J Occup Health 2011; 53: 23–35
Journal of
Occupational Health
Field Study
A Brief Cognitive-behavioral Stress Management Program for
Secondary School Teachers
Sharron SK Leung1, Vico CL Chiang2, YY Chui3, YW Mak2 and Daniel FK Wong4
School of Nursing, Li Ka Shing Faculty of Medicine, The University of Hong Kong, 2School of Nursing, The Hong
Kong Polytechnic University, 3Nursing School, Hong Kong Baptist Hospital and 4Department of Applied Social
Studies, City University of Hong Kong, China
1
Abstract: A Brief Cognitive-behavioral Stress
Management Program for Secondary School
Teachers: Sharron SK Leung, et al. School of
Nursing, Li Ka Shing Faculty of Medicine, The
University of Hong Kong, China—Objective: This
study aimed to assess the efficacy of a brief cognitivebehavioral program that was designed to reduce the
work-related stress levels of secondary school teachers.
Methods: A quasi-experimental design was used to
compare the intervention groups with the wait-list control
groups. Seventy teachers from the intervention groups
and 54 from the control groups completed a set of
validated scales at the baseline and 3–4 wk later. The
scales included the Depression Anxiety Stress Scale,
the Dysfunctional Attitude Scale—Form A, the HealthPromoting Lifestyle Profile II, and the Occupational
Stress Inventory Revised Edition. Results: After
controlling for the baseline measures, the intervention
groups had significantly lower role stress, personal
strain and overall work-related stress 3–4 wk after the
baseline measurements. The intervention groups also
had significantly higher stress management behaviors,
and less general stress and dysfunctional thoughts than
the control groups (all p≤0.05). The levels of dysfunctional
thoughts and stress management behaviors significantly
predicted general stress after intervention and personal
resource deficits. The level of dysfunctional thoughts
also predicted the personal strain of work-related
stresses (all p<0.05). Conclusions: The brief program
reported in this study was efficacious in reducing the
work-related stress of secondary teachers. Teachers
experienced less work-related stress after the program,
and they reported reduced dysfunctional thoughts and
Received Mar 15, 2010; Accepted Oct 4, 2010
Published online in J-STAGE Nov 10, 2010
Correspondence to: S. SK Leung, School of Nursing, Li Ka Shing
Faculty of Medicine, The University of Hong Kong, 4/F, William
MW Mong Block, 21 Sassoon Road, Hong Kong, PRC.
(e-mail: [email protected])
enhanced stress management behaviors. This program
may be considered as an initial strategy for teachers to
develop skills to cope with their work-related stress in
the short term and could be incorporated with other
strategies to achieve longer-term effects.
(J Occup Health 2011; 53: 23–35)
Key words: Cognitive behavior therapy, Intervention
studies, Occupational health, Psychological, School
teachers, Stress
High levels of work-related stress in secondary school
teachers have negative consequences for individuals and
for society. Work-related stress is defined as unpleasant
and negative emotions resulting from some aspects of
work1). This type of stress leads not only to a loss of
productivity to society but can also lead to mental health
problems, such as depression and anxiety2–4). In cases of
chronic and extended stress, burnout can be a long-term
consequence5). Negative emotional responses in stressful
situations, such as anxiety, frustrations and depression,
have also been shown to be strongly associated with stress
and burnout in teachers4). A recent study reported that
38.6% of secondary school teachers in Hong Kong
experienced strong maladaptive stress due to vocational
strain, 32.5% experienced moderate to extremely severe
levels of depression, and that teachers had limited
resources for coping6). There is an urgent need for
effective stress management program for teachers to
promote their psychological and physical health and to
prevent losses in productivity. This study tested a brief
cognitive-behavioral (CB) stress management program,
which was designed for Chinese secondary school
teachers.
Among the different intervention approaches for
managing work-related stress, interventions that use CB
techniques have been shown to be the most effective7).
CB interventions specifically deal with negative emotional
responses by correcting irrational cognition. A meta-
24
analysis performed by van der Klink et al.7) revealed that
CB interventions achieved the largest effect size (d=0.68)
compared to other interventions. Another meta-analysis
performed by Richardson and Rothstein8) yielded similar
results, and showed that CB interventions (d=1.164) and
alternative interventions (d=0.909) had the largest effect
sizes.
CB interventions aim to change cognition and
subsequently reinforce active coping skills and appear to
be effective at improving perceived quality of work life
and enhancing individuals’ psychological resources and
responses9). CB interventions encourage individuals to
proactively respond to stress by reducing their negative
thoughts9), and they have been shown to be effective in
reducing daily hassles and psychological distress10), and a
variety of psychiatric problems, such as mood and anxiety
disorders11). These conditions also seem to be related to
work-related stress9). However, there are relatively few
studies on the effectiveness of CB interventions for
teachers for occupational stress management. Richardson
and Rothstein8) reviewed 36 studies on occupational stress
management, but only four of these studies targeted
teachers and among them only three studies tested CB
interventions and used other multimodal interventions.
Bertoch et al.12) used a 12-week multimodal intervention
including deep breathing and social support without a
cognitive-behavioral element. They found significantly
lower stress among the 15 teachers in the intervention
group than in the control group. Among those three studies
which used the CB interventions, Tunnecliffe et al.13)
reported that a collaborative behavioral consultation
intervention with five weekly sessions for seven teachers
was superior at reducing teachers’ stress than relaxation
training, or no intervention for the control. This
intervention mainly focused on problem-solving skills
with consultant-modelled good practices and helped
teachers to plan and develop their own stress reduction
interventions. Two other studies compared stress
inoculation training that incorporated primarily the
cognitive-behavioral approach to other interventions and
control. One compared classroom management skills
training in four weeks14) and the other compared a coworker support intervention in six weeks15). Sharp and
Forum14) found that both the stress inoculation training
and classroom management skills training were more
effective at reducing general anxiety and teaching anxiety
for participants in an experimental group than in the
control (n=30 per group). Cecil and Forum15) found that
stress inoculation training (n=17), but not co-worker
support (n=17), was effective at reducing teachers’ selfreported stress level. Although Richardson and Rothstein
reviewed studies published through 2006, only these four
studies12–15) targeted teachers and all of them were reported
two decades ago, having taken place between 1985 and
1990. Only two studies14, 15) reported details of the use of
J Occup Health, Vol. 53, 2011
the cognitive-behavioral approach in an intervention, and
one14) assessed anxiety rather than stress as the outcome
measure. The sample sizes of these studies were relatively
small, ranging from 21 to 60 with only seven to 30
participants per experimental or control group, and none
of these four studies were conducted with Chinese
populations.
