Directory UMM :Data Elmu:jurnal:I:International Journal of Educational Management:Vol15.Issue1.2001:
A survey of the health of Victorian primary school
principals
Rod Green
Senior Lecturer, La Trobe University, Bundoora, Australia
Susan Malcolm
Senior Lecturer, La Trobe University, Bundoora, Australia
Ken Greenwood
La Trobe University, Bundoora, Australia
Michael Small
Victorian Primary Principals Association, North Melbourne, Australia
Gregory Murphy
La Trobe University, Bundoora, Australia
Keywords
Teachers, Education, Schools,
Surveys
Abstract
In recent years responsibility for
the administration of schools
internationally has shifted from
education departments towards
self-governing schools. This trend
has resulted in major changes to
the role of school principals. Such
changes in role may impact on the
psychological and physical health
of principals, but there has been
very little research into this
population. A survey of the health
and wellbeing of a representative
sample of 50 principals of State
primary schools in Victoria,
Australia is reported. Subjects
completed questionnaires
measuring health-related
behaviour and stress and arousal
levels and participated in
comprehensive health appraisals.
Principals reported better smoking
patterns than the population as a
whole. Despite a higher
socioeconomic status than the
population as a whole, the health
status of the principals was not
apparently better. Principals
reported higher stress levels and
worse physical health than a group
of white-collar employees of
similar socioeconomic status.
This is the first of two
related articles taken from
the authors' recent research
into the health of primary
school principals in Victoria,
Australia
The International Journal of
Educational Management
15/1 [2001] 23±30
# MCB University Press
[ISSN 0951-354X]
Introduction
There is a large body of literature
internationally which has documented the
high prevalence of stress and burnout in the
teaching profession (Kyriacou, 1987). The
majority of this research relates to classroom
teachers while comparatively few papers
report on stress amongst school principals
(Tung and Koch, 1980; Cooper et al., 1988;
Savery and Detiuk, 1986). One study which
has examined stressors amongst school
principals and other school administrators
indicated that the major stressors for
principals related to the following four
dimensions in decreasing order of
importance: role-based stress, task-based
stress, conflict-mediating stress and
boundary-spanning stress (Tung and Koch,
1980). Principals scored higher than regionbased administrators on all of these
dimensions except boundary-spanning,
which relates to the interaction between the
school and the external environment. A
survey of Western Australian principals
found that primary principals had higher
stress levels than secondary principals
(Savery and Detiuk, 1986). Another study
attempted to correlate continuously
monitored heart rate as an indicator of
physiological stress to principal behaviour,
and found variations in heart rate related to
personality type (Cooper et al., 1988).
Despite concerns about the relationship
between occupational stress and stressrelated illness (Cooper and Marshall, 1976;
Kyriacou, 1987; Savery and Detiuk, 1986),
there has been far less work conducted on the
general health characteristics of teachers
and, again, even less relating to the health
status of principals (Smith et al., 1988). One
report using self-report of health
characteristics on the multidimensional
The current issue and full text archive of this journal is available at
http://www.emerald-library.com/ft
General Wellbeing Questionnaire suggested
that principals had above average health
status (Smith et al., 1988). This study reported
that principals had low drug (tobacco,
alcohol) usage, but some health behaviours,
particularly diet and exercise needed to be
improved. Another study involving selfreport of various stress-related illnesses
found a higher incidence of stress-related
illness, particularly amongst primary
principals, than the general population
(Savery and Detiuk, 1986).
There is evidence that prolonged
occupational stress can lead to both mental
and physical ill-health (Kyriacou, 1987;
Cooper and Marshall, 1976), and a study of
school administrators revealed a strong
inverse relationship between stress levels
and self-reported physical health status
(Tung and Koch, 1980). Concern about high
stress levels amongst teachers in part relates
to the cost of workers' compensation and
potentially the loss of staff through
associated ill-health retirement (Kyriacou,
1987). Such concerns about consequent
workers' compensation claims led to the
instigation of a government initiated report
in Victoria, Australia (University of
Melbourne Applied Psychology Research
Group, 1990). This report included 47 specific
recommendations to reduce teacher stress;
however one general comment of interest in
the recommendations section was that ``the
Ministry (of Education) should slow the
perceived rate of change, and so reduce the
extent to which it creates uncertainty in the
school environment'' (University of
Melbourne Applied Psychology Research
Group, 1990).
Well-documented changes in public
education, both internationally and in
Australia, over the last decade have involved
devolution of financial, staffing and planning
processes from Departments of Education to
individual schools (Cranston, 1994; Caldwell,
1992). The introduction in 1993 of the Schools
of The Future (SoF) program in government
[ 23 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
schools in the State of Victoria has been
typical of this type of change. Such changes
have modified the traditional role of the
principal to a great extent such that much
more managerial, as opposed to the previous
administrative, skills are required (Cranston,
1994). It has been suggested that this
increasing responsibility for principals may
lead to stress-related illness (Thornton, 1996).
In light of the potential impact of these factors
on principal stress levels and principal health,
and the lack of existing data on the health and
health-related behaviour characteristics of
school principals, this study was undertaken
to ascertain the health status of principals
working under the SoF system.
Method
This study was initiated by the Victorian
Primary Principals Association (VPPA) out
of concern for the overall health and
wellbeing of its members associated with the
introduction of SoF. The VPPA's concern
related to rapidly changing work practices
and lifestyles reported by principals in order
to cope with the increased workload,
accountability and responsibility inherent in
the SoF reforms.
Although the DoE was initiating a free statewide screening program for principals, the
VPPA felt that this program was inadequate to
assist principals in maintaining their health
and wellbeing in the face of increasing workrelated demands. Hence this project involved
two stages; a survey of principal health, the
results of which are presented in this paper,
and the introduction of a comprehensive
health promotion (HP) program to improve
principal health. The outcomes of the HP
program are presented elsewhere (Green et
al., 2000).
The VPPA, the DoE, HBA Health
Management (the provider of the health
promotion program), and principals in the
program made equal contributions to the cost
of the program. The initial health survey,
which was the first phase of the HP program,
took place in March 1996.
Participants
Approximately 1,450 VPPA members were
contacted and asked to give an expression of
interest in participating in a HP program
conducted by an external provider which
included a heavily subsidised comprehensive
health assessment as a component of a
broader HP program. The 99 members who
responded comprised approximately 7
percent of this population. After stratifying
these respondents on the basis of age, gender
[ 24 ]
and school location (metropolitan vs.
country), 50 participants were randomly
selected to match the distribution of all VPPA
members and hopefully all principals. This
method was adopted in an attempt to make the
intervention group representative of the
entire principal class in the DoE on the basis
of the demographic variables available.
As the program was largely funded by the
VPPA and the DoE, the decision was made
that participation in the research program
would be obligatory for all participants in
the HP program. It was recognised that this
could dissuade some employees from
participating in the HP program, but it was
necessary to maximise participant numbers
and therefore gain maximum benefit from
available funding. This was made explicit
in all communications with VPPA
members.
The demographic characteristics of the
participants (PRIN) and the full DoE
population of principal class employees
(S. Stephen, DoE, personal communication)
are given in Table I. The PRIN group differed
from the DoE group only on the basis of
seniority ( 2 (2, n = 50) = 22.0, p < 0.001). This
was despite the fact that approximately 33
percent of all VPPA members are assistant
principals. The PRIN group represented
principals, rather than the broader principal
class employee population which includes
assistant principals.
All subjects completed an informed
consent form and the project had approval
from the La Trobe University Human Ethics
Committee.
Measures
Participant characteristics were assessed by
self-report questionnaire and by physical
assessment. All questionnaires were numbercoded to ensure anonymity. They were
distributed to the PRIN group at the launch
of the associated HP program and returned
via pre-paid reply envelopes.
The questionnaire has been described in
detail elsewhere (Green, 1997), but, for
comparative purposes, questions relating to
health-related behaviour were drawn from
those used in previous health surveys of the
Australian population by the National Heart
Foundation (NHF) (National Heart
Foundation of Australia, 1991). The questions
covered the following four areas: smoking
history, exercise patterns in the previous two
weeks, alcohol consumption and dietary
patterns. In addition the Stress Arousal
Checklist was included as a reliable, two
dimensional measure of self-reported mood
(Gotts and Cox, 1990).
