Skaalvik,
2011). In addition, the relations between the burnout dimensions

102

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

Fig. 1. Theoretical structural model. The variables RELCOL, RELSTU, and RELPAR were grouped together to increase the clarity of the image. Note: SPC support with policy changes;
RELSUP relationship with supervisor; RELCOL relationship with colleagues; RELSTU relationship with students; RELPAR relationship with parents; AUT autonomy;
EE emotional exhaustion; DP depersonalization; PA personal accomplishment.

are often not investigated (e.g., Grayson et al., 2008; Kokkinos,

2007; Skaalvik & Skaalvik, 2009; Van Der Doef & Maes, 2002). In
this article we try to add to the existing literature by differentiating
between four interpersonal relationships and investigating how
these are separately related to the three burnout dimensions. We
also take into account the relations between the burnout dimensions. This is important because, rst, it can be expected that
not all interpersonal relationships at work have the same inuence.
Previous research revealed that teachers are quite likely to accept
changes when they believe they will have a positive effect on pupils
(Smylie, 1994). This pupil-centrism suggests that the quality of the
studenteteacher relationships will be of particular importance for
teachers' well-being. Second, an inverse relationship can be expected between the amount of direct paths from the interpersonal
relationships to the burnout dimensions and the strength of the
relationships between the burnout dimensions. If the relationships
between the burnout dimensions are strong, we expect that many
of the postulated direct paths will be mediated by one of the dimensions of burnout.
1.2.2. Teachers' autonomy and teaching-related and non-teachingrelated workload
Teaching is considered by many teachers to be a vocation
(Roness, 2011). When teachers are asked why they want to teach
they primarily refer to intrinsic and altruistic reasons, such as the
desire to teach, to work with children, or to make a contribution to
society (Brookhart & Freeman, 1992; OECD, 2005; Struyven,
Jacobs, & Dochy, 2012). Since the desire to teach and to help

children is so central in the motivation of teachers, it should not

surprise us that externally imposed demands for high-stakes
summative assessments, record keeping, and accountability are
experienced as a burden, as these tasks distract teachers from
their perceived core job of teaching students. Often teachers feel
that the time investment is disproportionate to the benets for the
student (Ballet et al., 2009). Several authors have indicated that
these administrative burdens are considered an important stressor
and a main cause of job dissatisfaction among teachers (Byrne,
1991; Dinham & Scott, 1998; Scott, Stone, & Dinham, 2001; Van
Droogenbroeck & Spruyt, 2014; Varlaam, Nutall, & Walker, 1992).
For this reason we argue that it is important to distinguish between teaching-related (e.g., class preparation, teaching) and nonteaching-related workload (e.g., administration, meetings). It is
plausible and even likely that teachers, trained and motivated to
teach, react differently to the workload that is directly related to
teaching and the well-being of their students than to the workload
many of them perceive as being generated by high-stakes
accountability demands. Previous research has indicated that
time pressure and workload are signicantly related to burnout
(Hakanen et al., 2005; Schaufeli & Bakker, 2004). When burnout
dimensions are analyzed separately, time pressure and workload
are positively correlated with emotional exhaustion but not with
the other dimensions (Dorman, 2003; Kokkinos, 2007; Mazur &
Lynch, 1989; Skaalvik & Skaalvik, 2009, 2010, 2011; Van Der
Doef & Maes, 2002). We expect teaching-related and nonteaching-related workload to be directly related to emotional
exhaustion. Following the intensication theory, we also

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

hypothesize that non-teaching-related workload is negatively

related to personal accomplishment.
Ballet et al. (2006) argued that intensication has multiple
sources and is experienced differently by teachers. Supervisors, for
example, play a central role in how policy changes are implemented. Through the support that supervisors provide and the
extent of teacher involvement in school policy-decisions they
allow, supervisors can inuence teachers' experienced job autonomy (Skaalvik & Skaalvik, 2009). Autonomy, in turn, is known to be
negatively related to emotional exhaustion and depersonalization
and to be positively related with personal accomplishment
(Skaalvik & Skaalvik, 2009, 2010). Following the literature, we
expect that supervisor support is positively related to feelings of
autonomy. More autonomy can lead to less teaching-related and
non-teaching-related workload, as teachers have more chances to
regulate the way they perform these tasks. We also expect autonomy to be negatively related to emotional exhaustion and cynical
depersonalization and to be positively related to personal
accomplishment.
1.2.3. Gender, age, and burnout
We also examined the impact of age and gender on burnout. Age
differences have been found for the three dimensions, although the
effects have been inconsistent. In previous studies younger teachers reported signicantly higher emotional exhaustion than older
teachers (Anderson & Iwanicki, 1984; Byrne, 1991; Russell et al.,
1987). Noor and Zainuddin (2011), on the other hand, in their
sample of female teachers found that older female teachers experienced more emotional exhaustion than younger teachers did. For
the other dimensions the ndings are less consistent. In some
studies younger teachers reported higher depersonalization scores
(Lau, Yuen, & Chan, 2005; Pierce & Molloy, 1990; Van Der Doef &
Maes, 2002), while in other studies reverse or non-signicant age
differences for cynical depersonalization were found (Anderson &
Iwanicki, 1984; Byrne, 1991; Leithwood, Menzies, Jantzi, &
Leithwood, 1999). In terms of personal accomplishment, several
studies (Byrne, 1991; Lau et al., 2005) show that younger teachers
experience lower personal accomplishment than older teachers.
Others reported reverse (Heus & Diekstra, 1999) or non-signicant
associations (Anderson & Iwanicki, 1984; Van Der Doef & Maes,
2002). The different age ranges of the studies considered and the
differential exposure of their samples to the healthy worker effect
may provide an explanation for the inconsistencies (Schaufeli &
Enzmann, 1998, p. 74). The healthy worker effect refers to a selection effect in which those teachers who burnout early in their careers are likely to quit their jobs, leaving behind those who exhibit
low levels of burnout. This healthy worker effect leads to an underestimation of the effects of burnout, because only working and
healthier employees are investigated. With regard to senior
teachers this becomes interesting because previous research indicates that teachers suffering from burnout have a higher chance
of leaving the profession and retiring early (Leung & Lee, 2006;
Martin et al., 2012; Rudow, 1999; Skaalvik & Skaalvik, 2011). We
hypothesize that a healthy worker effect will be visible at the end of
the career, since those teachers who feel burned out will leave the
profession sooner. Thus, we expect that older teachers, compared
with younger teachers, will report less emotional exhaustion,
cynical depersonalization, and more personal accomplishment.
Gender differences have been found in all three burnout dimensions. In most studies women report higher emotional
exhaustion than men do (e.g., Byrne, 1991; Grayson et al., 2008; Lau
et al., 2005); only a few studies reported a reverse association
(Anderson & Iwanicki, 1984). Byrne (1991) and Lau et al. (2005)
reported signicantly lower perceptions of personal accomplishment for female teachers when compared with male teachers,