A recent study in China16) reported an effective stress
management program for teachers, which applied both
organizational and individual approaches. A CB approach
generally provides clear instructions and explanation,
making it well suited for the Chinese population. Previous
research has shown that distinctive CB processes, such as
teaching cognitive skills, providing information, directing
session activities, and emphasising homework, were
highly compatible with Chinese culture17). In particular,
the Chinese prefer this type of structured session with
practical advice and instructions directly provided by
experts, which is effective at driving specific actions.
Assignments and homework also fit well with Confucian
work ethics, which is task- and action-oriented and
characterised by order and discipline. The Chinese
particularly benefit from CB processes that challenged
their irrational cognitions, which are hidden and deeply
rooted in their strict adherence to social norms17, 18). The
Chinese are culturally very concerned about the views and
instructions of people with authority and seniority, and
they tend to deal with life problems under the influence of
their parents, elders, and teachers.
However, most previous programs including the one
developed in China16) require time and resources (such as
manpower). The Chinese program spanned a year,
required substantial resources, collaboration and the firm
mutual commitment of the organization and individual
teachers throughout the entire period. In a review of 46
studies, van der Klink7) observed an inverse relationship
between the number of sessions and effect size, but there
was no optimal number of sessions identified. Another
report provided evidence that full attendance of a brief CB
program with six 30-minute sessions was effective at
reducing psychological distress and increasing job
performance among the participants19). A CB program
with a smaller number of sessions might still be
effective.
In this study, we designed a brief three-session group
CB program with activities and assignments that
specifically aimed to challenge teachers’ irrational
thoughts and illogical thinking in relation to their teaching
and interactions with senior staff, peers, and students. We
hypothesized that this type of group CB approach would
be particularly effective among Chinese teachers for workrelated stress management. Furthermore, Chinese teachers
are regarded by society as highly distinctive, respectable
role models. It is therefore particularly difficult for
teachers to admit having emotional difficulties, which are
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
often perceived as bringing shame to their families and
themselves as individuals, since they hold a senior status
as teachers18). Using group dynamics, the brief CB group
program aimed to generate collective reflections from the
participating teachers that challenged irrational thoughts
and stresses founded on cultural traditions. The goal of
the program was to encourage the teachers to exercise
positive and appropriate thinking when dealing with
teaching related stress.
The present study took a universal approach by targeting
all teachers rather than only those with high stress levels
or mental health problems. This type of educationally
focused, non-therapeutic program allowed the participants
to acquire new knowledge and skills without being
“labelled” with any problems or disorders. This strategy
is particularly suitable for Chinese teachers as it allows
them to “save face”, and if they prefer, they do not need
to explicitly disclose any difficulties or stresses they are
facing. All of the participants could learn skills to help
themselves, their colleagues, and/or junior staff to cope
with work-related stress.
Our study extended the literature with more up-to-date
data and evidence of a brief stress management program,
which used a CB approach to manage the work-related
stress of Chinese secondary school teachers. The
objectives of this study were (1) to evaluate the efficacy
of a brief CB intervention in reducing work-related and
general stress levels and dysfunctional thoughts, and
promoting stress management behaviors, and (2) to predict
post-intervention work-related and general stress levels
by measuring the changes in stress management behaviors
and dysfunctional thoughts after the intervention.
We tested two hypotheses:
· Hypothesis 1: the intervention groups show significantly
greater changes in general stress, depression, anxiety,
work-related stress, stress management behaviors, and
dysfunctional thoughts compared with the control
groups.
· Hypothesis 2: the general and work-related stress
levels of the intervention groups after the intervention can
be predicted by the changes in stress management
behaviors and dysfunctional thoughts after the
intervention.
Subjects and Methods
Design
A quasi-experimental study design was used to examine
the efficacy of a brief CB program for stress management
among secondary school teachers. The study design
compared the intervention groups with the wait-list control
groups. All participants in both types of groups completed
a set of validated questionnaires twice. The participants
in the intervention groups completed the pre-intervention
questionnaires 2–3 wk before the intervention in midNovember 2007 and the post-questionnaire one week after
25
the intervention in mid-December 2007. The participants
in the control groups completed the same set of
questionnaires at two different times in late April and late
May 2008, which were four weeks apart, before they were
offered the program.
Setting and sample
The eligible participants were secondary school teachers
working in government-aided or direct subsidized secondary
schools in Hong Kong, who were able to read and speak
Chinese. A letter of invitation and information about the
program were faxed to all of the secondary schools on the
2005 list of Hong Kong secondary schools20), thus reaching
all eligible secondary school teachers. Promotional
pamphlets were posted on the website of the Professional
Teacher Union, and a press conference publicized the
program by reporting the findings of a pilot run of the
program. Enquiries were received by telephone from five
schools, and three schools agreed to participate. The other
two schools declined because the teachers’ schedules did
not match the offered program periods. All of the
secondary school teachers from the three participating
schools were invited to participate in the study, and 161
teachers agreed to participate. The intervention was
incorporated as a teachers’ professional development
program and was conducted on their Teacher Development
Days in the participating schools. The participants from
the three schools were allocated into groups by matching
their Teacher Development Days with the program periods
offered by the research team. Those available earlier were
assigned to intervention groups. The control groups were
offered the program about six months after the intervention
groups due to re-scheduling of the Teaching Development
Day by the school of the control groups. According to this
school, such rescheduling was necessary because of some
changes of other school events. With the changes, the
related participants were only available to participate in
the survey and attend the program in the second semester.
This delay was inevitable in order to match the availability
of the participants from this school. Participants in the
control groups took all the assessments before attending
the program. The intervention groups took the baseline
and post-intervention assessments during the first semester,
whereas the control groups took the assessments during
the second semester. Assessment periods for both groups
were near the end of the semesters which avoided special
school events and examination periods. For the same
reason, the Teacher Development Days were used to
conduct the program because they were usually set around
less busy school periods. Ethics approval was obtained
from the Internal Review Board of The University of Hong
Kong. The purpose, risks and benefits of the study were
explained to all who agreed to participate. Written consent
was obtained from each participant before joining the
program.
26
J Occup Health, Vol. 53, 2011
Table 1. Demographic characteristics of the intervention and control groups
Gender*
Male
Female
Age (yr)
<26
26–35
36–45
>45
Employment
Permanent
Contract
Part-time
Monthly income (HK$)
<20,001
20,001–40,000
>40,000
Intervention groups (N=70)
N
%
Control groups (N=54)
N
%
43
27
61.4
38.6
26
28
48.1
51.9
6
14
37
13
8.6
20
52.6
18.8
6
15
19
14
11
27.8
35.2
26
49
21
0
70
30
0
42
11
1
77.8
20.4
1.8
5
34
31
7.1
48.6
44.3
8
22
24
14.8
40.7
44.5
*Difference between intervention and control group, c2=10.21, p<0.001.