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
The physical assessment for PRIN subjects
was conducted over one and a half hours at
the premises of HBA Health Management. It
included a full medical examination, 12-lead
exercise electrocardiogram, blood pathology,
and a number of physical measurements.
Subjects were given counselling on the day
by medical staff and exercise physiologists
regarding their diet and exercise patterns.
Further services provided as part of the HP
program have been described elsewhere
(Green et al., 2000).
As indicated previously, during the same
period that this study was conducted, a
screening program was offered to all
principal class employees by the DoE. This
program was conducted by MBF (as opposed
to HBA) Health Management using a mobile
testing unit. This program tested 1,507
employees (61 percent of all principal class
employees) and the data were provided to the
author by the DoE for comparative purposes
(O'Halloran, 1997). The number of subjects
who may have been involved in both
assessments is unknown.
Data analysis
The physical health and health-related
behaviour of the principals in this study
(PRIN group) were compared with those of
three other groups. The first of these was the
MBF screening data to ensure that they were
representative of the entire principal class.
Second, they were compared with
appropriate age-based data for the Victorian
population extracted from the most recent
National Heart Foundation survey (National
Heart Foundation & Australian Institute of
Health, 1995). Third, since the socioeconomic
status of principals is high relative to the
general population, they were compared with
Table I
Demographics of the participants (PRIN) and the entire DoE population of
principal class officers (MSD, unless specified otherwise)
Variable
PRIN (n = 50)
DoE (n = 1,758)
62
66
47.5 4.6
N/Ab
Age category (% under 44)
26
31
Age category (% over 49)
36
36
Seniority** (% assistant principals)
6
38
Duration employed (years)
27.7 5.4
N/Ab
Duration employed as principal (years)
6.5 5.8
N/Ab
Gender (%) male
Age (years)
28
33
Students enrolled at schoola (n = 47)
296 145
232
Salaried staff employed at schoola (n = 47)
19.9 8.9
Location at school (% rural)
a
16.1
b
Notes: Only mean data available from DoE, so no comparison possible; Not available;
**PRIN group differs from DoE, p < 0.01
45-49 year-old corporate clients tested at HBA
Health Management in the same calendar
year. Although the education and income
profiles of the latter group are unknown, the
HBA database is of similar socioeconomic
status to the principals as it comprises,
almost exclusively, white-collar employees in
middle and senior management (Green et al.,
1995).
In comparing the PRIN, MBF and HBA
groups, variables measured on interval or
ratio scales of measurement were analysed
using t-tests. Non-parametric data were
analysed between groups using tests for
independent samples (e.g. Chi-square,
Kruskal-Wallis). All chi-square tests
involving only one degree of freedom were
calculated using Yates' continuity
correction.
All data were inspected to detect problems
in their distributions which may have
affected the validity of subsequent statistical
analyses. Categorical variables with cells
containing fewer than five cases were
collapsed. The distribution of continuous
variables was examined. Variables which
deviated significantly from a normal
distribution were transformed (square root,
reciprocal or logarithm) to minimize
skewness prior to subsequent analysis.
Skewness was non-significant in 96 percent of
transformed variables after transformation.
Where the transformation was not sufficient
to reduce skewness to acceptable levels, the
least skewed transformation was chosen.
Transformed variables were used in all
subsequent analyses.
Results
Demographic characteristics of the PRIN
group in contrast to the general population
(NHF data) are presented in Table II. The
higher proportion of males employed full
time in the PRIN group approached
significance ( 2 = 3.5, p = 0.06). The
combination of 100 percent full-time
employment and higher education and
income levels of the PRIN group confirmed
their high socioeconomic status in contrast
with the general population.
Health-related behaviour and physical
and psychological health characteristics of
the PRIN and comparison groups are
presented separately for males and females
in Tables III-VI. Although the trend for PRIN
group males to be less likely to smoke than
the general population (NHF, Table III) did
not achieve significance after adjustment for
continuity correction ( 2 = 2.5, p = 0.11), this
may represent an artifact of small subject
[ 25 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
Table II
Demographic characteristics of PRIN group in contrast to population normative values for
45-49 year old males and females
The International Journal of
Educational Management
15/1 [2001] 23±30
Variable
Birthplace (% Australia)
PRIN
(n = 33)
Male
NHF
(n = 41)
88
58
sigd
PRIN
(n = 17)
Female
NHF
(n = 39)
sig4
2 6:4*
100
49
2 = 11.4**
Education (% post-secondary
educateda)
100
34
= 31.4**
100
20
2 = 27.1**
Employed (% full-time)
100
85
2 = 3.5, ns
100
54
2 = 9.5**
0
97
2 = 46.2**
Income (% less than $51,680 (male)
or $50,760 (female)p.a.)b,c
0
85
2
2
= 49.1**
Notes: aThis was defined as ``University, CAE or other tertiary institution'', for the NHF study, and includes ``trade
certificates'' for PRIN group; bCut-off for top NHF income category (1995 AUS$) adjusted for changes in average
weekly earnings (Australian Bureau of Statistics); cThis figure was the total remuneration package for PRIN
subjects, which includes employer superannuation contributions. However, the gross salary of even the lowest
paid subjects was still approximately equal to the top NHF category cut-off (depending on employer contributions
to superannuation); dSignificance of difference between groups: * p < 0.05; ** p < 0.01; ns = not significant
Source: National Heart Foundation in Australia, 1991
Table III
Health-related behaviour characteristics of male principals (PRIN) and three comparison groups
(NHF, HBA and MBF)a
PRIN
(n = 33)
NHF
(n = 41)
Current smoker (%)
6
22
8
5
Alcohol consumptionb (% intermediate
or greater risk)
3
16
N/Ac
N/A
2 = 1.9, ns
Usual diet (% modified in some way)
58
67
N/A
N/A
2 = 1.6, ns
Recreational exercise (% participating
in vigorous exercise in the last two
weeks)
45
24
N/A
N/A
2 = 2.7, ns
Recreational exercise (% participating
in walking exercise in the last two
weeks)
58
54
N/A
N/A
2 = 0.0, ns
No recreational exercise (% none in
last two weeks)
12
32
17
17
2 6:9; ns
Variable
HBA
MBF
(n = 317) (n = 1,090)
sigd
2 = 24.7**, NHF >
HBA, MBF, HBA > MBF
Notes: aSee text for details of comparison groups; bHealth risk of alcohol consumption. Classified as none, low,
intermediate, high or very high; cNot available; dSignificance of difference between groups: * p < 0.05; ** p < 0.01;
ns = not significant
numbers as the NHF group were more likely
to smoke than the larger survey of
principals (MBF) and the white-collar
comparison group (HBA). A similar trend
was evident in the contrast of smoking
patterns for females in the PRIN and NHF
groups after continuity correction ( 2 = 3.4,
p = 0.06, Table IV). In all other health-related
behaviours, there was no apparent
difference between the PRIN group and any
of the comparison populations for males or
females.
There were no differences in height, weight
or measures of obesity between the PRIN
group and any of the comparison populations
[ 26 ]
for either males or females (Tables V and VI).
Principals, represented by the PRIN, MBF or
sometimes both groups had worse scores on
blood pressure (systolic for males only), HDL
cholesterol, cholesterol ratio, cardiac risk
score and both psychological variables
(arousal for males only) than the white-collar
comparison group (HBA, Tables V and VI).
