high risk stress in kelatan schools pdf

STRESS AMONG SECONDARY SCHOOL TEACHERS

Vol 40 No. 6 November 2009 1359

Correspondence: Naing Nyi Nyi, Biostatisticsand Research Methodology Unit, School of Medi-cal Sciences, Universiti Sains Malaysia, HealthCampus, Kubang Kerian, Kelantan, Malaysia.E-mail: [email protected]

PREVALENCE AND FACTORS ASSOCIATED WITHSTRESS AMONG SECONDARY SCHOOL TEACHERS IN

KOTA BHARU, KELANTAN, MALAYSIA

Azlihanis Abdul Hadi1, Nyi Nyi Naing2, Aziah Daud1, Rusli Nordin1,3 andMohd Rahim Sulong1

1Department of Community Medicine, School of Medical Sciences;2Unit of Biostatistics and Research Methodology, School of Medical Sciences;

3School of Dental Sciences, Universiti Sains Malaysia, Health Campus, Kubang Kerian,Kelantan, Malaysia

Abstract. The teaching profession is an occupation at high risk for stress. This researchattempted to determine the prevalence of stress and the associated factors contribut-ing to stress among teachers in Malaysia. A cross-sectional study was conducted on580 secondary school teachers in Kota Bharu District. The instrument used to carryout the study was adopted and modified from the Depression, Anxiety and StressScale (DASS 21) and Job Content Questionnaire (JCQ). The questionnaire consisted oftwo parts: Part I consisting non-job factors (socio-demographic characteristics) andPart II consisting of psychosocial factors contributing to stress. Simple and multiplelinear regression analysis were carried out. The prevalence of stress was reported as34.0%. Seventeen point four percent of teachers experienced mild stress. Age, dura-tion of work and psychological job demands were significantly associated with stresslevel. This study indicates job-related factors did not contribute much to stress amongsecondary school teachers. Non-job-related factors should be further studied to deter-mine methods for stress reduction in teachers in Malaysia.

INTRODUCTION

Teacher stress is defined as experiencesin teachers of unpleasant, negative emotions,such as anger, frustration, anxiety, depres-sion and nervousness, resulting from someaspect of their work as teachers (Kyriacou,2001). The amount of research on teacherstress has increased steadily, and has nowbecome a major research topic in many coun-tries (Vandenberghe and Huberman, 1999;

Kyriacou, 2001; Hanizah, 2003). Social, cul-tural, economic and educational differencesbetween countries mean that one must becautious in applying research carried out inone country to another country. It is impor-tant for research regarding teacher stress tobe carried out in individual countries, wherelocal circumstances can be taken into accountin the design of the study.

The teaching profession has been cat-egorized as an occupation at high risk forstress (Chan and Hui, 1995; Pithers andForgaty, 1995). The Health and Safety Execu-tive (2000a) in the United Kingdom reportedthat teaching was the most stressful occu-pation, compared to other occupations, such

SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH

1360 Vol 40 No. 6 November 2009

as nursing, managing, professional and com-munity service occupations. It was also re-ported that two out of five teachers in theUnited Kingdom experienced stress, com-pared to one in five workers from other oc-cupations.

Okebukola and Jegede (1989) devel-oped a questionnaire in order to study fac-tors related to occupational stress amongteachers in Nigeria. They found five mainfactors related to stress: student factors,teacher factors, the school working environ-ment, administrative procedures and serviceconditions. Female teachers were more in-fluenced by the school environment and ad-ministrative procedures than male teachers.Those who were not married found studentfactors caused greater levels of stress thanin those who were married.

Borg et al (1991) produced a question-naire to investigate occupational stressamong teachers. They found the majorcauses of stress were problems of studentattitudes, problem with time and resourcemanagement, lack of professional recogni-tion and interpersonal relationships. Boyleet al (1995) validated these dimensions in or-der to form one model of factors associatedwith occupational stress; they foundworkload was another factor besides theabove four.

Studies carried out in Malaysia identi-fied several factors contributing to stressamong teachers, such as use of informationtechnology (Hanizah, 2003), years of expe-rience in teaching (Mokhtar, 1998), the work-ing environment and feelings of responsibil-ity (Ismail, 1998); the school type and per-ceptions of inadequate school facilities(Chan, 1998).