Of the 149 participants who returned the completed data
sheets on demographics, 63% were men and 36.9% were
women. More than a quarter of the participants (26.8%)
were between 36 and 40 yr old, and 33.6% were aged 36
yr or younger. Around half of the participants were single
(49%), and 45.6% were married. Most of the participants
had permanent employment (74.5%), and 24.8% of the
participants had a monthly household income of
HK$40,001 or above (equivalent to US$5,128). Only
14.1% had a monthly income below HK$20,001
(US$2,564). The age range of our sample was comparable
to those reported previously in Hong Kong statistics, and
the teachers’ median age was 36. There were comparatively
more females in the population, the male to female teacher
ratio was 45:5521).
Of the 149 participants, 124 completed the pre- and
post-intervention questionnaires. The questionnaires for
the remainder were discarded because significant
proportions of the data were missing. The final data set
for the analyses included 70 participants in the intervention
groups and 54 in the control groups. Table 1 shows the
demographic characteristics of the participants in the
intervention and control groups. There were no significant
differences between the two groups except for gender. The
intervention groups had significantly more male
participants than the control groups (χ 2 =10.21,
p<0.001).
CB Intervention
A brief three-session CB stress management program
was designed specifically to alleviate the work-related
stress of secondary school teachers. The program was
modified from a 10-session CB program that significantly
reduced depressive symptoms among adults in Hong
Kong22). The CB program was developed using Beck’s
cognitive model of depression23), and we hypothesized that
after the modifications, the program would be useful for
equipping teachers with the cognitive skills to deal with
their work-related stress. The aim of the program was to
enable the participants to use the CB approach23) to manage
their own emotions in stressful work-related situations that
are common among secondary school teachers in the
Chinese educational context. Mini-lectures and active
participation in group discussions were the key activities
in the program. Worksheets were designed to facilitate
the application of the model in stressful situations that the
participants had experienced in schools. The program
manual, teaching materials and worksheets were tested
previously on two groups of 20 secondary school
teachers24).
In the first session, using the worksheets, the participants
identified the discrepancies between their life priorities
and actual schedule of work and social activities. They
also identified the sources of work-related stress at their
schools. They were introduced to the CB model of Aaron
Beck23) and identified the activating event and the four
responses (physiological, emotional, behavioral and
cognitive) under stressful situations.
In the second session, the participants learned to identify
the automatic thoughts from their cognitive responses and
the five behavioral and cognitive strategies to correct
irrational thoughts22). That is, the participants learned to
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
be aware of their own stress signals, stop irrational
thoughts, reflect and self-assess the irrational thoughts,
replace them with positive and rational statements, and
engage in pleasurable activities. Stressful situations and
related irrational thoughts that are common in teaching
and school administration in Hong Kong were included
on the worksheets to stimulate reflection and discussions.
Participants were asked to apply these strategies to break
the cycle of irrational and automatic thoughts in their own
stressful situations.
In the third session, the participants discussed in groups
the core beliefs that lead to irrational thoughts, and
explored ways to relax those strongly held core beliefs.
Participants also made plans to engage in pleasurable
activities based on the level of enjoyment and level of
feasibility. Relaxation exercises were introduced and
practiced during the session.
The participants were arranged in groups with 10 to 20
participants per group, and each group was led by one
facilitator. The facilitators all had nursing backgrounds,
and they had all received training on running groups with
the CB approach.
Measures
The set of questionnaires included the Depression
Anxiety Stress Scale (DASS-21), the Health-Promoting
Lifestyle Profile II (HPLP-II), the Dysfunctional Attitude
Scales—Form A (DAS-A), and the demographic data
before the intervention. The same set of questionnaires
was used after the intervention with the addition of the
Occupational Stress Inventory Revised Edition (OSI-R).
Since the OSI-R is a long questionnaire, and the time lapse
between the first and second data collections was short, it
was only used after the intervention.
The DASS-21, a short version of the DASS25), has 21
items and was originally developed to measure the
negative emotional states of depression, anxiety and stress.
The instrument has good psychometric properties26) and
has been widely used27, 28). The stress subscale was used
as an indicator of the general stress level in this study.
Participants were asked to rate the degree to which
statements applied to them in the past week on a four-point
scale from 0 (does not apply to me at all) to 3 (applies to
me very much, or most of the time). The total score for
each scale ranges from 0 to 21, and higher scores indicate
a higher level of depression, anxiety and stress. In the
present study, Cronbach’s alpha was 0.95 for the whole
scale, and 0.89, 0.86, 0.89 for the depression, anxiety and
stress subscales, respectively.
The HPLP-II was used to measure stress management
behaviors. The scale measures six dimensions of selfreported health-promoting behaviors29) including health
responsibility, physical activity, nutrition, spiritual growth,
interpersonal relations and stress management. In this
study, we only used the stress management subscale, which
27
has eight items. Participants were asked to rate the
frequency of stress management behaviors using a fourpoint Likert scale as 1 (never), 2 (sometimes), 3 (often),
or 4 (routinely). The total score ranges from 8 to 32, and
higher scores represent more engagement in stress
management behaviors. For this scale, Cronbach’s alpha
was 0.75 for the participants.
The DAS-A was used to measure cognitive distortions,
particularly those related to depression. This 40-item scale
was developed based on Beck’s constructs of cognitive
dysfunction30). It has been translated and used in Chinese
populations in Taiwan and mainland China. The DAS-A
was validated in a Chinese community sample in Hong
Kong and had a reported Cronbach’s alpha of 0.95 and
split-half reliability of r=0.87 31). The participants
responded to the statements on a Likert scale that ranges
from 1 to 7. The total score ranges from 40 to 280, and
higher scores indicate more maladaptive beliefs and
cognitive distortions. In this study, the Cronbach’s alpha
was 0.89.
OSI-R32) was used to measure work-related stress. It
was revised from the Occupational Stress Inventory
(OSI)33). The OSI-R consisted of three subscales: the 60item Occupational Roles Questionnaire (ORQ), the 40item Personal Strain Questionnaire (PSQ), and the 40-item
Personal Resources Questionnaire (PRQ). The ORQ
measures stress from six areas relating to occupational
roles including role overload, role insufficiency, role
ambiguity, role boundary, responsibility and physical
environment. The PSQ measures four dimensions of
personal strain including vocational strain, psychological
strain, interpersonal strain and physical strain. The PRQ
measures deficits of four types of coping resources
including recreation, self-care, social support and rational/
cognitive. The OSI-R has good internal consistency, and
it has been widely used in 60 studies in various contexts34).