Although the larger survey of principals
(MBF) had a lower total cholesterol than the
HBA group, their HDL cholesterol was not
measured to allow comparison of the
cholesterol ratio. Principals did not differ
from the general population on any physical
characteristics. The cardiac risk score for the
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
Table IV
Health-related behaviour characteristics of female principals (PRIN) and three comparison
groups (NHF, HBA and MBF)a
The International Journal of
Educational Management
15/1 [2001] 23±30
Current smoker (%)
PRIN
(n = 17)
NHF
(n = 39)
HBA
(n = 26)
MBF
(n = 417)
6
33
8
5
Alcohol consumptionb (% intermediate
or greater risk)
12
10
N/Ac
N/A
2 = 0.0, ns
Usual diet (% modified in some way)
29
38
N/A
N/A
2 = 0.1, ns
Recreational exercise (% participating
in vigorous exercise in the last two
weeks)
18
18
N/A
N/A
2 = 0.0, ns
Recreational exercise (% participating
in walking exercise in the last two
weeks)
71
59
N/A
N/A
2 = 0.3, ns
No recreational exercise (% none in
last two weeks)
24
36
4
21
2 = 11.0,* NHF > HBA
Variable
sigd
2 = 41.7**, NHF >
HBA, MBF
Notes: aSee text for details of comparison groups; bHealth risk of alcohol consumption. Classified as none, low,
intermediate, high or very high; cNot available; dSignificance of difference between groups: * p < 0.05;
** p < 0.01; ns = not significant
PRIN group was positively correlated with
stress (r = 0.39, p < 0.01).
The psychological characteristics of the
PRIN and HBA groups are presented
diagrammatically in Figure 1. The axes on
the Stress Arousal Checklist (SACL)
quadrant are centred on the mean for a
full-time employed Australian population,
and the description of the mood state for
each quadrant as described by the authors
of the questionnaire (Cox and Griffiths,
1995). Figure 1 shows the coordinates of the
subjects with males and females pooled
since there was no difference between
sexes. There was no relationship between
stress and either age or experience.
Table V
Physical and psychological characteristics of male principals (PRIN) and three comparison groups (NHF, HBA and MBF)a
Variable
PRIN
(n = 33)
NHF
(n = 41)
HBA
(n = 317)
MBF
(n = 1,090)
sigf
Age (years)
47.7 4.5
47.9 1.5
46.9 1.4
47.6 5.0
F = 1.98, ns
Height (cm)
175.3 6.3
173.8 7.5
177.1 11.9
N/A
F = 0.78, ns
Weight (kg)
82.6 10.7
81.0 10.9
83.4 12.3
N/A
F = 0.37, ns
Body mass index (kg/mb)
26.5 3.0
26.5 3.7
26.4 3.4
N/A
F = 0.00, ns
Waist-to-hip ratio
F = 1.00, ns
0.93 0.05
0.92 0.08
0.92 0.06
N/A
Systolic blood pressure (mm Hg)
133 14
134 16
127 14
128 13
F = 5.3**, HBA < PRIN, NHF
Diastolic blood pressure (mm Hg)
81 9
85 10
80 10
84 9
F = 24.9**, HBA < NHF, MBF
Cholesterol *mmol/1)
5.77 0.94
5.53 1.09
5.85 1.01
5.59 1.04
F = 5.6**, MBF < HBA
HDL cholesterolb (mmol/1)
1.06 0.37
1.12 0.26
1.22 0.32
N/A
F = 4.3*, PRIN < HBA
Cholesterol ratioc
5.99 1.99
5.15 1.52
5.19 1.74
N/A
F = 3.2*, HBA < PRIN
Cardiac risk scored
19.4 8.9
N/A
15.9 8.5
N/A
t = 23*
Stresse
7.4 5.2
N/A
4.5 4.7
N/A
t = 3.3**
Arousale
7.7 3.5
N/A
9.1 3.0
N/A
t = 2.6*
a
b
Notes: See text for details of comparison groups; High density lipoprotein cholesterol. The fraction of cholesterol which serves a preventive role
in terms of cardiovascular disease; cCholesterol ratio = cholesterol/HDL cholesterol. Higher results indicate increased risk of cardiovascular
disease; dCardiac risk score (arbitrary units). Calculated on the basis of known risk factors measured during the assessment. Risk classification:
0-8 = ``low''; 9-16 = ``below average''; 17-23 = ``average''; 24-31 = ``above average''; > 31 = ``high''; eStress arousal checklist (SACL) typical scale
(Mackay et al., 1978). Response to 30 scale items, resulting in a score from 0-12 for arousal and 0-18 for stress; fSignificance of difference
between groups: * p < 0.05; ** p < 0.01; ns = not significant
[ 27 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
Discussion
As indicated above, the low proportion of
assistant principals in the PRIN group means
that the results presented here reflect
principal health rather than the health of all
principal class employees. Sixty-one percent
of all principal class employees participated
in the MBF survey (O'Halloran, 1997), but the
classification levels of this group were
unknown. However, as the mean age of the
PRIN and MBF health survey groups is
identical for males and females, the MBF
survey may also have been underrepresented in terms of assistant principals.
Although there were no differences in the
health related behaviour, physical or
psychological health characteristics of the
PRIN and MBF groups, the lower total
cholesterol of the MBF males in contrast with
the HBA males (in the absence of any
difference for the PRIN group) may be due to
the methodology adopted in the two surveys.
It has been suggested previously that
cholesterol measured in a whole blood
sample (as used in the MBF survey) is likely
to be up to 10 percent lower than that
measured from serum samples (as used in the
other three groups) (Kingery et al., 1994). The
lack of difference between these two groups
on any parameters may relate to the small
sample numbers in the PRIN group, but
given the relatively high representation of
principal class employees in the MBF survey
(61 percent) it would appear that the results
of this survey accurately represent the health
of Victorian principals.
The high socioeconomic status of the PRIN
group was indicated by their education and
income levels compared with the NHF group
and was consistent with the high proportion
of the PRIN group born in Australia. High
socioeconomic status is typically associated
with better health status and better healthrelated behaviour (Australian Institute of
Health, 1990; Marmot, 1994). The
socioeconomic status of the PRIN group was
consistent with lower smoking rates but no
other health-related behaviour (although the
comparative measure of diet used here is
very limited). However, the health status of
Victorian principals was no better than that
of the average population, as represented by
the NHF group and appears to be worse than
that of a comparative group of similar
socioeconomic status (HBA). The low
incidence of smoking and the need for better
exercise and dietary behaviour are in
agreement with previous surveys of
principal health (Smith et al., 1988). However,
the comparatively poor health status of
principals in this study contrasts with
previous self-reports of above average health
status (Smith et al., 1988). The current
findings may be related to the high stress
levels reported, as previous work has found
Table VI
Physical and psychological characteristics of female principals (PRIN) and three comparison groups (NHF, HBA and MBF)a
PRIN
(n = 17)
Variable
NHF
(n = 39)
HBA
(n = 26)
MBF
(n = 417)
sigf
Age (years)
47.1 5.0
47.6 1.5
46.7 1.3
47.2 5.1
F = 0.17, ns
Height (cm)
163.0 6.4
162.5 6.3
163.8 6.2
N/A
F = 0.07, ns
F = 0.93, ns
Weight (kg)
69.3 11.8
73.0 13.2
64.2 13.7
N/A
Body mass index (kg/mb)
25.8 5.4
27.1 4.4
23.8 4.2
N/A
F = 0.83, ns
Waist-to-hip ratio
0.79 0.05
0.79 0.07
0.77 0.06
N/A
F = 0.59, ns
Systolic blood pressure (mm Hg)
131 17
127 16
121 12
123 14
Diastolic blood pressure (mm Hg)
83 14
80 9
73 9
80 9
Cholesterol *mmol/1)
5.78 1.02
5.26 1.05
5.22 0.74
5.37 10.96
HDL cholesterolb (mmol/1)
1.33 0.44
1.40 0.26
1.62 0.40
N/A
F = 4.2*, PRIN < HBA
4.75 1.68
3.88 1.11
3.43 1.10
N/A
F = 5.6*, HBA < PRIN
17.4 7.6
N/A
11.1 7.3
N/A
t = 27*
Stress
7.5 5.2
N/A
3.0 4.2
N/A
t = 3.1**
Arousale
8.8 3.0
N/A
7.5 3.8
N/A
t = 1.2, ns
Cholesterol ratio
c
Cardiac risk scored
e
a
b
F = 2.6, ns
F = 5.4**, HBA < NHF, MBF, PRIN
F = 1.4, ns
Notes: See text for details of comparison groups; High density lipoprotein cholesterol. The fraction of cholesterol which serves a preventive role
in terms of cardiovascular disease; cCholesterol ratio = cholesterol/HDL cholesterol. Higher results indicate increased risk of cardiovascular
disease; dCardiac risk score (arbitrary units). Calculated on the basis of known risk factors measured during the assessment. Risk classification:
0-8 = ``low''; 9-16 = ``below average''; 17-23 = ``average''; 24-31 = ``above average''; > 31 = ``high''; eStress arousal checklist (SACL) typical
scale (Mackay et al., 1978). Response to 30 scale items, resulting in a score from 0-12 for arousal and 0-18 for stress; fSignificance of difference
between groups: * p < 0.05; ** p < 0.01; ns = not significant
[ 28 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
that higher stress levels are associated with
poor self-reported health status amongst
principals (Tung and Koch, 1980). The
relationship between stress levels and
cardiac risk score in the current study is
consistent with this.