103

whereas Unterbrink et al. (2007) found the reverse. Men generally

report higher cynical depersonalization than women do (Byrne,
1991; Lau et al., 2005; Russell et al., 1987; Stoeber & Rennert,
2008). Following the literature we expect female teachers would
experience less cynical depersonalization and personal accomplishment and higher levels of emotional exhaustion than male
teachers would.
2. Method
2.1. Data
This study focuses on senior teachers 45e65 years of age. The
threshold age of 45 aligns with previous studies on older employees
(e.g., Elovainio et al., 2005; Schreurs, Van Emmerik, De Cuyper,
Notelaers, & De Witte, 2010). The data are based on the End of
career survey conducted in 2011 among a random sample of the
population of active and non-active teachers between 45 and 65
years of age in Flanders (Belgium). We contacted 5200 working and
retired Flemish teachers using their private addresses. Because of
the sensitive information in the survey, a higher response rate could
be achieved by contacting teachers directly instead of using the
school to distribute the surveys. To optimize the response rate we
followed the Dillman procedure (Dillman, Smyth, & Christian, 2009)
and used a mixed methods design. The teachers were contacted in
four waves using mail surveys and reminder cards. Respondents
were invited to ll out the questionnaire on the Internet if they
preferred so (44% of those who participated in the survey did so).
With a response rate of 62% or 3124 lled-out questionnaires,
the survey was very well received (in comparison, the European
Social Survey 2010 had an average response rate of 61% using faceto-face interviews). Our sample only differed slightly from the
population of teachers. The data were weighted for gender, age, and
educational level (highest weight factor: 2.32). Of the 3124 respondents 1246 were retired and 1878 were working at the time of
the survey. Because burnout is primarily work related, only the data
from the 1878 working teachers were used in the empirical part of
the study.
2.2. Measures
2.2.1. Teacher burnout
Teacher burnout was assessed with the Dutch version of the
Maslach Burnout Inventory e The Dutch Educators Survey (MBI-NLES; Horn & Schaufeli, 1998) adapted for teachers. Emotional
exhaustion was measured with ve items (e.g., I feel emotionally
drained from my work. a .89), cynical depersonalization was
measured with ve items (e.g., I doubt the signicance of my job
a .82) and personal accomplishment with six items (e.g., I can
effectively solve the problems that rise in my work a .81). Responses were given on an eight-point Likert scale ranging from
0 Never to 7 Every day. Higher scores indicate more emotional
exhaustion, cynical depersonalization and personal accomplishment.
2.2.2. Interpersonal relationships
Four dimensions of interpersonal relationships at work were
analyzed. We measured the relationships with students (four items
e.g., Most students treat me with respect a .70), colleagues (ve
items e.g., When I have a hard time at school I seek support from
colleagues a .80), supervisor (ve items e.g., The principal
supports me on an adequate way during conicts with students and
parents a .90), and parents (ve items e.g., I have good contacts
with parents a .75). All items were rated on a ve-point Likert
scale ranging from 1 Strongly disagree to 5 Strongly agree.
Higher scores indicate more positive relationships.

104

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

2.2.3. Autonomy and support with policy changes

We also analyzed the autonomy of teachers and the support
teachers receive when policy changes occur. The perception of
autonomy was assessed by three items (e.g., I have enough say in
school policy a .78). The perceived support teachers receive
when policy changes occur was also assessed by three items (e.g., I
receive adequate support for the implementation of policy
changes a .70). All items were rated on a ve-point Likert scale
ranging from 1 Strongly disagree to 5 Strongly agree. Higher
scores indicate more autonomy and more perceived support with
the implementation of policy changes.

2.2.4. Teaching-related and non-teaching-related workload

As argued in the theoretical Section 1.2.2 we analyze teachingrelated and non-teaching-related workload separately instead of
using a general workload indicator. We asked teachers how satised they were with (1) workload because of teaching-related activities such as class preparation, teaching itself and (2) workload
because of non-teaching-related activities such as paperwork and
meetings. Both items were rated on a ve-point Likert scale ranging
from 1 Very satised to 5 Very dissatised.