Participants were asked to rate how well the statement
fitted them on a scale from 1 (rarely or never true) to 5
(true most of the time). Higher scores indicate higher
work-related stress levels. In the present study, Cronbach’s
alpha was 0.91 for the whole scale and 0.87, 0.86 and 0.87
for ORQ, PSQ and PRQ, respectively.
Data analysis
A repeated measures analysis of variance (ANOVA)
model was used with the group assignment to examine the
differences between the intervention and control groups
after the intervention. Differences in general stress,
depression, anxiety, stress management behaviors and
dysfunctional thoughts were evaluated. Separate models
were tested for each of these variables, separately for males
and females first and then controlled for gender as a
covariate for testing of all participants. Gender was
controlled as a confounder due to the significant difference
in gender between the intervention and control groups.
28
Group-by-time interactions were used to test whether time
differed between the intervention and control groups and
effectively controlled for any differences at baseline
between the groups. Analysis of covariance (ANCOVA)
was used to test the effect of the intervention on workrelated stress, which was measured after the intervention.
Separate models were tested for each of the three
dimensions of work-related stress (ORQ, PSQ, PRQ) and
the overall work-related stress score. To control for
differences at the baseline, the baseline measures were
entered as covariates in each model. Analysis was done
separately for males and females and then controlled for
gender as a covariate for testing of all participants.
Cohen’s d was used to represent the effect sizes of the
intervention when comparing the intervention groups with
the control groups. It was the standardized measure of the
intervention effect and calculated as the difference between
the means of the intervention and control groups divided
by a pooled standard deviation35). According to Cohen,
the effect sizes of 0.20, 0.5, and 0.8 are small, medium,
and large, respectively35).
Linear multiple regressions were used to identify the
predictors of general and work-related stress levels at the
follow-up from the changes in stress management and
dysfunctional thoughts in the intervention groups.
Changes in the study variances were calculated by
subtracting the T1 score from the T2 score. Separate
models were run for general and work-related stress
including the subscales and the whole scale. Each model
was controlled for baseline measures including general
stress level, stress management behaviors and dysfunctional
thoughts.
Results
Changes in stress, stress management behaviors and
dysfunctional thoughts
Table 2 shows the severity of the level of stress,
depression, anxiety and work-related stress in the
intervention and control groups. Six percent or less of
participants in the intervention groups, and no participants
in the control groups, had severe or extremely severe levels
of depression or anxiety. Strong maladaptive stress mainly
arose from role ambiguity and vocational strain. Regarding
personal resources, there was a significant lack of rational/
cognitive coping in both the intervention and control
groups.
Table 3 shows the gender-specific as well as the group
differences in general stress management behaviors and
dysfunctional thoughts between the intervention and
control groups at follow-up. The intervention groups had
significantly higher general stress, depression and anxiety
levels at baseline than the control groups, but these
differences were controlled for statistically in the repeated
measures ANOVA. The intervention groups had
significantly lower general stress levels, higher stress
J Occup Health, Vol. 53, 2011
management behaviors and lower dysfunctional thoughts
than the control groups after controlling for the baseline
measures (all p<0.05).
When males and females were considered separately,
the differences in general stress were significant among
females but not males, and differences in depression were
only observed among males. The effects of the intervention
on males and females were similar, and there was a greater
effect size for females in terms of reducing anxiety and
dysfunctional thoughts and increasing stress management
behaviors. Table 4 compares the post-intervention workrelated stress in the intervention and control groups. The
intervention groups, both as a whole and considering only
the females, had lower role stress, personal strain and
overall work-related stress (all p<0.05) than the control
groups. Among males, the differences between the
intervention and control groups were significant only for
personal strain and overall occupational stress. Cohen’s
d ranged from 0.03 (whole group general stress) to 0.71
(dysfunctional thoughts among females), with the
exception of the value for stress management behaviors
among females (d=0.92). Aside from this one exception,
the values were regarded as small to medium effect sizes,
and the highest effect sizes were for dysfunctional thoughts
and stress management behaviors. These data largely
confirmed Hypothesis 1.
Prediction of stress levels at follow-up
Table 5 shows predictions of post-intervention general
and work-related stress levels based on the changes in
stress management behaviors and dysfunctional thoughts
in the intervention groups. The change in dysfunctional
thoughts was a significant predictor of post-intervention
general stress after controlling for the baseline stress level
of the whole group as well as for males (both p<0.01). On
the other hand, the change in stress management behaviors
significantly predicted only the post-intervention personal
resources deficits after controlling for the baseline stress
level both for the whole group and for males (p=0.007 and
p=0.025, respectively). No significant predictors were
found for females. These findings partly support
Hypothesis 2 in that the changes in stress management
behaviors after the intervention predicted personal
resources deficits, and the change in dysfunctional
thoughts predicted general stress. However, when males
and females were considered separately, the predictions
were significant only for males.