Previous reports have suggested that
teachers have very high occupational stress
levels (Cox and Brockley, 1984). The high
stress scores for the principals in this study
may indicate similarly high levels amongst
principals and also reflect the change in role
of the principal under the organisational
changes inherent in the SoF program. Over
30 years ago, Kahn et al. clearly identified
role changes as significantly correlated with
such important emotional and behavioural
reactions as less job satisfaction, less
confidence in the larger organisation and less
interaction with other role senders in the
organisation (Kahn et al., 1964). The current
findings are consistent with the changed role
of the principal with the devolution of
responsibilities from an Education
Department to individual self-managing
schools (Cranston, 1994). Although in this
study only a single measure of stress was
measured, it may be expected that the SoF
changes would increase the role-related
stress and also boundary-spanning stress
which was previously in the domain of the
Education department (Tung and Koch, 1980).
There has been a shift in recruitment policies
for principals in that merit and ability now
take precedence over seniority and
experience (Cranston, 1994). Such changes
may mean that a ``new breed'' of principals
Figure 1
Typical mood stress arousal checklist coordinates for PRIN and HBA
groups (males and females pooled)
capable of working under the changed school
conditions is being employed. However,
recent international evidence suggests that
many primary school principals received
very little training for their new role prior to
commencing as a principal (Dunning, 1996).
Assuming that adequate training is provided,
the problems reported in this study may
diminish with time. However, the lack of
relationship between stress and either age or
years of experience as a principal does not
support this proposition. This analysis may
be somewhat simplistic because if a
multidimensional measure of stress is
conducted it is apparent that, while taskrelated stress may decrease with age and
experience, boundary-spanning stress may
increase (Tung and Koch, 1980). Regardless of
the precise nature and extent of these
temporal correlates of stress, for those
principals still adjusting to change assistance
in coping with the new expectations may be
required.
Conclusion
Principals in Victorian primary schools
reported lower smoking prevalence than the
typical Victorian population but were no
better than the general population on a
number of other health-related behaviours.
The lower smoking prevalence is consistent
with their high socioeconomic status relative
to the remainder of the population but,
perhaps because of the lack of other
improvements in health-related behaviour,
was not reflected in better health status.
Their cardiovascular health status was worse
than a group of similar socioeconomic status
and may be related to the higher stress levels
reported by the principals in contrast to a
group of similar socioeconomic status. The
high stress levels may have been associated
with the introduction of the SoF selfgovernance programs introduced prior to
this health survey. Regular surveys of the
physical and psychological health status of
principals are required so as to better
understand the relationship of occupational
stress to key indices of healthy living in this
important group of educational leaders.
References and further reading
Australian Institute of Health (1990), Australia's
Health, Australian Institute of Health.
Caldwell, B. (1992), ``The principal as leader of the
self-managing school in Australia'', Journal of
Educational Administration, Vol. 30, pp. 6-19.
Cooper, B., Sieverding, J. and Muth, R. (1988),
``Principals' management behaviour,
personality types and psychological stress'',
[ 29 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
[ 30 ]
The Journal of Educational Administration,
Vol. 26, pp. 197-219.
Cooper, C.L. and Marshall, J. (1976),
``Occupational sources of stress: a review of
the literature relating to coronary heart
disease and mental health'', Journal of
Occupational Psychology, Vol. 49, pp. 11-28.
Cox, T. and Brockley, T. (1984), ``The experience
and effects of stress in teachers'', British
Educational Research Journal, Vol. 10, pp. 83-7.
Cox, T. and Griffiths, A. (1995), ``The nature and
measurement of work stress: theory and
practice'', in Wilson, J.R. and Corlett, E.N.
(Eds), Evaluation of Human Work, Taylor &
Francis, Washington, DC, pp. 783-803.
Cranston, N. (1994), ``Strategic management in
schools: a new role for the principal'', Asia
Pacific Journal of Human Resources, Vol. 32,
pp. 100-9.
Dunning, G. (1996), ``Preparing for primary school
management: training providers, task
preparation and support ± perceptions of new
teachers'', International Journal of
Educational Management, Vol. 10, pp. 33-43.
Gotts, G. and Cox, T. (1990), Stress and Arousal
Checklist. A Manual for its Administration,
Scoring and Interpretation, Swinburne Press,
Melbourne.
Green, R., Lang, J. and Hatcher, D. (1995),
``Fitness testing'', in Zuluaga, M., Briggs, C.,
Carlisle, J., McDonald, V., McMeeken, J.,
Nickson, W., Oddy, P. and Wilson, D. (Eds),
Sports Physiotherapy: Applied Science and
Practice, Churchill Livingstone, Melbourne,
pp. 161-79.
Green, R., Malcolm, S., Greenwood, K. and
Murphy, G. (2001), ``Impact of a health
promotion program on the health of primary
school principals'', International Journal of
Educational Management, Vol. 15 No. 1,
pp. 31-8.
Green, R.A. (1997), Benefits of Worksite Health
Promotion in Australia, PhD Thesis, La Trobe
University, Melbourne, p. 500.
Kahn, R., Wolfe, D., Quinn, R. and Shoek, J. (1964),
Organisational Stress: Studies in Role Conflict
and Ambiguity, Wiley, New York, NY.
Kingery, P.M., Ellsworth, C.G., Corbett, B.S.,
Bowden, R.G. and Brizzolara, J.A. (1994),
``High-cost analysis: a closer look at the case
for work-site health promotion'', Journal of
Occupational Medicine, Vol. 36, pp. 1341-7.
Kyriacou, C. (1987), ``Teacher stress and burnout:
an international review'', Educational
Research, Vol. 29, pp. 146-52.
Mackay, C.J., Cox, T., Burrows, G.C. and
Lazzerini, A.J. (1978), ``An inventory for the
measurement of self-reported stress and
arousal'', British Journal of Social and
Clinical Psychology, Vol. 17, pp. 283-4.
Marmot, M. (1994), ``Work and other factors
influencing coronary health and sickness
absence'', Work & Stress, Vol. 8, pp. 191-201.
National Heart Foundation & Australian Institute
of Health (1995), ``The risk factor prevalence
survey, 1989'', Social Sciences Data Archives,
Australian National University, Canberra.
National Heart Foundation of Australia (1991),
Risk Factor Prevalence Study No. 3, 1989: Cities
Analysis, National Heart Foundation of
Australia & Australian Institute of Health,
Canberra.
O'Halloran, P. (1997), MBF Cardiovascular
Testing of Principal Class Officers,
Directorate of School Education (Victoria).
Savery, L. and Detiuk, M. (1986), ``The perceived
stress levels of primary and secondary school
principals'', The Journal of Educational
Administration, Vol. 24, pp. 272-81.
Smith, D.W., Bibeau, D.L., Altschuld, J.W. and
Heit, P. (1988), ``Health-related characteristics
of selected school principals'', Journal of
School Health, Vol. 58, pp. 397-400.
Thornton, P. (1996), ``The physiological,
psychological and work stress of primary
school principals'', International Journal of
Educational Management, Vol. 10, pp. 42-55.
Tung, R.L. and Koch, J.L. (1980), ``School
administrators: sources of stress and ways of
coping with it'', in Cooper, C.L. and
Marshall, J. (Eds), White-Collar and
Professional Stress, John Wiley & Sons,
Chichester, pp. 63-91.
