|
RESULTS
Scale Statistics
Based on the approach described in the preceding section, optimal
reliability coefficients (Cronbach Coefficient alpha) were computed
for each scale (see Table 3). These results show that all scales
had at least satisfactory internal consistency. Indices ranged from
.66 for role
ambiguity to .91 for school environment and compared favourably
with those reported in previous learning environment and burnout
research (e.g. Byrne, 1994; Dorman, Adams, & Ferguson, in press;
Dorman et al., 1997; Maslach et al., 1996). Table 3 also shows values
for 
and 
for each scale which provide a sound basis for examining the postulated
structural
model. Means and standard deviations for each scale are also shown
in Table 3.
LISREL Analyses
The LISREL analysis of the postulated (initial) model shown in Figure
1 revealed a sound but not outstanding fit to the data. Fit and
modification indices shown in Table 4 indicated that better fit
might be achieved if one path (self-esteem ¡æ emotional
exhaustion) was added to the model. Additionally, two paths (role
ambiguity ¡æ self-esteem and role ambiguity ¡æ
personal accomplishment) were not statistically significant (¥â
=.16 , p = .11 and ¥â = .31, p = .17 respectively). Accordingly,
role ambiguity was deleted from the model. Fit indices for this
modified model improved with RMSEA reduced to .05. Two further model
revisions involving the addition of two paths resulted in significant
changes in ¥ö2 (see Table 4). The final model fitted the
data very well with an RMSEA of .03 and TLI of 1.00. Structural
components of this model with standardized path coefficients are
shown in Figure 2. All path coefficients in the final model were
statistically significant (p <.05). The strength and direction
of these coefficients are plausible. For example, role overload
was a strong, positive predictor of work pressure (¥â
= .91) which, in turn, positively predicted emotional exhaustion.
Self-esteem was negatively related to both emotional exhaustion
(¥â = -.37) and depersonalisation (¥â = -.30)
but positively related to personal accomplishment (¥â
= .22). The strongest negative relationship was between role conflict
and school environment (¥â = -1.07). The squared multiple
correlation coefficient for the prediction of personal accomplishment
(R2) was computed to be .64 which indicates that 64% of variance
in personal
accomplishment could be explained by its contributing variables
(viz. depersonalisation, selfesteem, teaching efficacy). Similarly,
work pressure, classroom environment and self-esteem accounted for
69% of variance in emotional exhaustion. Over 46% of variance in
depersonalisation was attributable to school environment, classroom
environment, emotional exhaustion and self-esteem. The coefficient
of determination for all structural equations jointly was computed
to be .98. That is 98% of variance in the set of dependent variables
(viz.
school environment, work pressure, self-esteem, emotional exhaustion,
depersonalisation and personal accomplishment was explained by the
set of independent variables (viz. role overload, role conflict,
classroom environment, teaching efficacy and external locus of control).
Overall, Figure 2 provides a comprehensive structural model for
burnout based on the data collected in the present study.
DISCUSSION
The results of this study can be discussed in terms of both the
full model and specific organizational and personality variables.
Holistically, the results support, to some degree, the burnout models
for primary, intermediate and secondary school teachers developed
by Byrne (1994). Those models involved role conflict, work overload,
classroom climate, decision making, peer support, self-esteem and
external locus of control as predictors of the same three burnout
dimensions of the present study: emotional exhaustion, depersonalisation
and personal accomplishment. However, differences in model structure
among these three levels of schooling were evident. For example,
Byrne found that external locus of control was a predictor of one
burnout dimension (viz. personal accomplishment) for secondary teachers
only. In the present study, the burnout predictors were role overload,
role conflict, classroom environment, school environment, work pressure,
teaching efficacy, self-esteem and external locus of control. It
is particularly noteworthy that while role ambiguity was included
in the present study as a possible burnout predictor, this variable
was deleted from the final model.
This is consistent with Byrne’s findings. Clearly, role ambiguity
is not a significant predictor of burnout and its retention in future
modelling is questionable. One point of contrast between Byrne’s
research and the present study concerns the number of statistically
significant paths. While Byrne found six statistically significant
paths common to all three types of school teachers, the present
study identified 12 such paths. While this difference could be due
to the additional variables included in the present study (e.g.
school
environment, work pressure, teaching efficacy), it could also be
due to two other characteristics of the present study: improved
instrumentation and the data modelling approaches employed. The
classroom and school environment scales were based on the latest
instrumentation in learning environment research. Teaching efficacy
was assessed with a well-validated scale developed by Midgley et
al. (1997). In addition, data modelling involved congeneric measurement
models that maximized reliabilities and facilitated the fixing of
path
loadings and error variances for observed variables in structural
equation modelling.
With regard to the organizational variables that impacted on teacher
burnout, role overload was a potent variable. It influenced work
pressure and subsequently emotional exhaustion. Clearly, teachers
who experience significant role overload reported high levels of
work
pressure and this pressure leads to increased levels of emotional
exhaustion. Role conflict was also a potent negative predictor of
school environment which, in turn predicted negatively depersonalisation.