Discussion
This study provides evidence that a brief CB intervention
was effective at reducing overall work-related stress levels
in the short term in a community sample of secondary
school teachers. Our short stress management program
was efficacious in reducing general stress among females,
depression among males, and anxiety in both genders. It
5.59 (4.59)
5.24 (4.51)
Depression
Anxiety
4(5.7)
1(1.4)
0
Extremely
severe &
severe
4 (5.7)
4 (5.7)
5 (7.1)
Moderate
7 (10)
4 (5.7)
3 (4.3)
Mild
7(10)
31.5 (7.88)
29.3 (7.26)
30.3 (6.29)
29.9 (6.18)
29.5 (6.96)
Role overload
Role insufficiency
Role ambiguity
Role boundary
Responsibility
27.2 (6.89)
27.2 (6.89)
26.5 (7.03)
26.5 (6.89)
Vocational strain
Psychological strain
Interpersonal strain
Physical strain
7(10)
12(17.1)
11(15.7)
48(68.5)
38 (54.3)
35 (50)
37 (52.8)
11 (15.7)
24 (34.3)
20 (28.6)
20 (28.6)
7 (10)
1 (1.4)
3 (4.3)
2 (2.9)
4 (5.7)
1 (1.4)
1 (1.4)
0
1 (1.4)
2 (2.9)
≤39 (relative
absence of
stress)
Physical
27.2 (8.01)
18(25.7)
35 (50)
16 (22.9)
environment
T-scores of subscale of OSI-R: Personal Strain Questionnaire (PSQ)
23 (32.9)
14 (20.3)
47 (67.2)
40 (57.1)
40–59
(normal
range)
3 (4.3)
13 (18.6)
36 (51.4)
36 (51.4)
20 (28.6)
21 (30)
60–69 (mild
maladaptive
stress)
55 (78.6)
61 (87.1)
62 (88.6)
Normal
N (%)
53 (75.7)
1(1.4)
9(12.9)
20(28.6)
2(2.9)
≥70 (strong
maladaptive
stress)
Dimensions
T-scores of subscale of OSI-R: Occupational Role Questionnaire (ORQ)
8.37 (5.23)
Stress
Sub-scales
The Depression Anxiety Stress Scale (DASS)
N (%)
23.1 (5.23)
24.9 (4.74)
23.8 (6.12)
23.4 (5.65)
19.5 (7.71)
26.3 (5.57)
28.5 (5.25)
30.9 (5.31)
29.1 (3.96)
32.7 (7.42)
2.94 (3.02)
3.61 (3.74)
5.70 (4.08)
0
2 (3.7)
3 (5.6)
3 (5.6)
15 (27.8)
4 (7.4)
6 (11.1)
5 (9.3)
20 (37)
0
13 (24.1)
≥70 (strong
maladaptive
stress)
0
0
Extremely
severe &
severe
N (%)
Mean (SD)
N (%)
Mean (SD)
N (%)
Control groups (N=54)
Intervention groups (N=70)
9 (16.7)
23 (42.6)
17 (31.5)
18 (33.3)
10 (18.5)
45 (83.3)
20 (37)
17 (31.5)
14 (25.9)
8 (14.8)
60–69 (mild
maladaptive
stress)
1 (1.9)
1 (1.9)
1 (1.9)
Moderate
N (%)
43 (79.6)
27 (50)
34 (63)
20 (37)
39 (72.2)
3 (5.6)
29 (53.7)
17 (31.5)
40 (74.1)
33 (61.1)
40–59
(normal
range)
3 (5.6)
2 (3.7)
1 (1.9)
Mild
N (%)
0
1 (1.9)
0
0
1 (1.9)
0
0
0
0
0
≤39 (relative
absence of
stress)
50 (92.6)
51 (94.4)
52 (96.3)
Normal
N (%)
Table 2. Severity level on subscales of the Depression Anxiety Stress Scale as well as subscales and T-scores of the Occupational Stress Inventory-R (OSI-R) and maladaptive stress
level on these subscales at baseline
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
29
N (%)
N (%)
27.5 (7.43)
27.7 (8.12)
28.1 (7.84)
31.2 (7.66)
Recreation
Self-care
Social support
Rational/cognitive
coping
Dimensions
17 (24.3)
9 (12.9)
0
<30
(significant
t lack of
coping)
0
23 (32.9)
36 (51.4)
6 (8.6)
8 (11.4)
30–39
(mild
deficit)
N (%)
26 (36.1)
25 (35.7)
47 (67.1)
41 (58.5)
40–59
(average)
T-scores of subscale of OSI-R: Personal Resources Questionnaire (PRQ)
Mean (SD)
Intervention groups (N=70)
4 (5.7)
0
17 (24.3)
≥60
(increasingly
strong
coping)
21 (30)
N (%)
30.7 (6.59)
38.2 (6.75)
24.2 (4.89)
22.7 (5.62)
Mean (SD)
16 (29.6)
2 (3.7)
0
<30
(significant
t lack of
coping)
1 (1.9)
N (%)
Control groups (N=54)
16 (29.6)
9 (16.7)
13 (24.1)
13 (24.1)
30–39
(mild
deficit)
N (%)
20 (37)
40 (74.1)
40 (74.1)
2 (3.7)
3 (5.6)
1 (1.9)
2 (3.7)
≥60
(increasingly
strong coping)
40–59
(average)
38 (70.4)
N (%)
N (%)
Table 2. Severity level on subscales of the Depression Anxiety Stress Scale as well as subscales and T-scores of the Occupational Stress Inventory-R (OSI-R) and maladaptive stress
level on these subscales at baseline (continued)
30
J Occup Health, Vol. 53, 2011
31
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
Table 3. Descriptive statistics and comparison of intervention and control groups with regard to general stress, depression, anxiety,
stress management behaviors and dysfunctional thoughts at follow-up
Variables
Intervention Baseline
Mean (SD)
Follow-up
Mean (SD)
Males (n=54)
General stressa
8.20 (5.42)
6.52 (4.74)
Depression
5.67 (4.60)
4.54 (5.11)
Anxiety
5.04 (4.76)
5.24 (4.26)
Stress management
13.37 (5.06)
17.54 (3.16)
behaviorsb
Dysfunctional thoughtsc 163.06 (27.15)
158.41 (30.23)
Females (n=16)
General stressa
8.94 (4.70)
8.56 (4.77)
Depression
5.31 (4.67)
4.88 (3.61)
Anxiety
5.94 (3.59)
5.63 (4.50)
Stress management
13.69 (4.08)
16.38 (1.75)
behaviorsb
Dysfunctional thoughtsc 166.50 (17.24)
159.50 (20.18)
Controlled for gender Whole group (n=70)
General stressa
8.37 (5.23)
6.99 (4.79)
Depression
5.59 (4.59)
4.61 (4.79)
Anxiety
5.24 (4.51)
5.32 (4.29)
Stress management
13.44 (4.83)
17.27 (2.93)ø*
behaviorsb
Dysfunctional thoughtsc 163.84 (25.16)
158.66 (28.12)
Control
Baseline
Mean (SD)
Follow-up
Mean (SD)
Repeated Measures
Cohen’s d
Males (n=24)
5.38 (4.08)
7.00 (3.97)
3.21 (2.52)
2.63 (2.98)
3.17 (2.55)
4.21 (3.37)
18.88 (3.04)
19.63 (3.84)
F(1,77), (p)
2.24 (0.138)
8.13 (0.006)
4.63 (0.035)
25.22 (<0.001)
0.45
0.27
0.61
174.96 (15.05)
8.20 (0.005)
0.54
172.50 (19.24)
Females (n=30)
5.97 (4.13)
7.27 (4.19)
3.93 (4.51)
2.93 (3.39)
2.77 (3.39)
3.77 (4.26)
18.53 (3.04)
18.73 (3.37)
F(1,45), (p)
6.27 (0.016)
3.60 (0.064)
10.12 (0.003)
26.45 (<0.001)
174.80 (17.56)
8.31 (0.006)
0.71
Whole group (n=54)
5.70 (4.08)
7.15 (4.06)
3.61 (3.74)
2.80 (3.19)
2.94 (3.02)
3.96 (3.86)
18.69 (3.01)
19.13 (3.58)
F(1,123), (p)
7.00 (0.009)
11.75 (0.001)
13.06 (<0.001)
47.86 (<0.001)
0.03
0.43
0.33
0.58
174.87 (16.34)
14.99 (<0.001)
0.59
173.40 (18.95)
173 (18.90)
0.29
0.43
0.92
General stress, measured by DASS-21, repeated measures controlled for general stress at baseline and gender. bStress management
behaviors, measured by HPLP-II, repeated measures controlled for stress management behaviors at baseline and gender. cDysfunctional
thoughts, measured by DAS-A, repeated measures controlled for dysfunctional thoughts at baseline and gender. øSignificant within
group differences before and after the intervention.