Teachers are at increased risk for burn-out. Measuring teacher stress is importantand can play an important role in under-standing the processes that lead to teacher

burnout. Burnout is described as the inabil-ity to perform both functionally and effec-tively in employment settings due to exten-sive exposure to job-related stress (Dorman,2003). The aim of this study was to explorestress among teachers in secondary schoolsin Kota Bharu, Kelantan, Malaysia. The re-searchers sought to determine the preva-lence of teacher stress and its associated fac-tors.

MATERIALS AND METHODS

Subjects

A cross-sectional study was conductedin 20 secondary schools under the authorityof the Kota Bharu District Education Office,Kota Bharu, Kelantan, Malaysia. A simplerandom sampling technique was applied toselect study subjects. All subjects were re-cruited at the school office after given writ-ten consent. Self-administered question-naires were distributed to 580 teachers. Theteachers were asked to reform the question-naire three days later. The returned question-naires were checked on site to assure com-pleteness.

The study protocol was approved by theResearch and Ethics Committee, School ofMedical Sciences, Universiti Sains Malaysiain January, 2005.

Job Content Questionnaire

The Job Content Questionnaire (JCQ)was based on the Karasek’s Demand-Con-trol Model and was used to determine thepsychosocial factors contributing to stress.The JCQ has four sections: the first was toassess for psychosocial strain; the secondwas assessing psychological and physicalstrain; the third was to evaluate technologyand the fourth was to assess wages andhours. All questionnaires were scored on aLikert scale of 1 to 4 (strongly disagree, dis-agree, agree and strongly agree). In this



study, job factors investigated were psycho-logical job demand, decision latitude (skilldiscretion, decision authority), supervisorsupport, coworker support, job insecurity,physical exertion and hazardous conditions.All the job factors were from section one ofthe JCQ.

Reliability and construct validity of theMalay version of the questionnaire was doneamong secondary school teachers in KotaBharu, Kelantan, Malaysia. A total of 68 teach-ers consented to participate in the study. Dataregarding their responses were collected us-ing a Malay version of the JCQ. Reliabilitywas determined using Cronbach’s alpha forinternal consistency whilst construct validitywas assessed using factor analysis. TheCronbach’s alpha coefficients revealed de-cision latitude of 0.75, psychological job de-mand of 0.50 and social support of 0.84. Fac-tor analysis showed three meaningful com-mon factors that could explain the constructof the Karasek’s demand-control-social sup-port model. The study demonstrated thethree scales of the JCQ were reliable andvalid for assessing the psychosocial workconditions of secondary school teachers, al-though further studies are needed to im-prove the psychological job demand scale(Azlihanis et al, 2006).

Depression Anxiety and Stress 21 ItemsQuestionnaire

Stress level was measured using theDepression Anxiety and Stress 21 ItemsQuestionnaire (DASS 21). It is a shorter ver-sion of the DASS 42. DASS questionnaire isa set of three self reported scales designedto measure the negative emotional states ofdepression, anxiety and stress. The DASSwas developed by Lovibond and Lovibond(1995) which has been increasingly used indiverse settings. The DASS questionnairemeasures negative emotional states based onclinical symptoms and meets the require-

ments of both researchers and scientists (pro-fessional clinicians). The use of the DASSquestionnaire as an objective measure ofhealth indicator (depression, anxiety andstress) in combination with the JCQ showthe direct effect of job stress on the healthproblem as recommended by Harmy (2001).

The DASS is not meant for clinical di-agnosis according to discrete diagnostic cat-egories postulated in classificatory systems,such as the DSM and ICD. This is becausethe DASS is based on a dimensional ratherthan a categorical conception of psychologi-cal disorders. A key strength of the DASS isits ability to assess depression, anxiety andstress in a brief and psychometrically soundmanner (Brown et al, 1996).

Even though the DASS 42 gives a morereliable score and more information aboutspecific symptoms, the DASS 21 has the ad-vantage of taking only half the time to ad-minister. There are several published stud-ies showing that the DASS 21 has the samefactor structures and gives results similarto the DASS 42 (Antony et al, 1998; Henryand Crawford, 2005). The DASS 42 is pref-erable for clinical work and the DASS 21 isoften used for research purposes.

All questions were scored on a Likertscale of 0 to 3, “Did not apply to me at all”,“applied to me to some degree or some ofthe time”, “applied to me to a considerabledegree or a good part of time”, “applied tome very much or most of the time”. Subjectswere asked to answer to question based ontheir experiences over the past week.

Scores for stress was calculated by sum-ming the scores for the item using the DASS21 answer template. The severity rating forstress depended on the score: normal, mild,moderate, severe and extremely severe,stress.