While Byrne’s model did not include school environment as a
possible predictor or mediating variable of burnout, it did find
that role conflict influenced depersonalisation in secondary teachers
only. That role conflict influences negatively teachers’ perceptions
of overall school environment is not surprising. If teachers are
confused
about their work demands, they are not likely to report a positive
overall working environment. The negative relationship between school
environment and depersonalisation supports the finding of the only
study linking school environment and burnout in Australian
and Singaporean schools (see Ball, Moselle, & Fraser, 1995).
In that study, significant associations were found between six of
the seven dimensions of the School-Level Environment Questionnaire
(SLEQ) (Fraser, 1994) and depersonalisation. Positive school environments
were associated with reduced levels of depersonalisation.
The final organizational variable, classroom environment was found
to have significant negative relationships with emotional exhaustion
and depersonalisation. While this finding supports Byrne’s
contention that classroom environment is a key nomological predictor
of
burnout, it should be noted that the present study used a set of
items with a Cronbach Coefficient alpha of .80 from the WIHIC –
an instrument developed within contemporary learning environment
research (see Fraser, 1998). Byrne used Bacharach, Bauer and Conley’s
(1986) Classroom Environment Scale which reported a Cronbach Coefficient
alpha of .60. It is recommended that future burnout research employ
scales that have been developed within the generally accepted framework
for studying classroom environments.
The three personality variables, teaching efficacy, external locus
of control and self-esteem were all predictors of personal accomplishment.
However, while teaching efficacy has a significant direct effect
on personal accomplishment and an indirect effect via self-esteem,
external locus of control had only an indirect negative effect via
self-esteem. Clearly, teachers with an elevated external locus of
control tended to have reduced self-esteem which then reduced personal
accomplishment. Additionally, self-esteem had significant negative
effects on emotional exhaustion and depersonalisation. These effects
are highly plausible.
Finally, it is clear that the present modelling supports the hypothesized
intraburnout relationships of emotional exhaustion influencing positively
depersonalisation which subsequently influences negatively personal
accomplishment. It is noteworthy that the direction and strength
of these relationships are consistent with those reported by Byrne
(1994) for intermediate and secondary school teachers.
CONCLUSION
The issue of teacher burnout is very important to schools and school
systems. For too long, teacher burnout has been explained largely
in terms of individual teacher personality characteristics. Such
a narrow set of predictors has suited employers who do not wish
to accept any moral or legal responsibility for burning out teachers
through poor organizational and managerial processes. The present
research focused on both organizational variables and personality
factors of individual teachers. It has reported the validation of
a nomological
network of burnout for teachers by building upon and extending prior
research in the burnout area, principally the work of Byrne (1994).
This Australian research needs to be complemented by further research
in Western countries so that a widely generalizable model for teacher
burnout can be developed and a theory of teacher burnout firmly
established. One direction for future research would be the validation
of the model of teacher burnout postulated recently by Maslach and
Leiter (1999). In this elaborate model, political, policy
and economic contexts, school ecology, task qualities (the work
of teachers), organizational characteristics and teacher person
qualities combine to predict burnout. Consequences of burnout are
teacher behaviour which influences student perceptions and evaluation,
and
subsequent student behaviours and outcomes. While it might be desirable
to have simple solutions to the teacher burnout issue, the reality
is that, as Maslach (1999) suggests, such a complex social phenomenon
will require complex solutions.
REFERENCES
Abel, M. H., & Sewell, J. (1999). Stress and burnout in rural
and urban secondary school teachers. Journal of Educational Research,
92, 287- 293.
Aldridge, J. M., & Fraser, B. J. (2000). A cross-cultural study
of classroom learning
Anderson, M. B. & Iwanicki, E. F. (1984). Teacher motivation and
its relationship to burnout.
Boyle, G. J., Borg, M. G., Falzon, J. M., & Baglioni, A. J. (1995).
A structural model of the dimensions of teacher stress. British Journal
of Educational Psychology, 65, 49-67.
Bransgrove, E. (1994). A decade of teacher stress: The changing nature
of the determinants of teacher stress, 1981 to 1991. South Pacific
Journal of Teacher Education, 22, 39-52.
Brouwers, A., & Tomic, W. (2000). A longitudinal study of teacher
burnout and perceived self-efficacy in classroom management. Teaching
and Teacher Education, 16, 239-253.
Byrne, B. M. (1991). Burnout: Investigating the impact of background
variables for elementary, intermediate, secondary, and university
educators. Teaching and Teacher Education, 7, 197-209.
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,
645-673.
Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS,
and SIMPLIS: Basic concepts, applications and programming. Mahwah,
NJ: Lawrence Erlbaum.
Coopersmith, S. (1981). Self-esteem inventories. Palo Alto, CA: Consulting
Psychologists press.