University of Melbourne Applied Psychology
Research Group (1990), Teacher Stress in
Victoria, a Survey of Teachers' Views: The
Rreport, Ministry of Education, Victoria,
Melbourne.
principals
Rod Green
Senior Lecturer, La Trobe University, Bundoora, Australia
Susan Malcolm
Senior Lecturer, La Trobe University, Bundoora, Australia
Ken Greenwood
La Trobe University, Bundoora, Australia
Michael Small
Victorian Primary Principals Association, North Melbourne, Australia
Gregory Murphy
La Trobe University, Bundoora, Australia
Keywords
Teachers, Education, Schools,
Surveys
Abstract
In recent years responsibility for
the administration of schools
internationally has shifted from
education departments towards
self-governing schools. This trend
has resulted in major changes to
the role of school principals. Such
changes in role may impact on the
psychological and physical health
of principals, but there has been
very little research into this
population. A survey of the health
and wellbeing of a representative
sample of 50 principals of State
primary schools in Victoria,
Australia is reported. Subjects
completed questionnaires
measuring health-related
behaviour and stress and arousal
levels and participated in
comprehensive health appraisals.
Principals reported better smoking
patterns than the population as a
whole. Despite a higher
socioeconomic status than the
population as a whole, the health
status of the principals was not
apparently better. Principals
reported higher stress levels and
worse physical health than a group
of white-collar employees of
similar socioeconomic status.
This is the first of two
related articles taken from
the authors' recent research
into the health of primary
school principals in Victoria,
Australia
The International Journal of
Educational Management
15/1 [2001] 23±30
# MCB University Press
[ISSN 0951-354X]
Introduction
There is a large body of literature
internationally which has documented the
high prevalence of stress and burnout in the
teaching profession (Kyriacou, 1987). The
majority of this research relates to classroom
teachers while comparatively few papers
report on stress amongst school principals
(Tung and Koch, 1980; Cooper et al., 1988;
Savery and Detiuk, 1986). One study which
has examined stressors amongst school
principals and other school administrators
indicated that the major stressors for
principals related to the following four
dimensions in decreasing order of
importance: role-based stress, task-based
stress, conflict-mediating stress and
boundary-spanning stress (Tung and Koch,
1980). Principals scored higher than regionbased administrators on all of these
dimensions except boundary-spanning,
which relates to the interaction between the
school and the external environment. A
survey of Western Australian principals
found that primary principals had higher
stress levels than secondary principals
(Savery and Detiuk, 1986). Another study
attempted to correlate continuously
monitored heart rate as an indicator of
physiological stress to principal behaviour,
and found variations in heart rate related to
personality type (Cooper et al., 1988).
Despite concerns about the relationship
between occupational stress and stressrelated illness (Cooper and Marshall, 1976;
Kyriacou, 1987; Savery and Detiuk, 1986),
there has been far less work conducted on the
general health characteristics of teachers
and, again, even less relating to the health
status of principals (Smith et al., 1988). One
report using self-report of health
characteristics on the multidimensional
The current issue and full text archive of this journal is available at
http://www.emerald-library.com/ft
General Wellbeing Questionnaire suggested
that principals had above average health
status (Smith et al., 1988). This study reported
that principals had low drug (tobacco,
alcohol) usage, but some health behaviours,
particularly diet and exercise needed to be
improved. Another study involving selfreport of various stress-related illnesses
found a higher incidence of stress-related
illness, particularly amongst primary
principals, than the general population
(Savery and Detiuk, 1986).
There is evidence that prolonged
occupational stress can lead to both mental
and physical ill-health (Kyriacou, 1987;
Cooper and Marshall, 1976), and a study of
school administrators revealed a strong
inverse relationship between stress levels
and self-reported physical health status
(Tung and Koch, 1980). Concern about high
stress levels amongst teachers in part relates
to the cost of workers' compensation and
potentially the loss of staff through
associated ill-health retirement (Kyriacou,
1987). Such concerns about consequent
workers' compensation claims led to the
instigation of a government initiated report
in Victoria, Australia (University of
Melbourne Applied Psychology Research
Group, 1990). This report included 47 specific
recommendations to reduce teacher stress;
however one general comment of interest in
the recommendations section was that ``the
Ministry (of Education) should slow the
perceived rate of change, and so reduce the
extent to which it creates uncertainty in the
school environment'' (University of
Melbourne Applied Psychology Research
Group, 1990).
Well-documented changes in public
education, both internationally and in
Australia, over the last decade have involved
devolution of financial, staffing and planning
processes from Departments of Education to
individual schools (Cranston, 1994; Caldwell,
1992). The introduction in 1993 of the Schools
of The Future (SoF) program in government
[ 23 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
schools in the State of Victoria has been
typical of this type of change. Such changes
have modified the traditional role of the
principal to a great extent such that much
more managerial, as opposed to the previous
administrative, skills are required (Cranston,
1994). It has been suggested that this
increasing responsibility for principals may
lead to stress-related illness (Thornton, 1996).
In light of the potential impact of these factors
on principal stress levels and principal health,
and the lack of existing data on the health and
health-related behaviour characteristics of
school principals, this study was undertaken
to ascertain the health status of principals
working under the SoF system.
Method
This study was initiated by the Victorian
Primary Principals Association (VPPA) out
of concern for the overall health and
wellbeing of its members associated with the
introduction of SoF. The VPPA's concern
related to rapidly changing work practices
and lifestyles reported by principals in order
to cope with the increased workload,
accountability and responsibility inherent in
the SoF reforms.
Although the DoE was initiating a free statewide screening program for principals, the
VPPA felt that this program was inadequate to
assist principals in maintaining their health
and wellbeing in the face of increasing workrelated demands. Hence this project involved
two stages; a survey of principal health, the
results of which are presented in this paper,
and the introduction of a comprehensive
health promotion (HP) program to improve
principal health. The outcomes of the HP
program are presented elsewhere (Green et
al., 2000).
The VPPA, the DoE, HBA Health
Management (the provider of the health
promotion program), and principals in the
program made equal contributions to the cost
of the program. The initial health survey,
which was the first phase of the HP program,
took place in March 1996.
Participants
Approximately 1,450 VPPA members were
contacted and asked to give an expression of
interest in participating in a HP program
conducted by an external provider which
included a heavily subsidised comprehensive
health assessment as a component of a
broader HP program. The 99 members who
responded comprised approximately 7
percent of this population. After stratifying
these respondents on the basis of age, gender
[ 24 ]
and school location (metropolitan vs.
country), 50 participants were randomly
selected to match the distribution of all VPPA
members and hopefully all principals. This
method was adopted in an attempt to make the
intervention group representative of the
entire principal class in the DoE on the basis
of the demographic variables available.
As the program was largely funded by the
VPPA and the DoE, the decision was made
that participation in the research program
would be obligatory for all participants in
the HP program. It was recognised that this
could dissuade some employees from
participating in the HP program, but it was
necessary to maximise participant numbers
and therefore gain maximum benefit from
available funding. This was made explicit
in all communications with VPPA
members.
The demographic characteristics of the
participants (PRIN) and the full DoE
population of principal class employees
(S. Stephen, DoE, personal communication)
are given in Table I. The PRIN group differed
from the DoE group only on the basis of
seniority ( 2 (2, n = 50) = 22.0, p < 0.001). This
was despite the fact that approximately 33
percent of all VPPA members are assistant
principals. The PRIN group represented
principals, rather than the broader principal
class employee population which includes
assistant principals.
All subjects completed an informed
consent form and the project had approval
from the La Trobe University Human Ethics
Committee.
Measures
Participant characteristics were assessed by
self-report questionnaire and by physical
assessment. All questionnaires were numbercoded to ensure anonymity. They were
distributed to the PRIN group at the launch
of the associated HP program and returned
via pre-paid reply envelopes.
The questionnaire has been described in
detail elsewhere (Green, 1997), but, for
comparative purposes, questions relating to
health-related behaviour were drawn from
those used in previous health surveys of the
Australian population by the National Heart
Foundation (NHF) (National Heart
Foundation of Australia, 1991). The questions
covered the following four areas: smoking
history, exercise patterns in the previous two
weeks, alcohol consumption and dietary
patterns. In addition the Stress Arousal
Checklist was included as a reliable, two
dimensional measure of self-reported mood
(Gotts and Cox, 1990).