a
Table 4. Descriptive statistics and comparison of intervention and control groups with regard to work-related stress at follow-up
Variables
Mean Role stressd
Personal straind
Personal resources deficitd
Overall occupational stressd
Males (n=54)
169.58
24.51
95.67
13.60
116.13
17.77
381.38
47.28
Males (n=24)
178.30
23.00
108.76
22.76
120.35
23.89
407.41
52.67
F(1,77), (p)
2.29 (0.134)
6.83 (0.011)
0.60 (0.440)
4.31 (0.04)
Role stressd
Personal straind
Personal resources deficitd
Overall occupational stressd
Females (n=16)
164.83
23.88
94.53
20.32
115.43
15.07
374.80 48.38
Females (n=30)
187.60
31.23
110.73
21.99
112.60
17.88
410.93
65.37
F(1,45), (p)
7.38 (0.009)
6.02 (0.018)
0.31 (0.579)
4.10 (0.042)
Controlled for gender Role stressd
Personal straind
Personal resources deficitd
Overall occupational stressd
Whole group (n=70)
166.94
24.05
95.04
17.51
115.74
16.17
377.72
47.56
Whole group (n=54)
180.32
25.06
109.18
22.44
118.67
22.82
408.17
55.17
F(1,123), (p)
8.54 (0.004)
12.86 (<0.001)
0.16 (0.688)
8.70 (0.004)
SD
Intervention
Mean Control
SD
ANCOVA
Cohen’s d
0.68
0.53
0.84
0.78
0.65
0.55
0.72
0.60
General stress, measured by DASS-21, repeated measures controlled for general stress at baseline and gender. Role stress, personal
strain, personal resources deficit and overall work-related stress, measured by OSI-R, ANCOVA controlled for stress level at baseline
and gender. øSignificant within group differences before and after the intervention.
a
d
a
a
p
–0.15
–0.17
>0.05
–0.30
–0.83
>0.05
0.30
0.98
>0.05
–0.24
0.28
>0.05
–0.37
0.93
>0.05
0.21, 0.52
Female (n=16)
–0.22
–0.85
>0.05
–0.15
–0.69
>0.05
0.14
0.73
>0.05
0.01
0.03
>0.05
0.07
0.34
>0.05
0.06, 0.60
Male (n=54)
p
–0.59
–0.24
0.26
–0.17
–0.68
0.31, 0.89
–0.26
–0.53
0.28
–0.39
0.55
0.27, 3.50**
b
>0.05
> 0.05
0.017
0.049
>0.05
0.002
>0.05
>0.05
>0.05
>0.05
>0.05
>0.05
0.01
>0.05
>0.05
0.004
p
a
Stress management behaviors, measured by HPLP-II. bDysfunctional thoughts, measured by DAS-A. cGeneral stress, measured by DASS-21. dPersonal strain, personal resources
deficits, measured by OSI-R. Occupational role stress and the overall occupational role stress scores were not predicted by any of the independent variables reported in this table.
**p<0.05; ***p<0.001. Change of stress predicted by change of DAS for the whole group (b=–0.29, t=–3.16, p=0.002) and for males (b=–0.32, t=–3.05, p=0.004).
1.59
–1.32
–2.46
2.01
–1.42
3.16
0.24, 3.36**
–0.70
–0.72
0.90
–0.21
–1.85
–1.15
–2.70
1.70
–1.72
3.05
t
General stress after intervention
–1.46
>0.05
0.18
2.27
0.027
–0.27
–0.34
>0.05
–0.41
0.21
>0.05
0.27
–2.81
0.007
–0.29
0.95
>0.05
0.48
0.18, 2.29*
1.09
1.26
>0.05
–0.31
–0.91
>0.05
0.001
0.004
>0.05
–1.30
1.61
>0.05
–0.15
-0.40
>0.05
0.28, 0.77
0.45
1.89
>0.05
0.11
0.54
>0.05
0.09
0.53
>0.05
–0.57
–2.32
0.025
0.25
1.30
>0.05
0.15, 1.68
Work-related stress-personal resources deficitsd
b
t
–0.05
–0.37
>0.05
0.17
–0.15
–0.66
>0.05
0.49
–0.18
–0.97
>0.05
–0.006
–0.20
–1.35
>0.05
0.03
–0.09
0.39
>0.05
–0.61
–0.09
–0.50
>0.05
0.15
0.08, 0.90
Whole group (n=70)
Gender
Stress management behaviorsa at baseline
Dysfunctional thoughtsb at baseline
General stressc at baseline
Change in stress management behaviorsa
Change in dysfunctional thoughtsb
R2, F (6,63)
Stress management behaviors at baseline
Dysfunctional thoughtsb at baseline
General stressc at baseline
Change in stress management behaviorsa
Change in dysfunctional thoughtsb
R2, F (5, 10)
Stress management behaviors at baseline
Dysfunctional thoughtsb at baseline
General stressc at baseline
Change in stress management behaviorsa
Change in dysfunctional thoughtsb
R2, F (5,48)
Work-related stress-personal straind
b
t
Table 5. Prediction of work-related and general stress levels by the change in stress management behaviors and dysfunctional thoughts after controlling for baseline measures
32
J Occup Health, Vol. 53, 2011
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
was also efficacious in reducing work-related stress, except
personal resources, and dysfunctional thoughts, as well as
enhancing stress management behaviors when comparing
the intervention and control groups. The effect size on
general stress was very small, while there was a medium
effect size on work-related personal strain. The effect size
was greatest for increasing stress management behaviors
among females. Nevertheless, the program was primarily
designed to address work-related stress. All the worksheets
were designed specifically to guide participants to review
their experiences in managing work-related stress at
schools. Thus, the program raised the awareness of the
participants and reduced the specific work-related stressors
more than lowering general stress. These findings extend
those reported by Shimazu et al.36) for 24 teachers in Japan,
in which a five-session program combining CB and
relaxation training over two weeks was reported to have
been ineffective at reducing general stress responses in
the intervention compared to the control group. The study
measured stress responses including vigour, anger, fatigue,
anxiety, depression, and somatic stress responses.
Our program targeted both cognitive and behavioral
responses to stresses. It was designed to change the
participants’ cognition by identifying irrational thoughts
and practicing cognitive reconstruction. At the same time,
participants were alerted to the importance of increasing
stress management behaviors, and they practiced relaxation
exercises during the last session. This type of program
design may have more effect on dysfunctional thoughts
and stress management behaviors rather than ultimately
changing the perception of general or work-related stress
levels. The effects were similar in both males and females,
although the effects were relatively greater for females.