2.3. Data analysis

Conrmatory factor analysis (CFA) and structural equation
n &
modelling (SEM) as implemented by MPLUS 6.2 (Muthe
n, 1998e2012) were used to analyze the data. We used
Muthe
SEM instead of other multivariate approaches, because SEM follows
a conrmatory approach, corrects for measurement error, and allows for a clear distinction between direct and indirect effects
(Byrne, 2011). The goodness-of-t of the models was evaluated by
t indices such as the c2 statistic, root mean square error of
approximation (RMSEA), comparative t index (CFI), TuckereLewis
t index (TLI), and standardized root mean square residual (SRMR).
For the RMSEA and SRMR indices, values less than .05 indicate a
good t, and values as high as .08 represent an acceptable t
(Browne & Cudeck, 1993). As a rule of thumb for the TLI and CFI
indices, values greater than .95 are indicative of good t, whereas
values greater than .90 are considered acceptable (Hu & Bentler,
1999). Our preliminary CFA analysis of the burnout dimensions
indicates the presence of non-normality in the variables. This does
not directly affect the estimates, but it could lead to biased standard
errors and an inated number of statistically signicant parameters. As suggested by Byrne (2011), we used the robust maximum
likelihood estimator MLM that estimates RMSEA, CFI, TLI, standard
errors, and mean-adjusted c2 test statistics that are robust to nonnormality.
To evaluate our research questions, two measurement models
were created using CFA to test the multidimensionality of the
theoretical constructs. The rst measurement model specied the
three Maslach burnout dimensions, whereas the second specied
autonomy, support with policy changes, and the interpersonal relationships with students, colleagues, supervisors, and parents. In
the next step we combined both measurement models and tested a
theoretical model by means of SEM where gender, age, and
teaching-related and non-teaching-related workload were added.
To discriminate between signicantly better tting models we used
the c2 difference test for the SatorraeBentler1 scaled c2.

1
The MLM method estimates the SatorraeBentler scaled c2 and a scaling
correction factor instead of the regular c2. See http://www.statmodel.com/chidiff.
shtml for more information on the adjusted c2 difference test when using the
MLM estimator.

3. Results
Table 1 shows the zero order correlations between the variables
and the Cronbach's alphas. The correlations between the three
burnout dimensions varied from moderate to strong. The strongest
correlation was found between emotional exhaustion and depersonalization (.74).2 The correlations between the four interpersonal
relationships were positive and varied from weak to moderate. The
strongest correlations were found between relationships with the
supervisor and the relationships with colleagues (.42) and the relationships with the parents and the relationships with the students (.48). Furthermore, autonomy was positively related to the
four interpersonal relationships and negatively related to
emotional exhaustion (.32) and depersonalization (.43). Autonomy was most strongly correlated with relationships with the
supervisor (.71). Teaching-related and non-teaching-related workload had a moderate negative correlation (.39).
Table 2 summarizes the results for the CFA and SEM analyses.
We rst tested a measurement model for burnout. Results of the
CFA revealed an acceptable t for the hypothesized threedimensional MBI structure. The incorporation of one crossloading and two error-covariances signicantly improved the t
(model 2). The second measurement model specied six aspects of
the school context, four interpersonal relationships (students, colleagues, supervisors, and parents), autonomy, and the support with
policy changes. The baseline model had a good t with the data. Fit
indices suggested a signicantly better t with one additional
error-covariance (see model 4). In the next step the two measurement models were combined, also revealing a good t of data
(model 5).
Based on previous empirical ndings, a theoretical model (Fig. 1)
was specied. Since previous research predicts direct paths between
all interpersonal relationships and the three burnout dimensions, we
grouped the interpersonal relationships with students, colleagues,
and parents in the gure for the sake of clarity. As argued in the
theoretical section (see Section 1.2.1) we expect several of these
direct paths to become non-signicant because we take into account
four interpersonal relationships at the same time and we also specify
the relationships between the three burnout dimensions. This
should cause a mediation of several of the direct paths from the
interpersonal relationships to the burnout dimensions.
For the SEM analysis we followed the method suggested by
Byrne (2011). The best tting model is determined rst, then the
most parsimonious model. Once the nal model is established, the
remaining non-signicant parameters are deleted.
The SEM analysis of the postulated model (Fig. 1) revealed a
satisfactory t to the data (model 6). The modication indices
suggested that a better t could be achieved (model 7) if one path
was added (relationships with students / teaching-related
workload). The c2 difference test for the SatorraeBentler scaled c2
(Dc2(1) 18.07, p < .001) and other t indices indicated that the
addition of this path leads to a statistically signicant improvement
of the model. In addition, the modication indices suggested adding one further path (relationships with colleagues / autonomy),
which equally resulted in a statistically signicant improvement of
the model (model 8). In the nal step non-signicant paths were
removed. The nal model revealed an acceptable t (model 9).
Fig. 2 presents the standardized regression estimates. Interestingly,
the relationship with parents was not signicantly related to any
indicator in the nal model. When this latent factor was removed,

2
Because of these strong correlations we controlled for multicollinearity between all the study variables. The highest Variance Ination Factor (VIF) was 2.4,
with a threshold of 10, multicollinearity can be excluded.

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

105

Table 1
Zero order correlations and Cronbach's alphas.