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
The physical assessment for PRIN subjects
was conducted over one and a half hours at
the premises of HBA Health Management. It
included a full medical examination, 12-lead
exercise electrocardiogram, blood pathology,
and a number of physical measurements.
Subjects were given counselling on the day
by medical staff and exercise physiologists
regarding their diet and exercise patterns.
Further services provided as part of the HP
program have been described elsewhere
(Green et al., 2000).
As indicated previously, during the same
period that this study was conducted, a
screening program was offered to all
principal class employees by the DoE. This
program was conducted by MBF (as opposed
to HBA) Health Management using a mobile
testing unit. This program tested 1,507
employees (61 percent of all principal class
employees) and the data were provided to the
author by the DoE for comparative purposes
(O'Halloran, 1997). The number of subjects
who may have been involved in both
assessments is unknown.
Data analysis
The physical health and health-related
behaviour of the principals in this study
(PRIN group) were compared with those of
three other groups. The first of these was the
MBF screening data to ensure that they were
representative of the entire principal class.
Second, they were compared with
appropriate age-based data for the Victorian
population extracted from the most recent
National Heart Foundation survey (National
Heart Foundation & Australian Institute of
Health, 1995). Third, since the socioeconomic
status of principals is high relative to the
general population, they were compared with
Table I
Demographics of the participants (PRIN) and the entire DoE population of
principal class officers (MSD, unless specified otherwise)
Variable
PRIN (n = 50)
DoE (n = 1,758)
62
66
47.5 4.6
N/Ab
Age category (% under 44)
26
31
Age category (% over 49)
36
36
Seniority** (% assistant principals)
6
38
Duration employed (years)
27.7 5.4
N/Ab
Duration employed as principal (years)
6.5 5.8
N/Ab
Gender (%) male
Age (years)
28
33
Students enrolled at schoola (n = 47)
296 145
232
Salaried staff employed at schoola (n = 47)
19.9 8.9
Location at school (% rural)
a
16.1
b
Notes: Only mean data available from DoE, so no comparison possible; Not available;
**PRIN group differs from DoE, p < 0.01
45-49 year-old corporate clients tested at HBA
Health Management in the same calendar
year. Although the education and income
profiles of the latter group are unknown, the
HBA database is of similar socioeconomic
status to the principals as it comprises,
almost exclusively, white-collar employees in
middle and senior management (Green et al.,
1995).
In comparing the PRIN, MBF and HBA
groups, variables measured on interval or
ratio scales of measurement were analysed
using t-tests. Non-parametric data were
analysed between groups using tests for
independent samples (e.g. Chi-square,
Kruskal-Wallis). All chi-square tests
involving only one degree of freedom were
calculated using Yates' continuity
correction.
All data were inspected to detect problems
in their distributions which may have
affected the validity of subsequent statistical
analyses. Categorical variables with cells
containing fewer than five cases were
collapsed. The distribution of continuous
variables was examined. Variables which
deviated significantly from a normal
distribution were transformed (square root,
reciprocal or logarithm) to minimize
skewness prior to subsequent analysis.
Skewness was non-significant in 96 percent of
transformed variables after transformation.
Where the transformation was not sufficient
to reduce skewness to acceptable levels, the
least skewed transformation was chosen.
Transformed variables were used in all
subsequent analyses.
Results
Demographic characteristics of the PRIN
group in contrast to the general population
(NHF data) are presented in Table II. The
higher proportion of males employed full
time in the PRIN group approached
significance ( 2 = 3.5, p = 0.06). The
combination of 100 percent full-time
employment and higher education and
income levels of the PRIN group confirmed
their high socioeconomic status in contrast
with the general population.
Health-related behaviour and physical
and psychological health characteristics of
the PRIN and comparison groups are
presented separately for males and females
in Tables III-VI. Although the trend for PRIN
group males to be less likely to smoke than
the general population (NHF, Table III) did
not achieve significance after adjustment for
continuity correction ( 2 = 2.5, p = 0.11), this
may represent an artifact of small subject
[ 25 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
Table II
Demographic characteristics of PRIN group in contrast to population normative values for
45-49 year old males and females
The International Journal of
Educational Management
15/1 [2001] 23±30
Variable
Birthplace (% Australia)
PRIN
(n = 33)
Male
NHF
(n = 41)
88
58
sigd
PRIN
(n = 17)
Female
NHF
(n = 39)
sig4
2 6:4*
100
49
2 = 11.4**
Education (% post-secondary
educateda)
100
34
= 31.4**
100
20
2 = 27.1**
Employed (% full-time)
100
85
2 = 3.5, ns
100
54
2 = 9.5**
0
97
2 = 46.2**
Income (% less than $51,680 (male)
or $50,760 (female)p.a.)b,c
0
85
2
2
= 49.1**
Notes: aThis was defined as ``University, CAE or other tertiary institution'', for the NHF study, and includes ``trade
certificates'' for PRIN group; bCut-off for top NHF income category (1995 AUS$) adjusted for changes in average
weekly earnings (Australian Bureau of Statistics); cThis figure was the total remuneration package for PRIN
subjects, which includes employer superannuation contributions. However, the gross salary of even the lowest
paid subjects was still approximately equal to the top NHF category cut-off (depending on employer contributions
to superannuation); dSignificance of difference between groups: * p < 0.05; ** p < 0.01; ns = not significant
Source: National Heart Foundation in Australia, 1991
Table III
Health-related behaviour characteristics of male principals (PRIN) and three comparison groups
(NHF, HBA and MBF)a
PRIN
(n = 33)
NHF
(n = 41)
Current smoker (%)
6
22
8
5
Alcohol consumptionb (% intermediate
or greater risk)
3
16
N/Ac
N/A
2 = 1.9, ns
Usual diet (% modified in some way)
58
67
N/A
N/A
2 = 1.6, ns
Recreational exercise (% participating
in vigorous exercise in the last two
weeks)
45
24
N/A
N/A
2 = 2.7, ns
Recreational exercise (% participating
in walking exercise in the last two
weeks)
58
54
N/A
N/A
2 = 0.0, ns
No recreational exercise (% none in
last two weeks)
12
32
17
17
2 6:9; ns
Variable
HBA
MBF
(n = 317) (n = 1,090)
sigd
2 = 24.7**, NHF >
HBA, MBF, HBA > MBF
Notes: aSee text for details of comparison groups; bHealth risk of alcohol consumption. Classified as none, low,
intermediate, high or very high; cNot available; dSignificance of difference between groups: * p < 0.05; ** p < 0.01;
ns = not significant
numbers as the NHF group were more likely
to smoke than the larger survey of
principals (MBF) and the white-collar
comparison group (HBA). A similar trend
was evident in the contrast of smoking
patterns for females in the PRIN and NHF
groups after continuity correction ( 2 = 3.4,
p = 0.06, Table IV). In all other health-related
behaviours, there was no apparent
difference between the PRIN group and any
of the comparison populations for males or
females.
There were no differences in height, weight
or measures of obesity between the PRIN
group and any of the comparison populations
[ 26 ]
for either males or females (Tables V and VI).
Principals, represented by the PRIN, MBF or
sometimes both groups had worse scores on
blood pressure (systolic for males only), HDL
cholesterol, cholesterol ratio, cardiac risk
score and both psychological variables
(arousal for males only) than the white-collar
comparison group (HBA, Tables V and VI).
Although the larger survey of principals
(MBF) had a lower total cholesterol than the
HBA group, their HDL cholesterol was not
measured to allow comparison of the
cholesterol ratio. Principals did not differ
from the general population on any physical
characteristics. The cardiac risk score for the
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
Table IV
Health-related behaviour characteristics of female principals (PRIN) and three comparison
groups (NHF, HBA and MBF)a
The International Journal of
Educational Management
15/1 [2001] 23±30
Current smoker (%)
PRIN
(n = 17)
NHF
(n = 39)
HBA
(n = 26)
MBF
(n = 417)
6
33
8
5
Alcohol consumptionb (% intermediate
or greater risk)
12
10
N/Ac
N/A
2 = 0.0, ns
Usual diet (% modified in some way)
29
38
N/A
N/A
2 = 0.1, ns
Recreational exercise (% participating
in vigorous exercise in the last two
weeks)
18
18
N/A
N/A
2 = 0.0, ns
Recreational exercise (% participating
in walking exercise in the last two
weeks)
71
59
N/A
N/A
2 = 0.3, ns
No recreational exercise (% none in
last two weeks)
24
36
4
21
2 = 11.0,* NHF > HBA
Variable
sigd
2 = 41.7**, NHF >
HBA, MBF
Notes: aSee text for details of comparison groups; bHealth risk of alcohol consumption. Classified as none, low,
intermediate, high or very high; cNot available; dSignificance of difference between groups: * p < 0.05;
** p < 0.01; ns = not significant
PRIN group was positively correlated with
stress (r = 0.39, p < 0.01).