Data analysis

Data analysis was done using the Sta-



tistical Program for Social Science (SPSS) ver-sion 12.0 for Windows. For job factors, suchas decision latitude, skill discretion, decisionauthority, psychological job demand, super-visor support, co-worker support, physicalexertion, hazardous conditions and job in-security, these were calculated using the for-mulae of the Job Content Instrument.

The data was first analyzed using de-scriptive statistic to give an overview of thedistribution of the data. For socio-demo-graphic characteristics, job characteristicsand prevalence of stress, means and stan-dard deviations were used to describe nor-mally distributed continuous variables andmedians and inter-quartile ranges for non-normally distributed continuous variables.Frequencies and percentages were used forcategorical variables.

Associations between stress score andjob factors were analyzed using multiple lin-ear regressions analysis. Before proceedingto multiple linear regression (MLR), scatterplots between outcome variables (stressscore) and numerical independent variableswere plotted to find any associations. Onunivariable analysis, simple linear regres-sion (SLR) was used for the numerical andcategorical independent variables. Categori-cal variables with small cells, which were notsignificant at the univariate level, were col-lapsed and the small cells were combinedwhere clinically meaningful and reanalyzedusing SLR.

For MLR analysis, to obtain the prelimi-nary main effect model, variable selectionwas done using an automatic forward andbackward stepwise procedure. The modelwith the variables chosen from those twoprocedures were rechecked with only the se-lected variables because in the stepwise pro-cedure, only subjects with full data were ana-lyzed and subjects with incomplete data ormissing values were excluded. After that,

manual backward elimination was carriedout to get only the variables with a signifi-cant association with the outcome. Beforeobtaining the preliminary main effect model,manual forward inclusion was carried out,whereby all the previously excluded vari-ables were tested one by one to ensure thatno significant variables were left out beforemodel refinement was done.

For fine modeling, all two-way-interac-tion terms of independent variables chosenin the preliminary main effect model werechecked. Multicollinearity was checked withthe variables in the preliminary main effectmodel and with all the other excluded vari-ables to ensure that they were not excludeddue to multicollinearity problems with othervariables. A serious multicollinearity prob-lem was assumed present if the variance in-flation factor (VIF) was equal to or greaterthan ten, which required remedial action.

Before abtaining the final model, as-sumptions, overall model fitness, functionalforms of variables and outliers werechecked. Unstandardized predicted values(linear prediction) and standardized residu-als (error terms) were calculated using soft-ware from the fitted model. Normality as-sumption was checked by plotting a histo-gram of standardized residuals and check-ing the normality of the histogram distri-bution. A scatter plot of standardized re-siduals on the y-axis and unstandardizedpredicted values on the x-axis was made tocheck for linearity and equal variance as-sumptions. Linearity was assumed if theerror terms (standardized residuals) ap-peared randomly scattered on both sides ofand along the zero line. This also reflecteda good overall fitness of the model. Equalvariance assumption was satisfied if thevariance of the error terms (dispersion fromthe zero line) appeared to be constant alongthe unstandardized predicted value.



A scatter plot of the standardized re-siduals on the y-axis and the numericalindependent variables on the x-axis wasmade to check the appropriateness of thefunctional forms of the variables. A scatterplot of the standardized residuals on they-axis and the numerical independent vari-ables not in the model on the x-axis was alsomade to check for any relationship with out-come variables.

After the assumptions and fitness weresatisfied, the result was the best fit model,which without interaction, an interpretationof the model was obtained. Results were pre-sented with crude and adjusted regressioncoefficients, 95% confidence intervals (CI),

t-statistics with degrees of freedom, their cor-responding p-values and overall R2 values.

RESULTS

A total of 580 teachers participated inthe study. Six hundred sixty-five completedthe questionnaires giving a response rate of97.4%. Those who did not completely fill outthe questionnaire (n=15) were those who didnot respond to the question about income.

Table 1 describes the demographic char-acteristics of the study population. The meanage was 40.5 years (SD=6.41) with 404 femaleteachers (69.7%). Five hundred forty (93.1%)teachers were Malay, 544 were married

Variable Mean (SD) Median (IQR) n (%)

Age (years) 40.5 (6.41)Gender

Male 176 (30.3)Female 404 (69.7)

RaceMalay 540 (93.1)Chinese 36 (6.2)Indian 2 (0.3)Siamese 2 (0.3)

Marital statusMarried 544 (93.8)Single/Divorce 36 (6.2)