1 EE
2 DP
3 PA
4 RELSUP
5 RELCOL
6 RELPAR
7 RELSTU
8 AUT
9 SPC
10 Teaching-related workload
11 Non-teaching-related workload
12 Age
13 Gender

10

11

12

13

.74
e

.29
.38
e

.26
.31
.17
e

.24
.33
.18
.42
e

.26
.35
.24
.25
.25
e

.28
.36
.26
.23
.19
.48
e

.32
.43
.22
.71
.43
.30
.25
e

.31
.35
.20
.40
.30
.24
.18
.48
e

.40
.35
.19
.25
.17
.19
.23
.25
.24
e

.36
.31
.16
.33
.16
.21
.16
.39
.29
.39
e

.13
.05
.10
.01
.03
.02
.03
.00
.00
.13
.07
e

.06
.08
.11
.02
.11
.05
.03
.00
.05
.13
.06
.17
e

.89

.82

.81

.90

.80

.75

.78

.70

Cronbach's alpha

Note: all correlations are signicant (p < .001); EE emotional exhaustion; DP depersonalization; PA personal accomplishment. RELSUP relationship with supervisor;
RELCOL relationship with colleagues; RELPAR relationship with parents; RELSTU relationship with students; AUT autonomy; SPC support with policy changes.

the model t further improved to a good t (c2 (712) 1546.541,

RMSEA .033, CFI .952, TLI .947, SRMR .041).
As expected, the three burnout dimensions inuenced one
another. We found that higher levels of emotional exhaustion were
associated with higher levels of cynical depersonalization. The
relationship between emotional exhaustion and cynical depersonalization was very strong (b .71). Further, higher levels of
cynical depersonalization were associated with lower levels of
personal accomplishment. We expected that the strong relationships between the burnout dimensions would cause several direct
effects of interpersonal relationships to become non-signicant. To
verify this we tested the nal model with and without associations
between the burnout dimensions. In the model without relations
between the burnout dimensions there are eight direct effects from
interpersonal relationships to the burnout dimensions (not displayed). In the model with relations between the burnout dimensions the amount of direct effects was reduced to ve (see
Fig. 2). The latter model had a signicantly better t
(Dc2(1) 56.51, p < .001) than the model that did not specify relations between the burnout dimensions.

Also in accordance with our expectations, both teaching-related

and non-teaching-related workload had a direct positive but
modest effect on emotional exhaustion. In a similar vein, our hypothesis assuming that teaching-related and non-teaching-related
workload mediated the relationship between autonomy and
emotional exhaustion was supported. Autonomy was most strongly
related to non-teaching-related workload. Autonomy itself was
most strongly predicted by positive relationships with the supervisor and in lesser extent by positive relationships with colleagues
and feelings of being supported when policy changes occur.
For the interpersonal relationships, the central role of relationships with students in the experience of burnout is noteworthy.
Relations with students were related to all three burnout dimensions. Positive relationships with students were negatively
related to emotional exhaustion and cynical depersonalization but
positively related to personal accomplishment. Relationships with
students also had an indirect effect on emotional exhaustion
through teaching-related workload. Higher levels of positive relationships with students were related to lower levels of teachingrelated workload. In contrast, relationships with parents were not

Table 2
Summary of goodness-of-t for the conrmatory factor analyzes and structural models.
Actions
Measurement model
1. Burnout
- Model 1 baseline
- Model 2 adapted

1 Cross-loading 2 Error
covariances

2. School context
- Model 3 baseline
- Model 4 adapted

1 Error covariance

Model 5 combined

Structural model
- Model 6 baseline
- Model 7 adapted
- Model 8 adapted
Model 9 nal model

RELSTU on teaching-related
workload
RELCOL on AUT
Deletion non-signicant paths

MLM c2
(df, scaling correction)

RMSEA

Upper limit
RMSEA 90%

CFI

TLI

SRMR

483.98
(101, 1.292)
319.629
(98, 1.280)

.053

.057

.954

.945

.039

.041

.046

.973

.967

.031

.039

.042

.952

.945

.038

.037

.040

.958

.951

.037

.031

.033

.955

.950

.039

.034

.036

.940

.935

.045

.033

.035

.942

.936

.043

.033

.035

.942

.937

.042

.033

.035

.942

.937

.043

761.689
(260, 1.201)
702.166
(259, 1.200)
1508.918
(739, 1.151)

1965.025
(906, 1.124)
1934.125
(905, 1.124)
1920.608
(904, 1.124)
1934.937
(913, 1.123)

106

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

Fig. 2. Final structural model of four interpersonal relations, teaching-related and non-teaching-related workload, autonomy, support with policy changes, age, and gender with
teacher burnout. Only signicant standardized regression weights are included in the gure. Note: SPC support with policy changes; RELSUP relationship with supervisor;
RELCOL relationship with colleagues; RELSTU relationship with students; RELPAR relationship with parents; AUT autonomy; EE emotional exhaustion;
DP depersonalization; PA personal accomplishment.

related to any dimension of burnout. Positive relationships with

colleagues were negatively related to emotional exhaustion and
cynical depersonalization. Finally, the effect of positive relationships with the supervisor on exhaustion was fully mediated by
autonomy.
For the demographic variables, female teachers reported less
cynical depersonalization and less personal accomplishment
compared with male teachers. Older teachers reported less
emotional exhaustion and more personal accomplishment,
although the latter relation was weak (.08).
Several expectations were not conrmed. We did not nd a
signicant effect for non-teaching-related workload, autonomy,
and relationships with colleagues and supervisor on personal
accomplishment. Moreover, contrary to expectations, age and relationships with the supervisor were not related to cynical depersonalization, nor was gender related to emotional exhaustion.
4. Discussion and conclusion
In this paper we assessed the impact of four interpersonal relationships (students, colleagues, supervisors, and parents) and
teaching-related and non-teaching-related workload on burnout
among senior teachers. Burnout among senior teachers deserves
special attention, because research has indicated that burnout is
related to attrition and early retirement (Leung & Lee, 2006; Martin
et al., 2012; Rudow, 1999; Skaalvik & Skaalvik, 2011). Together with
the coming retirement of the baby boom generation (Ekerdt, 2009),