All sessions of the program were conducted on the same
day and post-treatment measurements were made one
week after the intervention. Therefore, there were not
enough opportunities for the participants to practise their
skills and experience any impact of their new skills on
their stress perceptions.
The present results show that changes in dysfunctional
thoughts were predictive of general stress levels as well
as personal resources in the intervention groups upon
completion of the intervention. These findings support
the theoretical framework of Beck’s model23), which
emphasizes the importance of governing the cognitive
response to reduce the impact of stress. The insignificant
findings among females were likely due to the small
sample size. Future studies should include larger numbers
of both males and females.
Our target population was teachers in the community
who were experiencing some work-related and general
stress. However, our program may be insufficient to help
those who have already developed illnesses. For example,
de Vente et al.9) reported that even a long CB-based stress
management program with 12 sessions was unsuccessful
33
in treating employees with clinical illnesses related to work
stress. Therefore, we propose that the program may serve
as an important preventive measure to promote health
through better stress management and by controlling
dysfunctional thoughts. These strategies may serve to
prevent teachers from experiencing undue stress, which
may lead to illness or burnout.
The program in this study was limited by its short
delivery, as it took place during a single day. Homework
and post-session practices are common in CB programs,
but these methods could not be undertaken in the current
program. Therefore, the participants could only discuss
and practice the skills during the group sessions, and they
did not have a chance to apply these practices in real life
before moving on to the next session. The three sessions
were originally designed to be conducted individually, as
one session per week over three weeks, and they were
piloted in a secondary school. This design was later
modified to be an intensive program over one day in
response to the requests of teachers and principals from
schools, who were interested in taking part in the program.
The work schedules of teachers in Hong Kong are very
busy, and some teachers are even required to carry out
administrative duties after office hours. The program was
therefore arranged on a Teacher Development Day, which
is an official and regular training day for teachers. Such
training days are mandated by the Education Bureau of
the Hong Kong Government for all full-time secondary
school teachers. Professional development activities are
arranged for teachers on these Teacher Development Days.
Therefore, the arrangement of our program as a mandated
professional development activity might be viewed by the
teachers as a formality, and this possible perception of our
program may somewhat affect its impact and the results.
The program with three weekly sessions would allow the
participants to have more time to practice their skills and
to experience changes in behaviors and cognition between
sessions, which may lead to a greater effect.
This study was also limited by having pre- and postintervention assessments at two time points, which showed
only the short-term effects of the program. Follow-up
measurements at a later time during a third data collection
time point may allow examination of the effects of the
program in the medium and longer term. This would allow
for an assessment of the directional and even causal effects
of the predictors. The quasi-experimental and wait-list
design adopted in this study allowed for comparisons
between the intervention groups and the control groups.
However, the results of this study should be considered as
preliminary as participants were not randomly allocated
and there were some differences in baseline demographics
of the intervention and control groups. The differences in
data collection periods for the intervention and control
groups also posed threats to the internal validity of the
study37). Although we made our best effort to maintain the
34
same periods of assessments (i.e. 4 wk apart) and similar
school situations for data collection (i.e. near the end of
semesters and avoiding school events and examination
periods) for both the intervention and control groups,
interpretation of the findings has to be cautious. Teachers’
workload and related stress levels are commonly linked
to time factors and often increase during seasonal peaks
of work demands38). Studies have shown that teachers
experience seasonal peaks of heavy workload and
increased stress around the time of writing report cards39),
or in a particular season such as autumn40). Thus, our
findings may be biased due to the season-dependent school
situations or seasonality of mood that researchers are not
aware of. Random allocation into intervention and control
groups should be performed in future studies to reduce
selection bias. Comparison of groups with non-specific
psychosocial interventions should be added to test the
intervention effects of the cognitive and behavioral
strategies of the intervention. In addition, the current study
involved teachers from the same school, and they were
reluctant to share interpersonal or organizational problems
that caused their stress. Disclosure of personal issues was
also difficult in a few groups in which the participants did
not know each other very well or have strong relationships.
As mentioned above, some teachers were less motivated
to join the program because it was perceived as a reluctant
formality that was offered as a teacher development
activity on the Teacher Development Day. Because of
these objective circumstances, their level of participation
in the group interactions and exercises might have been
affected in some way41).
The program reported in this study was brief and it
effectively enhanced stress management behaviors in
secondary school teachers. In particular, the program was
effective at changing some dysfunctional attitudes and in
reducing stress. The program can be considered as an
initial program for teachers to develop skills to cope with
work-related stress, which can then be incorporated in a
more extensive program to achieve longer-term results.
Acknowledgements: This study was funded by Seed
Funding for Basic Research, The University of Hong
Kong. The team would like to acknowledge Dr. Ora Kwo
(Faculty of Education, The University of Hong Kong),
whose agreement to let us participate in a continuous
teacher training and development program inspired our
initial conception of this study. We also thank the teachers
who participated in the study.
References
1) Kyriacou C. Teacher stress: directions for future
research. Educ Rev 2001; 53: 27–35.
2) Wilson V. Feeling the strain: an overview of the literature
on teachers’ stress. Glasgow (UK): The Scottish Council
for Research in Education, The University of Glasgow;
J Occup Health, Vol. 53, 2011
2002.
3) Tang CS, Au W, Schwarzer R, Schmitz G. Mental health
outcomes of job stress among Chinese teachers: role of
stress resource factors and burnout. J Organ Behav 2001;
22: 887–901.
4) Montgomery C, Rupp AA. A meta-analysis for exploring
the diverse causes and effects of stress in teachers. Can
J Educ 2005; 28: 458–86.
5) Antoniou AS, Polychroni F, Vlachakis AN. Gender and
age differences in occupational stress and professional
burnout between primary and high-school teachers in
Greece. J Managerial Psychol 2006; 21: 682–90.
6) Leung SSK, Mak YW, Chui YY, Chiang VCL, Lee ACK.
Occupational stress, mental health status and stress
management behaviors among secondary school
teachers in Hong Kong. Health Educ J 2009; 68: 328–
43.
7) van der Klink JJL, Blonk RWB, Schene AH, van Dijk
FJH. The benefits of interventions for work-related
stress. Am J Public Health 2001; 91: 270–6.
8) Richardson KM, Rothstein HR. Effects of occupational
stress management intervention programs: a metaanalysis. J Occup Health Psychol 2008; 13: 69–93.
9) de Vente W, Kamphuis JH, Emmelkamp PMG, Blonk
RWB. Individual and group cognitive-behavioral
treatment for work-related stress complaints and
sickness absence: a randomized controlled trial. J Occup
Health Psychol 2008; 13: 214–31.