Educational statusSPM/STPM 13 (2.3)Diploma 29 (5.0)Bachelor degree 506 (87.2)Master’ s degree 32 (5.5)

aHousehold Income (RM) 2,736.4 (831.6)Duration of work (years) 11.0 (7.0,16.0)Number of children 4 (2)Smoking status

Yes 38 (6.6)No 542 (93.4)

Table 1Socio-demographic characteristics of 580 secondary school teachers in Kota Bharu.

a 15 teachers did not respond on the question of household income



Variable Mean (SD) n (%)

Type of schoolUrban 375 (64.7)Rural 205 (35.3)

Committee memberYes 351 (60.5)No 229 (39.5)

Number of classes 4.7 (1.6)Skill discretion 35.4 (3.6)Decision authority 14.4 (2.4)Decision latitude 49.8 (4.1)Psychological job demand 33.7 (4.2)Job insecurity 5.2 (1.9)Coworker support 12.3 (1.3)Supervisor support 11.7 (2.2)Social support 23.9 (2.8)Physical exertion 2.8 (0.7)Hazardous conditions 0.9 (1.5)

Table 2Job characteristics of 580 secondary

school teachers in Kota Bharu, Malaysia.

Stress level n (%)

Normal 383 (66.0)Mild 101 (17.4)Moderate 47 (8.1)Severe 29 (5.1)Extremely severe 20 (3.4)Total 580 (100.0)

Table 3Prevalence of stress among 580 secondaryschool teachers in Kota Bharu, Malaysia.

(93.8%). Five hundred six (87.2%) had a de-gree level of education and 542 (93.4%)teachers did not smoke. The median dura-tion of work was 11.0 years. The mean house-hold income and number of children wereRM 2,736.4 (SD=RM 831.6) and 4 (SD=2),respectively.

Table 2 shows the job characteristics ofthe secondary school teachers in Kota Bharu.

Three hundred seventy-five (64.7%) teach-ers taught in urban schools and 351 (60.5%)of them were members of a teachers’ union.The mean number of classes taught by theteachers was 4.7 (SD=1.6).

The number of teachers who had mildto extremely severe stress levels was 197,giving a prevalence of stress among second-ary school teachers of 34.0% (30.1, 37.8). Onehundred one teachers (17.4%) had a mildlevel of stress (Table 3).

Simple linear regression analysis of 8socio-demographic characteristics and 10 jobcharacteristics in the 580 secondary schoolteachers showed psychological job demand(p=0.037) was significantly associated withstress. Multiple linear regression analysisshowed age (p=0.002), work duration(p=0.002) and psychological job demands(p=0.027) were significantly associated withstress (Table 4).

The regression coefficient (b) was ap-plied to predict stress scores using multiplelinear regression equation y = a + bx; y = b0+ b1x1 + b2x2 + ... + bnxn where the regressioncoefficient (b) is the variation in value of theoutcome (y) when independent variable (x)is increased by one unit. Using the regres-sion equation resulting from linear regres-sion analysis, “b” was used in interpretingthe effect of independent “x” on outcome“y”.

For the age variable, one teacher 10years older than another teacher would havea higher level of stress by 4 points. This ex-plains why b = 0.39 (95% CI -0.65-0.14)meaning a teacher who is one year old willhave a 0.39 points increase in stress. There-fore, a teacher who is 10 years older (10 . 0.39= 3.9 ~ 4.0), will have stress score 4 pointshigher.

Similarly, a teacher with 10 years greaterwork experience had a higher level of stressby 3.8 points (One year more experience has



Tabl

e 4

Job

and

non

-job

fact

ors

asso

ciat

ed w

ith

stre

ss a

mon

g 58

0 se

cond

ary

scho

ol t

each

ers

in K

ota

Bha

ru, M

alay

sia.

Ind

epen

den

t var

iabl

ebc

(95%

CI)

t St

at. (

df)

p-va

lue

bd(9

5% C

I)t S

tat.

p-va

lue

Age

-0.0

55(0

.163

, 0

.053

)-1

.00

(578

)0.

317

-0.3

98(-

0.65

1, -0

.145

)-3

.09

0.00

2G

end

er 1

.125

(-0.

373,

2.6

23)

1.4

8 (5

78)

0.14

1M

arit

al s

tatu

s 2

.558

(-0.

294,

5.4

10)

1.7

6 (5

78)

0.07

9E

du

cati

onal

sta

tus

0.3

87(-

4.27

4, 5

.048

) 0

.16

(578

)0.

871

Hou

seho

ld in

com

e

<0.

001

(-0.