the early exit culture among teachers (Eurydice Network, 2012),

looming teacher shortages in several countries, and legislators
aiming to motivate employees to work longer (OECD, 2012),
studying burnout among senior teachers is highly relevant for society at large.
According to the intensication thesis the working conditions of
teachers changed signicantly during the last decades. Teachers
experienced a noticeable increase in high-stakes accountability
demands, administrative tasks and standardization which may
cause stress and feelings of deprofessionalization. This intensication of the profession is the result of teachers being increasingly
subjected to external pressures and demands from policymakers,
supervisors, parents, and experts. Maslach and Jackson (1984, p.
134) stated that the root cause of burnout comes from being
emotionally overextended and drained by one's contacts with other
people. In the case of teachers, the question may be asked, Contact
with whom? In accordance with previous research, our results
showed that the interpersonal relationships that teachers maintain
at work contribute signicantly to the prediction of burnout (e.g.,
Cano-Garca et al., 2005; Dorman, 2003; Friedman, 2003; Gavish &
Friedman, 2010; Grayson et al., 2008; Greenglass et al., 1997; Leung
& Lee, 2006; Skaalvik & Skaalvik, 2009, 2011). In line with the
recent renements of the intensication thesis we found, however,
that not all interpersonal relationships at work have the same
impact on burnout. The relationship with students, for example,
appears to play the most important role and relates directly to each
of the burnout dimensions. This is not surprising, since research

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

shows that teachers are mainly driven by intrinsic and altruistic

motivations, such as the desire to teach and help children
(Brookhart & Freeman, 1992; OECD, 2005; Struyven et al., 2012). It
is often taken for granted, but teachers nd themselves in the
peculiar situation where most of their time is spent with nonadults. When this time spent with students is experienced in a
positive and pleasant way this feeling may directly spill-over to
their job satisfaction and increase their perception of personal
accomplishment. In such a situation these teachers are accomplishing what they are mainly driven to do. In contrast, the relationship with the parents was not signicantly related to any of the
burnout dimensions. Teachers encounter parents only a couple of
times every year. This might explain why this relationship has no
impact on burnout. The strongest inuence appears to come from
the relationships that are encountered on a daily basis. Relationships with colleagues were directly related to emotional exhaustion
and cynical depersonalization. Although teachers spend most of
their time alone with children, the support that teachers receive
from colleagues can function as a buffer against burnout. In this
regard, Hargreaves (1994) points to the need to develop collaborative cultures where teachers can engage in collaborative inquiry
to collect and analyze local data so that data based decisions can be
made to improve teaching and learning (e.g., nding joint solutions
to shared problems, exchanging practices, co-teaching). Collaborating and talking about problems at work may induce feelings of
solidarity and empathy and prevent indifferent and negative attitudes towards others.
Finally, also the relationship with the supervisor plays an
important role. The effect of this relationship on emotional
exhaustion was fully mediated by autonomy. Supervisors play a
crucial role in giving teachers more autonomy. They have the power
to involve teachers in school policy-decisions and enable teachers
to take initiative. The importance of autonomy is illustrated by the
direct effect on depersonalization and the indirect effect on
emotional exhaustion through teaching-related and non-teachingrelated workload. Although the relationship with both types of
workload were signicant, it is interesting to nd that the one with
non-teaching related workload was strongest. This is particularly
interesting, because the intensication thesis states that, due to the
rise of accountability demands and standardization over the last
decades, the autonomy of teachers is diminishing (Ballet et al.,
2006). Our results show that school leaders play an important
role in mediating the experience of intensication. By increasing
the autonomy of teachers they can enable teachers to better cope
with teaching-related and especially non-teaching-related workload, such as paperwork and accountability demands. If teachers
are involved in school policy-decisions and maintain positive relations with their supervisor and colleagues it is probable that
external pressures and demands can be better identied and
mitigated. More autonomy may enable teachers to regulate these
tasks in ways that are more correspondent to their own views.
In the same line our results showed that teachers experience
more autonomy when they receive detailed information about
policy changes and experience support with policy change implementation. More information about policy changes helps teachers
to have a better understanding of the context of the imposed
changes. That nding also supports the notion of Ballet et al.
(2006): in the experience of intensication, teachers react in
complex ways and do not just become technical professionals
who blindly follow policy dictates. When teachers experience autonomy they are enabled to cope actively and creatively with
changes in the workplace.
Our results indicate, however, that non-teaching-related workload and especially teaching-related workload are positively
related to emotional exhaustion. This illustrates that the total