The psychological characteristics of the
PRIN and HBA groups are presented
diagrammatically in Figure 1. The axes on
the Stress Arousal Checklist (SACL)
quadrant are centred on the mean for a
full-time employed Australian population,
and the description of the mood state for
each quadrant as described by the authors
of the questionnaire (Cox and Griffiths,
1995). Figure 1 shows the coordinates of the
subjects with males and females pooled
since there was no difference between
sexes. There was no relationship between
stress and either age or experience.
Table V
Physical and psychological characteristics of male principals (PRIN) and three comparison groups (NHF, HBA and MBF)a
Variable
PRIN
(n = 33)
NHF
(n = 41)
HBA
(n = 317)
MBF
(n = 1,090)
sigf
Age (years)
47.7 4.5
47.9 1.5
46.9 1.4
47.6 5.0
F = 1.98, ns
Height (cm)
175.3 6.3
173.8 7.5
177.1 11.9
N/A
F = 0.78, ns
Weight (kg)
82.6 10.7
81.0 10.9
83.4 12.3
N/A
F = 0.37, ns
Body mass index (kg/mb)
26.5 3.0
26.5 3.7
26.4 3.4
N/A
F = 0.00, ns
Waist-to-hip ratio
F = 1.00, ns
0.93 0.05
0.92 0.08
0.92 0.06
N/A
Systolic blood pressure (mm Hg)
133 14
134 16
127 14
128 13
F = 5.3**, HBA < PRIN, NHF
Diastolic blood pressure (mm Hg)
81 9
85 10
80 10
84 9
F = 24.9**, HBA < NHF, MBF
Cholesterol *mmol/1)
5.77 0.94
5.53 1.09
5.85 1.01
5.59 1.04
F = 5.6**, MBF < HBA
HDL cholesterolb (mmol/1)
1.06 0.37
1.12 0.26
1.22 0.32
N/A
F = 4.3*, PRIN < HBA
Cholesterol ratioc
5.99 1.99
5.15 1.52
5.19 1.74
N/A
F = 3.2*, HBA < PRIN
Cardiac risk scored
19.4 8.9
N/A
15.9 8.5
N/A
t = 23*
Stresse
7.4 5.2
N/A
4.5 4.7
N/A
t = 3.3**
Arousale
7.7 3.5
N/A
9.1 3.0
N/A
t = 2.6*
a
b
Notes: See text for details of comparison groups; High density lipoprotein cholesterol. The fraction of cholesterol which serves a preventive role
in terms of cardiovascular disease; cCholesterol ratio = cholesterol/HDL cholesterol. Higher results indicate increased risk of cardiovascular
disease; dCardiac risk score (arbitrary units). Calculated on the basis of known risk factors measured during the assessment. Risk classification:
0-8 = ``low''; 9-16 = ``below average''; 17-23 = ``average''; 24-31 = ``above average''; > 31 = ``high''; eStress arousal checklist (SACL) typical scale
(Mackay et al., 1978). Response to 30 scale items, resulting in a score from 0-12 for arousal and 0-18 for stress; fSignificance of difference
between groups: * p < 0.05; ** p < 0.01; ns = not significant
[ 27 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
Discussion
As indicated above, the low proportion of
assistant principals in the PRIN group means
that the results presented here reflect
principal health rather than the health of all
principal class employees. Sixty-one percent
of all principal class employees participated
in the MBF survey (O'Halloran, 1997), but the
classification levels of this group were
unknown. However, as the mean age of the
PRIN and MBF health survey groups is
identical for males and females, the MBF
survey may also have been underrepresented in terms of assistant principals.
Although there were no differences in the
health related behaviour, physical or
psychological health characteristics of the
PRIN and MBF groups, the lower total
cholesterol of the MBF males in contrast with
the HBA males (in the absence of any
difference for the PRIN group) may be due to
the methodology adopted in the two surveys.
It has been suggested previously that
cholesterol measured in a whole blood
sample (as used in the MBF survey) is likely
to be up to 10 percent lower than that
measured from serum samples (as used in the
other three groups) (Kingery et al., 1994). The
lack of difference between these two groups
on any parameters may relate to the small
sample numbers in the PRIN group, but
given the relatively high representation of
principal class employees in the MBF survey
(61 percent) it would appear that the results
of this survey accurately represent the health
of Victorian principals.
The high socioeconomic status of the PRIN
group was indicated by their education and
income levels compared with the NHF group
and was consistent with the high proportion
of the PRIN group born in Australia. High
socioeconomic status is typically associated
with better health status and better healthrelated behaviour (Australian Institute of
Health, 1990; Marmot, 1994). The
socioeconomic status of the PRIN group was
consistent with lower smoking rates but no
other health-related behaviour (although the
comparative measure of diet used here is
very limited). However, the health status of
Victorian principals was no better than that
of the average population, as represented by
the NHF group and appears to be worse than
that of a comparative group of similar
socioeconomic status (HBA). The low
incidence of smoking and the need for better
exercise and dietary behaviour are in
agreement with previous surveys of
principal health (Smith et al., 1988). However,
the comparatively poor health status of
principals in this study contrasts with
previous self-reports of above average health
status (Smith et al., 1988). The current
findings may be related to the high stress
levels reported, as previous work has found
Table VI
Physical and psychological characteristics of female principals (PRIN) and three comparison groups (NHF, HBA and MBF)a
PRIN
(n = 17)
Variable
NHF
(n = 39)
HBA
(n = 26)
MBF
(n = 417)
sigf
Age (years)
47.1 5.0
47.6 1.5
46.7 1.3
47.2 5.1
F = 0.17, ns
Height (cm)
163.0 6.4
162.5 6.3
163.8 6.2
N/A
F = 0.07, ns
F = 0.93, ns
Weight (kg)
69.3 11.8
73.0 13.2
64.2 13.7
N/A
Body mass index (kg/mb)
25.8 5.4
27.1 4.4
23.8 4.2
N/A
F = 0.83, ns
Waist-to-hip ratio
0.79 0.05
0.79 0.07
0.77 0.06
N/A
F = 0.59, ns
Systolic blood pressure (mm Hg)
131 17
127 16
121 12
123 14
Diastolic blood pressure (mm Hg)
83 14
80 9
73 9
80 9
Cholesterol *mmol/1)
5.78 1.02
5.26 1.05
5.22 0.74
5.37 10.96
HDL cholesterolb (mmol/1)
1.33 0.44
1.40 0.26
1.62 0.40
N/A
F = 4.2*, PRIN < HBA
4.75 1.68
3.88 1.11
3.43 1.10
N/A
F = 5.6*, HBA < PRIN
17.4 7.6
N/A
11.1 7.3
N/A
t = 27*
Stress
7.5 5.2
N/A
3.0 4.2
N/A
t = 3.1**
Arousale
8.8 3.0
N/A
7.5 3.8
N/A
t = 1.2, ns
Cholesterol ratio
c
Cardiac risk scored
e
a
b
F = 2.6, ns
F = 5.4**, HBA < NHF, MBF, PRIN
F = 1.4, ns
Notes: See text for details of comparison groups; High density lipoprotein cholesterol. The fraction of cholesterol which serves a preventive role
in terms of cardiovascular disease; cCholesterol ratio = cholesterol/HDL cholesterol. Higher results indicate increased risk of cardiovascular
disease; dCardiac risk score (arbitrary units). Calculated on the basis of known risk factors measured during the assessment. Risk classification:
0-8 = ``low''; 9-16 = ``below average''; 17-23 = ``average''; 24-31 = ``above average''; > 31 = ``high''; eStress arousal checklist (SACL) typical
scale (Mackay et al., 1978). Response to 30 scale items, resulting in a score from 0-12 for arousal and 0-18 for stress; fSignificance of difference
between groups: * p < 0.05; ** p < 0.01; ns = not significant
[ 28 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
that higher stress levels are associated with
poor self-reported health status amongst
principals (Tung and Koch, 1980). The
relationship between stress levels and
cardiac risk score in the current study is
consistent with this.