10) Hamamci Z. Integrating psychodrama and cognitive
behavioral therapy to treat moderate depression. Arts
Psychother 2006; 33: 199–207.
11) Basco R, Glickman M, Weatherford P, Ryser N.
Cognitive-behavioral therapy for anxiety disorders: why
and how it works. Bull Menninger Clin 2000; 64:
A52–A70.
12) Bertoch MR, Nielson EC, Curley JR, Borg WR.
Reducing teacher stress. J Exp Edu 1989; 57: 117–28.
13) Tunnecliffe MR, Leach DJ, Tunnecliffe LP. Relative
efficacy of using behavioral consultation as an approach
to teacher stress management. J Sch Psychol 1986; 24:
123–31.
14) Sharp JJ, Forman SG. A comparison of two approaches
to anxiety management for teachers. Behav Therapy
1985; 16: 370–83.
15) Cecil MA, Forman SG. Effects of stress inoculation
training and coworker support groups on teachers’
stress. J Sch Psychol 1990; 28: 105–18.
16) Wu S, Li J, Wang M, Wang Z, Li H. Intervention on
occupational stress among teachers in the middle
schools in China. Stress Health 2006; 22: 329–36.
17) Hodges J, Oei TPS. Would Confucius benefit from
psychotherapy? The compatibility of cognitive
behaviour therapy and Chinese values. Beh Research
Therapy 2007; 45: 901–14.
18) Lin YN. The application of cognitive-behavioral therapy
to counselling Chinese. Am J Psychother 2002; 56:
46–58.
19) Umanodan R, Kobayashi Y, Nakamura M, KitaokaHigashiguchi K, Kawakami N, Shimazu A. Effects of a
worksite stress management training program with six
35
Sharron SK Leung, et al.: Stress Management for Secondary Teachers
20)
21)
22)
23)
24)
25)
26)
27)
28)
29)
30)
short-hour sessions: a controlled trial among Japanese
employees. J Occup Health 2009; 51: 294–302.
Education and Manpower Bureau. The Hong Kong
Government Information Center. The Government of
Special Administration Region, no date. [Online]. 2005
[cited 2006 Jan 17]; Available from: URL: http://www.
emb.gov.hk/index.aspx?langno=1&nodeid=1030
Education Bureau. Statistics on primary and secondary
school teachers 2008/09. School Education Statistics
Section. Hong Kong: Hong Kong Government SAR;
2008.
Wong FKD. Cognitive and health-related outcomes of
cognitive behavioral treatment groups for people with
depressive symptoms in Hong Kong: a randomized
waiting list control study. Aust N Z J Psychiatry 2008;
42: 703–12.
Beck AT. Depression: causes and treatment. Philadelphia:
University of Pennsylvania Press, 1967.
Leung SSK, Mak YW, Chiang VCL, Chui YY, Sun SYK,
Wong DFK. A study on a brief cognitive-behavioral
intervention for teachers’ work-related stress: a
discussion paper on recruitment and engagement issues,
HK J Mental Health 2008; 34: 43–50.
Lovibond SH, Lovibond PF. Manual for the depression
anxiety stress scales. 2nd ed. Sydney: Psychology
Foundation; 1995.
Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson
RP. Psychometric properties of the 42-item and 21-item
versions of the Depression Anxiety Stress Scales
(DASS) in clinical groups and a community sample.
Psychol Assess 1998; 10: 176–81.
Andrew MJ, Baker RA, Kneebone AC, Knight JL. Mood
state as a predictor of neuropsychological deficits
following cardiac surgery. J Psychosom Res 2000; 48:
537–46.
Barrett P, Shortt A, Healy L. Do parent and child
behaviours differentiate families whose children have
obsessive-compulsive disorder from other clinic and
nonclinic families? J Child Psychol Psychiatry 2002;
43: 597–607.
Walker SN, Sechrist R, Pender NJ. The health-promoting
lifestyle profile: development and psychometric
characteristics. Nurs Res 1987; 36: 76–81.
Weissman AN, Beck AT. Development and validation
of the Dysfunctional Attitude Scale: a preliminary
investigation. Paper presented at the 62nd Annual
31)
32)
33)
34)
35)
36)
37)
38)
39)
40)
41)
Meeting of the American Educational Research
Association, Toronto, Ontario (Canada); 1978. p.33.
Wong DFK, Chan KS, Lau Y. The reliability and validity
of the Chinese version of the Dysfunctional Attitudes
Scale form A (DAS-A) in a community sample. Int J
Psychiatry Med 2008; 38: 141–52.
Osipow SH. Occupational Stress Inventory Revised
Edition (OSI-R) professional manual. Lutz, FL:
Psychological Assessment Resources, Inc.; 1998.
Osipow SH, Spokane AR. Manual for measures of
occupational stress, strain, and coping (Form E-1).
Columbus (OH): Marathon Consulting & Press; 1981.
Spokane AR, Ferrara D, Samuel H. Osipow’s
contributions to occupational mental health and the
assessment of stress: The Occupational Stress Inventory.
In: Leong FTL, Barak A, eds. Contemporary models in
vocational psychology: A volume in honor of Samuel
H. Osipow. Mahwah (NJ): Lawrence Erlbaum; 2001. p.
79–96.
Cohen, J. A power primer. Psychol Bull 1992; 112:
155–9.
Shimazu A, Okada Y, Sakamoto M, Miura M. Effects
of stress management program for teachers in Japan: a
pilot study. J Occup Health 2003; 45: 202–8.
Krathwohl DR. Methods of educational and social
science research: The logic of methods (3rd ed.).
Illinois: Waveland Press Inc.; 2009.
Wilkinson RG. Teacher stress and coping strategies: a
study of Eastlake Comprehensive. Sch Organization
1988; 8: 185–95.
Naylor C. What do British Columbia’s teachers consider
to be the most significant aspects of workload and stress
in their work? BCTF Research Report. British Columbia:
British Columbia Teachers’ Federation. [Online].
2001[cited 2010 Sep 28]; Available from: URL: http://
www.bctf.ca/ResearchReports/2001wlc03
Salo K. Teacher stress as a longitudinal process.
Jyväskylä Studies in Education, Psychology and Social
Research, 208. [Online]. 2002 [cited 2010 Sep 28];
Av a i l a b l e f r o m : U R L : h t t p : / / u r n . f i /
URN:ISBN:951-39-1353-8
Eriksen HR, Olff MM, Ursin H. Coping with subjective
health problems in organizations. In: Dewe P, Leiter M,
Cox T, eds. Coping, health and organizations. New York:
Taylor & Francis; 2000. p.145–70.