107

workload is as important as the kind of workload experienced. That

is, teaching-related workload together with the relationships
maintained with the students has the strongest effect on emotional
exhaustion. This nding conrms the central role of the quality of
relationships in the experience of burnout. It might indicate that
the most draining and also buffering aspect is the workload coming
from the teaching activity itself and the quality of the relationships
experienced therein. Nevertheless, non-teaching-related workload
also plays an important role. Excessive paperwork and high-stakes
accountability demands are among the most important stressors
and causes of teacher job dissatisfaction (Byrne, 1991; Scott et al.,
2001; Van Droogenbroeck & Spruyt, 2014; Varlaam et al., 1992).
When teachers feel that they are accountable at all times and
constantly have to prove their competence, than this undermines
their professional self-condence (Kelchtermans, 2005). Senior
teachers may be especially vulnerable to this kind of pressure, since
they have experienced the rise of high-stakes accountability demands rsthand. It would help teachers if school leaders and policymakers try to align the non-teaching demands with the
professional standards teachers set for themselves, namely, the
well-being and achievements of their pupils. Today the added value
of administrative tasks for the well-being of their pupils is often
unclear or perceived as disproportionate to the time investment
required. Research by Smylie (1994) indicated that teachers are
more willing to accept changes which have a positive impact on the
learning process of pupils. This illustrates that for teachers the
problem probably does not lie in dealing with accountability or
policy changes themselves but in how accountability is dened and
how it is put to use. It is reasonable to expect that teachers account
for their work with students as part of their responsibility as a
professional and to some extent administrative tasks are of course
part and parcel of the contemporary teaching profession. In this
context one could raise the question whether teachers in training
are sufciently prepared for the administrative component of the
job. In several countries the job of teachers has traditionally been
dened in terms of classroom teaching hours (OECD, 2005).
Broader job proles may be needed; these should include the
complex and wide array of teachers' tasks and responsibilities.
For senior teachers, burnout can be a factor that makes them
leave the profession or retire early (Leung & Lee, 2006; Rudow,1999;
Skaalvik & Skaalvik, 2011; Van Droogenbroeck & Spruyt, 2014). As
expected, older teachers reported less emotional exhaustion and
more personal accomplishment than younger teachers. Previous
research indicated that a widespread early exit culture exists among
teachers (Eurydice Network, 2012). In Belgium teachers make
massive use of early exit schemes (Van Droogenbroeck & Spruyt,
2014). Our ndings might be an indication for the healthy worker
effect at the end of the career where those teachers who feel burned
out leave the profession as early as possible, while motivated, happy,
and healthy teachers are those who keep teaching.
We found a differential relation between the burnout dimensions and interpersonal relationships, autonomy, and
teaching-related and non-teaching-related workload. That nding
supports the multidimensional conceptualization of burnout; that
is, the three dimensions of burnout should not be added up to a
single measure (Byrne, 1994; Langballe, 2006; Skaalvik & Skaalvik,
2009). Although burnout is a multidimensional construct, this does
not mean that its dimensions are unrelated to each other. Often in
teacher burnout studies, the three burnout dimensions are
assumed to be unrelated (e.g., Grayson et al., 2008; Kokkinos, 2007;
Skaalvik & Skaalvik, 2009; Van Der Doef & Maes, 2002). This may
cause an overestimation of direct paths, because mediated effects
are not taken into account. We conrmed our expectation that
there indeed exists an inverse relation between the amount of
direct paths from interpersonal relationships to the burnout

108

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

dimensions and the strength of the associations between the

burnout dimensions. Based on our literature study, we postulated
11 direct paths from interpersonal relationships to the burnout
dimensions. Eventually, only ve of these direct paths remained in
the nal model when the relations between burnout dimensions
were specied. By regarding burnout as a sequential process, researchers may better understand the factors that contribute to its
development and take into account the mediation of effects.
Our study has several limitations. First, the cross-sectional
design based on self-reported measures does not allow causal inferences to be made between the investigated variables. Second, we
measured only four interpersonal relationships at work. Other
interpersonal relationships, such as studentepeer relations and
parentecommunity relations, can also have an inuence on
burnout (Grayson et al., 2008). Third, we used two one-item scales
to assess teaching-related and non-teaching-related workload. This
may have led to an underestimation of the inuence of these independent variables. Future research could benet by further
exploring this distinction. A multidimensional analysis of what the
work of teachers entails could give more insight in which aspects
inuence the burnout process.
Acknowledgement
FundingThe author(s) disclosed receipt of the following nancial support for the research of this article: This work was partly
supported by the Flemish Department of Education [grant number
OBPWO 09.06].
References
Alarcon, G. M. (2011). A meta-analysis of burnout with job demands, resources, and
attitudes. Journal of Vocational Behavior, 79(2), 549e562.
Anderson, M. B. G., & Iwanicki, E. F. (1984). Teacher motivation and its relationship
to burnout. Educational Administration Quarterly, 20(2), 109e132.
Apple, M. W. (1988). Teachers and texts: A political economy of class and gender relations in education. New York, NY: Routledge.
Ball, S. J. (2003). The teacher's soul and the terrors of performativity. Journal of
Education Policy, 18(2), 215.
Ballet, K., & Kelchtermans, G. (2009). Struggling with workload: primary teachers' experience of intensication. Teaching and Teacher Education, 25(8),
1150e1157.
Ballet, K., Kelchtermans, G., & Loughran, J. (2006). Beyond intensication towards a
scholarship of practice: analysing changes in teachers' work lives. Teachers and
Teaching: Theory and Practice, 12(2), 209e229.
Brookhart, S. M., & Freeman, D. J. (1992). Characteristics of entering teacher candidates. Review of Educational Research, 62(1), 37e60.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model t. In
K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136e162).
Newbury Park, CA: Sage.
Burke, R. J., & Greenglass, E. (1995). A longitudinal study of psychological burnout in
teachers. Human Relations, 48(2), 187e202.
Byrne, B. M. (1991). Burnout: investigating the impact of background variables for
elementary, intermediate, secondary, and university educators. Teaching and
Teacher Education, 7(2), 197e209.
Byrne, B. M. (1994). Burnout: testing for the validity, replication, and invariance of
causal structure across elementary, intermediate, and secondary teachers.
American Educational Research Journal, 31(3), 645e673.
Byrne, B. M. (2011). Structural equation modeling with Mplus: Basic concepts, applications, and programming (1st ed.). New York, NY: Routledge Academic.
~ oz, E. M., & Carrasco-Ortiz, M. A. (2005). Personality
Cano-Garca, F. J., Padilla-Mun
and contextual variables in teacher burnout. Personality and Individual Differences, 38, 929e940.
Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). Internet, mail and mixed-mode
surveys: The tailored design method (3rd ed.). Hoboken: Wiley.
Dinham, S., & Scott, C. (1998). A three domain model of teacher and school executive career satisfaction. Journal of Educational Administration, 36(4), 362e378.
Dorman, J. P. (2003). Relationship between school and classroom environment
and teacher burnout: a LISREL analysis. Social Psychology of Education, 6(2),
107e127.
Ekerdt, D. J. (2009). Frontiers of research on work and retirement. The Journals of
Gerontology Series B: Psychological Sciences and Social Sciences, 65B(1), 69e80.
ki, M., Sinervo, T., Sutinen, R., & Laine, M. (2005). Job
Elovainio, M., Forma, P., Kivima
demands and job control as correlates of early retirement thoughts in Finnish
social and health care employees. Work & Stress, 19(1), 84e92.