Previous reports have suggested that
teachers have very high occupational stress
levels (Cox and Brockley, 1984). The high
stress scores for the principals in this study
may indicate similarly high levels amongst
principals and also reflect the change in role
of the principal under the organisational
changes inherent in the SoF program. Over
30 years ago, Kahn et al. clearly identified
role changes as significantly correlated with
such important emotional and behavioural
reactions as less job satisfaction, less
confidence in the larger organisation and less
interaction with other role senders in the
organisation (Kahn et al., 1964). The current
findings are consistent with the changed role
of the principal with the devolution of
responsibilities from an Education
Department to individual self-managing
schools (Cranston, 1994). Although in this
study only a single measure of stress was
measured, it may be expected that the SoF
changes would increase the role-related
stress and also boundary-spanning stress
which was previously in the domain of the
Education department (Tung and Koch, 1980).
There has been a shift in recruitment policies
for principals in that merit and ability now
take precedence over seniority and
experience (Cranston, 1994). Such changes
may mean that a ``new breed'' of principals
Figure 1
Typical mood stress arousal checklist coordinates for PRIN and HBA
groups (males and females pooled)
capable of working under the changed school
conditions is being employed. However,
recent international evidence suggests that
many primary school principals received
very little training for their new role prior to
commencing as a principal (Dunning, 1996).
Assuming that adequate training is provided,
the problems reported in this study may
diminish with time. However, the lack of
relationship between stress and either age or
years of experience as a principal does not
support this proposition. This analysis may
be somewhat simplistic because if a
multidimensional measure of stress is
conducted it is apparent that, while taskrelated stress may decrease with age and
experience, boundary-spanning stress may
increase (Tung and Koch, 1980). Regardless of
the precise nature and extent of these
temporal correlates of stress, for those
principals still adjusting to change assistance
in coping with the new expectations may be
required.
Conclusion
Principals in Victorian primary schools
reported lower smoking prevalence than the
typical Victorian population but were no
better than the general population on a
number of other health-related behaviours.
The lower smoking prevalence is consistent
with their high socioeconomic status relative
to the remainder of the population but,
perhaps because of the lack of other
improvements in health-related behaviour,
was not reflected in better health status.
Their cardiovascular health status was worse
than a group of similar socioeconomic status
and may be related to the higher stress levels
reported by the principals in contrast to a
group of similar socioeconomic status. The
high stress levels may have been associated
with the introduction of the SoF selfgovernance programs introduced prior to
this health survey. Regular surveys of the
physical and psychological health status of
principals are required so as to better
understand the relationship of occupational
stress to key indices of healthy living in this
important group of educational leaders.
References and further reading
Australian Institute of Health (1990), Australia's
Health, Australian Institute of Health.
Caldwell, B. (1992), ``The principal as leader of the
self-managing school in Australia'', Journal of
Educational Administration, Vol. 30, pp. 6-19.
Cooper, B., Sieverding, J. and Muth, R. (1988),
``Principals' management behaviour,
personality types and psychological stress'',
[ 29 ]
Rod Green, Susan Malcolm,
Ken Greenwood,
Michael Small and
Gregory Murphy
A survey of the health of
Victorian primary school
principals
The International Journal of
Educational Management
15/1 [2001] 23±30
[ 30 ]
The Journal of Educational Administration,
Vol. 26, pp. 197-219.
Cooper, C.L. and Marshall, J. (1976),
``Occupational sources of stress: a review of
the literature relating to coronary heart
disease and mental health'', Journal of
Occupational Psychology, Vol. 49, pp. 11-28.
Cox, T. and Brockley, T. (1984), ``The experience
and effects of stress in teachers'', British
Educational Research Journal, Vol. 10, pp. 83-7.
Cox, T. and Griffiths, A. (1995), ``The nature and
measurement of work stress: theory and
practice'', in Wilson, J.R. and Corlett, E.N.
(Eds), Evaluation of Human Work, Taylor &
Francis, Washington, DC, pp. 783-803.
Cranston, N. (1994), ``Strategic management in
schools: a new role for the principal'', Asia
Pacific Journal of Human Resources, Vol. 32,
pp. 100-9.
Dunning, G. (1996), ``Preparing for primary school
management: training providers, task
preparation and support ± perceptions of new
teachers'', International Journal of
Educational Management, Vol. 10, pp. 33-43.
Gotts, G. and Cox, T. (1990), Stress and Arousal
Checklist. A Manual for its Administration,
Scoring and Interpretation, Swinburne Press,
Melbourne.
Green, R., Lang, J. and Hatcher, D. (1995),
``Fitness testing'', in Zuluaga, M., Briggs, C.,
Carlisle, J., McDonald, V., McMeeken, J.,
Nickson, W., Oddy, P. and Wilson, D. (Eds),
Sports Physiotherapy: Applied Science and
Practice, Churchill Livingstone, Melbourne,
pp. 161-79.
Green, R., Malcolm, S., Greenwood, K. and
Murphy, G. (2001), ``Impact of a health
promotion program on the health of primary
school principals'', International Journal of
Educational Management, Vol. 15 No. 1,
pp. 31-8.
Green, R.A. (1997), Benefits of Worksite Health
Promotion in Australia, PhD Thesis, La Trobe
University, Melbourne, p. 500.
Kahn, R., Wolfe, D., Quinn, R. and Shoek, J. (1964),
Organisational Stress: Studies in Role Conflict
and Ambiguity, Wiley, New York, NY.
Kingery, P.M., Ellsworth, C.G., Corbett, B.S.,
Bowden, R.G. and Brizzolara, J.A. (1994),
``High-cost analysis: a closer look at the case
for work-site health promotion'', Journal of
Occupational Medicine, Vol. 36, pp. 1341-7.
Kyriacou, C. (1987), ``Teacher stress and burnout:
an international review'', Educational
Research, Vol. 29, pp. 146-52.
Mackay, C.J., Cox, T., Burrows, G.C. and
Lazzerini, A.J. (1978), ``An inventory for the
measurement of self-reported stress and
arousal'', British Journal of Social and
Clinical Psychology, Vol. 17, pp. 283-4.
Marmot, M. (1994), ``Work and other factors
influencing coronary health and sickness
absence'', Work & Stress, Vol. 8, pp. 191-201.
National Heart Foundation & Australian Institute
of Health (1995), ``The risk factor prevalence
survey, 1989'', Social Sciences Data Archives,
Australian National University, Canberra.
National Heart Foundation of Australia (1991),
Risk Factor Prevalence Study No. 3, 1989: Cities
Analysis, National Heart Foundation of
Australia & Australian Institute of Health,
Canberra.
O'Halloran, P. (1997), MBF Cardiovascular
Testing of Principal Class Officers,
Directorate of School Education (Victoria).
Savery, L. and Detiuk, M. (1986), ``The perceived
stress levels of primary and secondary school
principals'', The Journal of Educational
Administration, Vol. 24, pp. 272-81.
Smith, D.W., Bibeau, D.L., Altschuld, J.W. and
Heit, P. (1988), ``Health-related characteristics
of selected school principals'', Journal of
School Health, Vol. 58, pp. 397-400.
Thornton, P. (1996), ``The physiological,
psychological and work stress of primary
school principals'', International Journal of
Educational Management, Vol. 10, pp. 42-55.
Tung, R.L. and Koch, J.L. (1980), ``School
administrators: sources of stress and ways of
coping with it'', in Cooper, C.L. and
Marshall, J. (Eds), White-Collar and
Professional Stress, John Wiley & Sons,
Chichester, pp. 63-91.
University of Melbourne Applied Psychology
Research Group (1990), Teacher Stress in
Victoria, a Survey of Teachers' Views: The
Rreport, Ministry of Education, Victoria,
Melbourne.