001,

0.0

01)

-0.6

1 (5

78)

0.54

0D

ura

tion

of w

ork

0.0

16(-

0.08

8, 0

.120

) 0

.30

(578

)0.

760

0.37

8(0

.134

, 0

.622

) 3

.04

0.00

2N

um

ber

of c

hild

ren

0.1

18(-

0.21

1, 0

.448

) 0

.70

(578

)0.

481

Smok

ing

stat

us

0.4

87(-

2.30

1, 3

.275

) 0

.34

(578

)0.

732

Typ

e of

sch

ool

0.7

12(-

0.73

1, 2

.154

) 0

.97

(578

)0.

333

Com

mit

tee

mem

ber

-0.5

86(-

1.99

7, 0

.825

)-0

.82

(578

)0.

415

Nu

mbe

r of

cla

sses

0.1

52(-

0.29

2, 0

.596

) 0

.67

(578

)0.

502

Dec

isio

n la

titu

de

-0.0

94(-

0.26

4, 0

.076

)-1

.09

(578

)0.

276

Psy

chol

ogic

al jo

b d

eman

d (P

JD)

0.1

74(0

.010

, 0

.337

) 2

.09

(578

)0.

037

0.1

84(0

.021

, 0

.346

) 2

.22

0.02

7Jo

b in

secu

rity

(JI)

0.3

12(-

0.05

3, 0

.677

) 1

.68

(578

)0.

094

Cow

orke

r su

pp

ort

-0.1

67(-

0.69

0, 0

.356

)-0

.63

(578

)0.

531

Sup

ervi

sor

sup

por

t

<0.

001

(-0.

318,

0.3

18)

0.0

0 (5

78)

0.99

9P

hysi

cal e

xert

ion

-0.2

52(-

1.25

7, 0

.753

)-0

.49

(578

)0.

623

Haz

ard

ous

cond

itio

n 0

.171

(-0.

291,

0.6

34)

0.7

3 (5

78)

0.46

7

SLR

aM

LR

b

a Si

mp

le li

near

reg

ress

ion

b M

ult

iple

lin

ear

regr

essi

on (

The

mod

el r

easo

nabl

y fi

ts;

mod

el a

ssu

mp

tion

s w

ere

met

; th

ere

wer

e no

mu

ltic

ollin

eari

ty p

robl

em o

r in

tera

ctio

nsbe

twee

n in

dep

end

ent v

aria

bles

)c C

rud

e re

gres

sion

coe

ffic

ient

d A

dju

sted

reg

ress

ion

coef

fici

ent

R2

= 0

.03



b = 0.38 (95% CI 0.13-0.62). Therefore, ateacher with 10 years more experience hada higher stress level by 3.8 points. A teacherwith a score 10 points higher in psychologi-cal job demand, had higher stress level by1.8 points. One point higher in psychologi-cal job demand, b = 0.18 (95% CI 0.02 - 0.35)resulted 1.8 points higher in stress levels).

On the linear regression model, age,years of work experience and psychologicaldemand explained 3% variation in stresslevel (R2=0.03). The rest of the 97% variationin stress score was explained by other vari-ables which were not in the final model.

DISCUSSION

The prevalence of stress ranging frommild to extremely severe, was 34.0% (95%CI 30.1, 37.8) in secondary school teachersin Kota Bharu, Kelantan, Malaysia. How-ever, the majority of teachers had a mild levelof stress (17.4%).

These findings show the teaching pro-fession is mildly stressful. These findings aresimilar to a study by Mokhtar (1998) of sec-ondary school teachers in one of the districtsin Kedah, Malaysia. He reported 17.5% hadhigh levels, 66% moderate levels and 16.5%mild levels of stress.

Hanizah (2003) studied the levels ofstress among secondary school teachers andfound the prevalence of stress among teach-ers was 55.7%. The high prevalence was dueto the working environment of the teachers,where all the teachers were using informa-tion technology whereas our study was ofgovernment-aided teachers. Noor Suhaida(2002) found the level of stress among ruraland urban secondary school teachers wasmild.

A report of occupational stress in theyear 2000 found the occupations in theUnited Kingdom with high stress (in order)

were teaching, nursing, management, pro-fessionals, other education, welfare, roadtransport and security. At least one in fivereported high levels of stress. The teacherswho reported high levels of stress were twoin five (Health and Safety Executive, 2000b).

Emotional health among secondaryschool teachers in Kota Bharu is worrisome.This problem should be addressed in orderto maintain the integrity of educational,which is important in developing a produc-tive and educated nation.