Eurydice Network. (2012). Key data on education in Europe 2012. Education, Audiovisual and Culture Executive Agency e European Commission.
Friedman, I. A. (1995). Student behavior patterns contributing to teacher burnout.
The Journal of Educational Research, 88(5), 281e289.
Friedman, I. A. (2003). Self-efcacy and burnout in teaching: the importance of
interpersonal-relations efcacy. Social Psychology of Education, 6(3), 191e215.
Gavish, B., & Friedman, I. A. (2010). Novice teachers' experience of teaching: a dynamic aspect of burnout. Social Psychology of Education, 13(2), 141e167.
Grayson, J. L., Alvarez, H. K., & Grayson, J. L. A. H. K. (2008). School climate factors
relating to teacher burnout: a mediator model. Teaching and Teacher Education,
24(5), 1349e1363.
Greenglass, E. R., Burke, R. J., & Konarski, R. (1997). The impact of social support on
the development of burnout in teachers: examination of a model. Work & Stress,
11(3), 267e278.
Hakanen, J. J., Bakker, A. B., & Schaufeli, W. B. (2005). Burnout and work engagement
among teachers. Journal of School Psychology, 43(6), 495e513.
Hargreaves, A. (1994). Intensication. Teachers' work e better or worse?. Changing
teachers, changing times: Teachers' work and culture in the postmodern age (pp.
117e140) London: Cassell.
Heus, P. D., & Diekstra, R. F. W. (1999). Do teachers burn out more easily? A comparison of teachers with other social professions on work stress and burnout
symptoms. In R. Vandenberghe, A. M. Huberman, R. Vandenberghe, &
A. M. Huberman (Eds.), Understanding and preventing teacher burnout (pp.
269e284). Cambridge: Cambridge University Press.
Horn, J. E., & Schaufeli, W. B. (1998). Maslach burnout inventory: The Dutch educators
survey (MBI-NL-ES) psychometric evaluations. Manual (unpublished manuscript).
Utrecht University, Department of Social and Organizational Psychology.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for t indexes in covariance structure
analysis: conventional criteria versus new alternatives. Structural Equation
Modeling, 6(1), 1e55.
Kelchtermans, G. (1999). Teaching career: between burnout and fading away? Reections from a narrative and biographical perspective. In R. Vandenberghe, &
A. M. Huberman (Eds.), Understanding and preventing teacher burnout (pp.
176e191). Cambridge: Cambridge University Press.
Kelchtermans, G. (2005). Teachers' emotions in educational reforms: selfunderstanding, vulnerable commitment and micropolitical literacy. Teaching
and Teacher Education, 21(8), 995e1006.
Kinman, G., Wray, S., & Strange, C. (2011). Emotional labour, burnout and job
satisfaction in UK teachers: the role of workplace social support. Educational
Psychology, 31(7), 843e856.
Kokkinos, C. M. (2007). Job stressors, personality and burnout in primary school
teachers. British Journal of Educational Psychology, 77, 229e243.
Langballe, E. M. (2006). The factorial validity of the Maslach burnout inventorygeneral survey in representative samples of eight different occupational
groups. Journal of Career Assessment, 14(3), 370e384.
Lau, P. S. Y., Yuen, M. T., & Chan, R. M. C. (2005). Do demographic characteristics
make a difference to burnout among Hong Kong secondary school teachers?
Social Indicators Research, 71(1e3), 491e516.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the
three dimensions of job burnout. Journal of Applied Psychology, 81(2), 123e133.
Leiter, M. P., & Maslach, C. (1988). The impact of interpersonal environment on burnout
and organizational commitment. Journal of Organizational Behavior, 9(4), 297e308.
Leithwood, K., Menzies, T., Jantzi, D., & Leithwood, J. (1999). Teacher burnout: a
critical challenge for leaders of restructuring schools. In R. Vandenberghe, &
A. M. Huberman (Eds.), Understanding and preventing teacher burnout (pp.
85e114). Cambridge: Cambridge University Press.
Leung, D. Y. P., & Lee, W. W. S. (2006). Predicting intention to quit among Chinese
teachers: differential predictability of the components of burnout. Anxiety,
Stress & Coping, 19(2), 129e141.
Martin, N. K., Sass, D. A., & Schmitt, T. A. (2012). Teacher efcacy in student
engagement, instructional management, student stressors, and burnout: a
theoretical model using in-class variables to predict teachers' intent-to-leave.
Teaching and Teacher Education, 28(4), 546e559.
Maslach, C. (2003). Burnout: The cost of caring. Los Altos, CA: Malor Books.
Maslach, C., & Jackson, S. E. (1984). Burnout in organizational settings. In S. Oskamp
(Ed.), Applied social psychology annual: Applications in organizational settings.
Beverly Hills, CA: Sage Publications.
Maslach, C., & Jackson, S. E. (1986). Maslach burnout inventory (2nd ed.). Palo Alto,
CA: Consulting Psychologists Press.
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of
Psychology, 52(1), 397e422.
Mazur, P. J., & Lynch, M. D. (1989). Differential impact of administrative, organizational, and personality-factors on teacher burnout. Teaching and Teacher Education, 5(4), 337e353.
n, L. K., & Muthe
n, B. O. (1998). Mplus user's guide (7th ed.). Los Angeles, CA:
Muthe
n & Muthe
n.
Muthe
Noor, N. M., & Zainuddin, M. (2011). Emotional labor and burnout among female
teachers: workefamily conict as mediator: emotional labor and burnout. Asian
Journal of Social Psychology, 14(4), 283e293.
OECD. (2005). Teachers matter: Attracting, developing and retaining effective teachers.
Education and training policy. OECD Publishing.
OECD. (2012). Education at a glance 2012: OECD indicators. OECD Publishing.
Park, H. I., Jacob, A. C., Wagner, S. H., & Baiden, M. (2013). Job control and
burnout: a meta-analytic test of the conservation of resources model. Applied
Psychology. http://dx.doi.org/10.1111/apps.12008.