Dorman, J. P., Adams, J. E., & Ferguson, J. M. (2002). Psychosocial
environment and student self-handicapping in secondary school mathematics
classes: A cross-national study. Educational Psychology, 22, 499-511.
Dorman, J. P., Fraser, B. J., & McRobbie, C. J. (1997). Relationship
between school-level and classroom-level environment in secondary
schools. Journal of Educational Administration, 35, 74-91.
Fraser, B. J. (1994). Research on classroom and school climate. In
D. Gabel (Ed.), Handbook of research on science teaching and learning
(pp. 493-541). New York: Macmillan.
Fraser, B. J. (1998). Science learning environments: Assessments,
effects and determinants. In B. J. Fraser & K. G. Tobin (Eds.),
International handbook of science education (pp. 527-564). Dordrecht,
The Netherlands: Kluwer.
Freudenberger, H. J. (1974). Staff burnout. Journal of Social Issues,
30, 159-165.
Guglielmi, R. S., & Tatrow, K. (1998). Occupational stress, Burnout,
and health in teachers: A methodological and theoretical analysis.
Review of Educational Research, 68, 61-99.
Hart, P. M., Wearing, A. J., & Conn, M. (1995). Conventional wisdom
is a poor predictor of the relationship between discipline policy,
student misbehaviour and teacher stress. British Journal of Educational
Psychology, 65, 27-48.
Holmes-Smith, P., & Rowe, K. J. (1994, January). The development
and use of congeneric measurement models in school effectiveness research:
Improving the reliability and validity of composite and latent variables
for fitting multilevel and structural equation models. Paper presented
at the International Congress for School Effectiveness and Improvement,
Melbourne.
Jackson, S. E., Schwab, R. L., & Schuler, R. S. (1986). Toward
an understanding of the burnout phenomenon. Journal of Applied Psychology,
71, 630-640.
Jöreskog, K. G., & Sörbom, D. (1989). LISREL 7: A guide
to the program and its applications (2nd ed.). Chicago, IL: SPSS Inc.
Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: User's
reference guide. Chicago, IL: Scientific Software International.
Kelloway, E. K. (1998). Using LISREL for structural equation modeling:
A researcher's guide. Thousand Oaks, CA: Sage.
Lens, W., & Jesus, S. N. D. (1999). A psychosocial interpretation
of teacher stress and burnout. In R. Vandenburghe & A. M. Huberman
(Eds.), Understanding and preventing teacher stress: A sourcebook
of international research and practice (pp. 192-202). Cambridge, UK:
Cambridge University Press.
MacDonald, A. P. (1974). Measures of internal-external control. In
J. P. Robinson & P. R.
Shaver (Eds.), Measures of social psychological attitudes (pp. 227-236).
Ann Arbor, MI: Institute for Social Research, University of Michigan.
Malik, J. L., Mueller, R. O, & Meinke, D. L. (1991). The effects
of teaching experience and grade level taught on teacher stress: A
LISREL analysis. Teaching and Teacher Education, 7, 57-62.
Maslach, C. (1999). Progress in understanding teacher burnout. In
R. Vandenburghe & A. M.
Huberman (Eds.), Understanding and preventing teacher stress: A sourcebook
of international research and practice (pp. 211-222.). Cambridge,
UK: Cambridge University Press. Maslach, C. & Jackson, S. E. (1981).
The measurement of experienced burnout. Journal of
Occupational Behaviour, 2, 99-113.
Maslach, C., Jackson, S. E., & Leiter, M. (1996). The Maslach
Burnout Inventory (3rd ed.). Palo Alto: Consulting Psychologists Press.
Maslach, C., & Leiter, M. P. (1999). Teacher burnout: A research
agenda. In R. Vandenburghe & A. M. Huberman (Eds.), Understanding
and preventing teacher stress: A sourcebook of international research
and practice (pp. 295-314). Cambridge, UK: Cambridge University Press.
Midgley, C., Maehr, M., Hicks, L., Roeser, R., Urdan, T., Anderman,
E., et al. (1997). Patterns of Adaptive Learning Survey (PALS). Ann
Arbor: University of Michigan.
Munck, I. M. E. (1979). Model building in comparative education: Applications
of the LISREL method to cross-national survey data. Stockholm: Almqvist
& Wiksell.
Otto, R. (1986). Teacher under stress. Melbourne: Hill.
Pettegrew, L. S., & Wolf, G. E. (1982). Validating measures of
teacher stress. American Educational Research Journal, 19, 373-396.
Rotter, J. B. (1966). Generalized expectancies for internal versus
external control of reinforcement. Psychological Monographs, 80, 1-28.
Schumacker, R. E., & Lomax, R. G. (1996). A beginner's guide to
structural equation modeling. Mahwah, NJ: Erlbaum.
Tang, T., & Yeung, A. S. (1999, February). Hong Kong teachers’
sources of stress, burnout and job satisfaction. Paper presented at
the International Conference on Teacher Education, Hong Kong. |
Ergopolis