The level of job stress is expected to cor-respond closely to the country’s economicdevelopment. Higher job stress is observedin subjects from countries with a lower de-gree of economic development, such as inKota Bharu, more so than in other states.Apart from methodological differences fromother studies, the lower levels of stress maybe explained by a tendency of Malay peopleto overlook their psychological problems.

In this study, younger teachers hadmore stress than the older teachers. Thisfinding corresponds with a study by NoorSuhaida (2002) of secondary school teachersin Terengganu and Selangor. She foundteachers between the age of 31 and 40 yearsold had high stress levels. At this age, thesubjects may have other requirements to ful-fill, such as family life and financial needs.

In a survey of English teachers in Brit-ish Columbia (BCTF, 2001) respondents wereasked whether they experienced stress andhow effectively they were able to cope withthese symptoms. When asked about “Lossof Time for Family or Friends”, almost 89%of teachers in their 20s and 30s said “Yes”while about 82% of teachers in their 50s an-swered “Yes” and 73% of teachers in their60s answered “Yes”. When asked about the“Loss of Personal Interest or Hobby Time”,almost 90% of subjects in their 20s and 30sanswered “Yes” compared to 10% fewer



teachers in their 50s or 60s. It appeares clearthat the youngest teachers fare worse interms of stress and its impact.

Antoniou et al (2006) found thatyounger and older teachers perceived stressat work differently. Younger teachers expe-rienced higher levels of burnout, specificallyin terms of emotional exhaustion and dis-engagement from the profession, while olderteachers experienced higher levels of stressin terms of the support they feel they re-ceived from the government.

Contrary to the findings of the BCTF(2001) where less-experienced, youngerteachers reported both more stress and lesserability to cope with it than their older andmore experienced colleagues, our studyfound more experienced teachers wereworse at coping with stress.

Teachers with more experience are oftenselected to become “expert teachers” (FederalCongregation of School Inspectors, 1995),meaning they are chosen by the ministry tobe an expert in certain subjects. They becomea referral teacher for that particular subjectboth for students and other teachers. Expertteachers give out their opinions, determinethe strengths and weaknesses of the educa-tion system, give training to other colleaguesand become consultants in the subjects theyteach. The multiple roles played by expertteachers can lead to high levels of stress.

However, Shafee (2002) found teacherswho were in service for more than 5 yearswere highly satisfied (in relation with task,colleagues, supervision and teaching envi-ronment). This is because of developing agood long relationship between the teacherand his or her colleagues and adapting tothe educational curriculum.

In this study, the psychological job de-mands were significantly associated withstress level. Rozihaya (1998) revealed a sig-nificant relationship between stress and

workload. There is an increasing demand forwork to be completed within a limited time,therefore, time constraints are a major fac-tor contributing to stress among teachers.Nor Salmi (2002) found “not enough time tofinish up the syllabus”, especially in thosewho taught exam classes, was one impor-tant factor causing stress among teachers.

NSW/ACT IEU (2002) conducted re-search projects into workloads and percep-tions of occupational stress among teachersemployed in Catholic and Independentschools in Victoria and New South Wales(NSW), Australia, respectively. The studyshowed there were four major areas per-ceived to be of particular concern as stres-sors: workload pressures, demands of pro-fessionalism, communications/managementand career prospects. Regarding workloadpressure, 85.1% of Victorian and 91.9% of theNSW teacher subjects recorded either “high”or “moderate” levels of stress deriving fromthe multiplicity of tasks to be performed bythe teacher given the time constraints; 75.9%of the Victoria and 86.2% of the NSW sub-jects reported “high” or “moderate” stresslevels ensuing from the constancy of thework effort.

In our study, the R2 for stress analysiswas low. This means job factors do not con-tribute much to stress among secondaryschool teachers. Other non-job related fac-tors, such as personal, social, health, tech-nology and financial factors, which were notcovered in this study, might be the main con-tributory factors towards stress levels amongthe teachers in our study.

Personal factors, such as the emotionsof the teachers, can be reasons for havingstress among teachers. Teachers in HongKong have found teaching to be more andmore stressful and suggest job stress haseroded their sense of excitement in pursu-ing a teaching career. Some reported havingfeelings of being emotionally drained, had



expressed negative attitudes in respondingto students, and did not value their ownachievements (Chan, 2006).