F. Van Droogenbroeck et al. / Teaching and Teacher Education 43 (2014) 99e109

Pierce, C. M. B., & Molloy, G. N. (1990). Psychological and biographical differences
between secondary school teachers experiencing high and low levels of
burnout. British Journal of Educational Psychology, 60(1), 37e51.
Roness, D. (2011). Still motivated? The motivation for teaching during the second
year in the profession. Teaching and Teacher Education, 27(3), 628e638.
Rudow, B. (1999). Stress and burnout in the teaching profession: European studies,
issues, and research perspectives. In R. Vandenberghe, & A. M. Huberman (Eds.),
Understanding and preventing teacher burnout (pp. 115e138). Cambridge:
Cambridge University Press.
Russell, D. W., Altmaier, E., & Vanvelzen, D. (1987). Job-related stress, social support,
and burnout among classroom teachers. Journal of Applied Psychology, 72(2),
269e274.
Schaufeli, W. B., & Bakker, A. B. (2004). Job demands, job resources, and their
relationship with burnout and engagement: a multi-sample study. Journal of
Organizational Behavior, 25(3), 293e315.
Schaufeli, W. B., Daamen, J., & Van Mierlo, H. (1994). Burnout among Dutch
teachers: an MBI-validity study. Educational and Psychological Measurement,
54(3), 803e812.
Schaufeli, W. B., & Enzmann, D. (1998). The burnout companion to study and practice:
A critical analysis. London: Taylor & Francis Group.
Schreurs, B., Van Emmerik, H., De Cuyper, N., Notelaers, G., & De Witte, H. (2010).
Job demandseresources and early retirement intention: differences between
blue-and white-collar workers. Economic and Industrial Democracy, 32(1),
47e68.
Scott, C., Stone, B., & Dinham, S. (2001). I love teaching but international patterns of teacher discontent. Education Policy Analysis Archives, 9, 28.
Skaalvik, E. M., & Skaalvik, S. (2009). Does school context matter? Relations with
teacher burnout and job satisfaction. Teaching and Teacher Education, 25(3),
518e524.
Skaalvik, E. M., & Skaalvik, S. (2010). Teacher self-efcacy and teacher burnout: a
study of relations. Teaching and Teacher Education, 26(4), 1059e1069.
Skaalvik, E. M., & Skaalvik, S. (2011). Teacher job satisfaction and motivation to leave
the teaching profession: relations with school context, feeling of belonging, and
emotional exhaustion. Teaching and Teacher Education, 27(6), 1029e1038.
Smeyers, P., & Depaepe, M. (Eds.). (2008). Educational research: The educationalization of social problems. Dordrecht: Springer Netherlands.

109

Smylie, M. A. (1994). Redesigning teachers' work: connections to the classroom

(Chapter 3) Review of Research in Education, 20(1), 129e177.
Stoeber, J., & Rennert, D. (2008). Perfectionism in school teachers: relations with
stress appraisals, coping styles, and burnout. Anxiety, Stress & Coping, 21(1),
37e53.
Struyven, K., Jacobs, K., & Dochy, F. (2012). Why do they want to teach? The multiple
reasons of different groups of students for undertaking teacher education. European Journal of Psychology of Education, 1e16.
Taris, T. W., Le Blanc, P. M., Schaufeli, W. B., & Schreurs, P. J. G. (2005). Are there
causal relationships between the dimensions of the Maslach burnout inventory? A review and two longitudinal tests. Work & Stress, 19(3), 238e255.
Toker, S., & Biron, M. (2012). Job burnout and depression: unraveling their temporal
relationship and considering the role of physical activity. Journal of Applied
Psychology, 97(3), 699e710.
Unterbrink, T., Hack, A., Pfeifer, R., Buhl-Griehaber, V., Mller, U., Wesche, H., et al.
(2007). Burnout and effortereward-imbalance in a sample of 949 German
teachers. International Archives of Occupational and Environmental Health, 80(5),
433e441.
Valli, L., & Buese, D. (2007). The changing roles of teachers in an era of high-stakes
accountability. American Educational Research Journal, 44(3), 519e558.
Van Der Doef, M., & Maes, S. (2002). Teacher-specic quality of work versus general
quality of work assessment: a comparison of their validity regarding burnout,
(psycho)somatic well-being and job satisfaction. Anxiety, Stress & Coping, 15(4),
327e344.
Van Droogenbroeck, F., & Spruyt, B. (2014). To stop or not to stop: an empirical
assessment of the determinants of early retirement among active and
retired senior teachers. Research on Aging. http://dx.doi.org/10.1177/
0164027513519449, 0164027513519449.
Varlaam, A., Nutall, D., & Walker, A. (1992). What makes teachers tick?. Clare market
papers no. 4. London: Centre for Education Research, London School of
Economics.
Woods, P. (1999). Intensication and stress in teaching. In R. Vandenberghe, &
A. M. Huberman (Eds.), Understanding and preventing teacher burnout (pp.
115e138). Cambridge: Cambridge University Press.