Our study has some limitations. First,the cross-sectional nature of this study lim-its our capacity to demonstrate a cause ef-fect relationship between individual and oc-cupational variables and stress level. Second,we did not take into account all the charac-teristics of the working environment thatcould have introduced a source of potentialbias (especially working hours, subjectstaught and working position). Third, thestudy was performed with participants froma single occupation rather than evaluatingand comparing various occupational groups.School teachers generally have similar re-sponsibilities. This means differences in jobdemands and job control are more a reflec-tion of individual perception of the worksituation than in studies where persons fromdifferent occupations are included. A reasonfor taking this approach is that professionsdiffering in social status, physical activityand other factors may influence stress level.The results can not be generalized to otherpopulations. DASS is not meant for diagno-sis according to discrete diagnostic catego-ries postulated in classificatory systems,such as the DSM IV and ICD 10. DASS isbased on a dimensional rather than a cat-egorical concept of a psychological disorder.However, they evaluate the same entity withdifferent methods of measurement.

The average stress level among second-ary school teachers in Kota Bharu was mild.However, based on the overall prevalenceof stress, it still poses a risk to teachers’health and well being. The factors signifi-cantly associated with stress were age, du-ration of work and psychological job de-mands.

The subjective characteristics related toworking conditions had more influence on

the subjects’ emotional health than the ob-jective characteristics. This shows that jobfactors did not contribute much to stressamong secondary school teachers.

Other non-job related factors, such aspersonal, social, technology and financialfactors which were not covered in this study,should be looked into, in order to determinethe main contributory factors towards stresslevel among secondary school teachers.

The findings of this study provide in-formation regarding the magnitude of theproblem among secondary school teachersin Kota Bharu. They point toward a need tofurther explore the underlying reasons orfactors leading to stress. This suggests pos-sible weaknesses in our existing teaching ser-vices as well as measures that improve them.

A number of the results of the presentstudy are important for the potential theyhave in terms of public health and policyimplications, especially because some easyto implement measures to change conditionsor behavior at work could help improvesome of the problems identified. Better in-formation for teachers and adjusting theirtraining could improve teacher behavior,which could have an impact on the occur-rence of emotional health problems, such asstress. Evaluation of some areas, such asadopting new teaching strategies, officialhours of duty and levels of extra-curricularinvolvement, could provide a better regula-tory framework within which education staffwould be able to find solutions to some oftheir problems. Other interventions, such asstress management, can be carried out toimprove the workplace. Intervention regard-ing non-job factors (eg, counseling) shouldnot be left out since these were the majorcause of stress in teachers in our study.

This study only evaluated the level ofstress among secondary school teachers. Itwould be helpful to compare primary and



secondary school teachers to determinewhich group has more psychological stress.Other factors may also affect the emotionalhealth status of teachers. We recommendother factors be considered, such as person-ality and technology use, in order to get atrue picture of factors associated with theemotional health of teachers. Most critical isthe need for prospective investigations toevaluate causality.

In our study, the R2 for stress analysiswas low, thus job factors did not contributemuch to stress among secondary schoolteachers. Other factors, such as personal, so-cial, technological and financial factorswhich were not being covered in this study,should be looked into more deeply in orderto obtain the main contributory factors to-wards stress among teachers.

ACKNOWLEDGEMENTS

The authors wish to thank ProfessorRobert A Karasek for the use of the Job Con-tent Questionnaire in this study. The authorsalso wish to thank the Ministry of Educa-tion and the principals, senior assistants andteachers of the several secondary schoolswho gave their full commitment to makingthis study possible.

REFERENCES

Antoniou AS, Polychroni F, Vlachakis AN. Gen-der and age differences in occupationalstress and professional burnout betweenprimary and high-school teachers in Greece.J Manage Psychol 2006; 21: 682-90.

Antony MM, Bieling PJ, Cox BJ, Enns MW,Swinson RP. Psychometric properties of the42-item and 21-item versions of the Depres-sion Anxiety Stress Scales (DASS) in clini-cal groups and a community sample.Psycholog Assess 1998; 10: 176-81.

Azlihanis AH, Nyi NN, Aziah BD, Rusli BN. Re-liability and construct validity of the Malay

version of the Job Content Questionnaire(JCQ) among secondary school teachers inKota Bharu, Kelantan, Malaysia. SoutheastAsian J Trop Med Public Health 2006; 37: 1254-59.

Borg MG, Riding RJ, Falzon JM. Stress in teach-ing: a study of occupational stress and itsdeterminants among primary school teach-ers. Br J Educ Psychol 1991; 11: 59-75.

Boyle GJ, Borg MG, Falzon JM, Baglioni AJ. Astructural model of the dimensions ofteacher stress. Br J Educ Psychol 1995; 65: 49-67.

British Columbia Teacher Federation (BCTF).Teacher workload and stress: an internationalperspective on human costs and systemicfailure. Research Report 2001; Section III,WLC-01. [Cited 2006 Jan 24]. Available from:URL: http://www.bctf.ca/ResearchReports /2001wlc01/report.html

Brown TA, Chorpita BF, Korotitscw K, BarlowDH. Psychometric properties of the Depres-sion Anxiety Stress Scales (DASS) in clini-cal samples. Behav Res Ther 1996; 35: 79-89.

Chan DW. Emotional intelligence and compo-nents of burnout among Chinese secondaryschool teachers in Hong Kong. TeachingTeacher Educ 2006; 22: 1042-54.

Chan DW, Hui EKP. Burnout and coping amongChinese secondary school teachers in HongKong. Br J Educ Psychol 1995; 65: 15-25.

Dorman, J. Testing a model for teacher burnout.Aust J Educ Develop Psychol 2003; 3: 35-47.

Federal Congregation of School Inspectors, Min-istry of Education. Annual report 1995: 80.

Hanizah MY. The effect of information technol-ogy usage on the prevalence of stress amongschool teachers in Selangor and factors af-fecting the stress. Kuala Lumpur: UniversitiKebangsaan Malaysia, 2003. 110 pp. Masterthesis.

Harmy MY. A comparative study of job stress innurses in Hospital Universiti Sains Malay-sia and clerical workers in Kota Bharu dis-trict in Kelantan. Kubang Kerian: UniversitiSains Malaysia, 2001. 107 pp. Master thesis.



Health and Safety Executive. Stress at work: Whois affected most? Press release E 246:00,2000a.

Health and Safety Executive. The scale of occu-pational stress: A further analysis of theimpact of demographic factors and type ofjob. 2000b. [Cited 2006 Jan 24]. Availablefrom: URL: http://www.hse.gov.uk/re-search/crr_pdf/2000/crr00311.pdf

Henry JD, Crawford JR. The short-form versionof the Depression Anxiety Stress Scales(DASS-21): construct validity and normativedata in a large non-clinical sample. Br J ClinPsychol 2005; 44: 227-39.

Ismail O. Work perception and its relationshipwith work stress: A study on mathematicsand science teachers in Kota Bharu, Kelantan.Kuala Lumpur: Universiti Kebangsaan Ma-laysia, 1998. 105 pp. Master thesis.

Kyriacou C. Teacher stress: directions for futureresearch. Educ Rev 2001; 53: 27-35.

Lovibond PF, Lovibond SH. Manual for the de-pression, Anxiety and Stress Scales. Sydney:Psychology Foundation, 1995.

Mokhtar A. Work stress among secondary schoolteachers: A study in Kulim-Bandar Bahru,Kedah, Darul Aman. Kota Samarahan:Universiti Malaysia Sarawak, 1998. 102 pp.Master thesis.

Noor Suhaida S. Work stress and associated fac-

tors among urban and rural secondaryschool teachers. Serdang: Universiti PutraMalaysia, 2002. 97 pp. Master thesis.

Nor Salmi L. Occupational stress among second-ary school teachers in Tumpat, Kelantan,Malaysia. Serdang: Universiti Utara Malay-sia, 2002. 112 pp. Master thesis.

NSW/ACT Independent Education Union. Stress!Burnout! Among Teachers 2002. [Cited 2006Feb 24]. Available from: URL: http://www.ieu.asn.au/ohs/stressburnout.html

Okebukola PA, Jegede OJ. Determinants of oc-cupational stress among teachers in Nige-ria. J Educ Stud 1989; 15: 23-36.

Pithers RT, Forgaty GJ. Occupational stressamong vocational teachers. Br J Educ Psychol1995; 65: 3-14.

Rozihaya Y. Organizational factors that contrib-ute to teacher stress. Kubang Kerian:Universiti Sains Malaysia, 1998. 102 pp.Master thesis.

Shafee H. Is duration of service has any relation-ship with level of job satisfaction: study inTanjung Rambutan and Ipoh, Perak, Malay-sia. Kuala Lumpur: Universiti KebangsaanMalaysia, 2002. 115 pp. Master thesis.

Vandenberghe R, Huberman M, eds. Under-standing and preventing teacher burnout:a sourcebook of international research andpractice. Cambridge: Cambridge UniversityPress, 1999.

high risk stress in kelatan schools pdf

Documents