Journal Prmrd
of Prrr ho,omorr< in Great Bntam.
Revearrh.
Vol
35. No. 2.3, ,,,I
lh3-17,.
,991. c
STRESS AMONG FACTORY WORKERS DEVELOPING COUNTRY J. SHANKAR and 0.0.
w22-399Y*Yl $3.00 + 00 1991 Pergmon Pm5 plc
IN A
FAMUYIWA*
Abstract-This cross-sectional study reveals that a significant proportion (35%) of the sampled population of factory workers experienced measurable health problems in form of anxiety. sleep disturbance. depression, somatic complaints and other clinical indicators of stress. Job stress and social support levels experienced by the group of workers judged ‘sick’ by the General Health Questionnaire (GOLL)BEKC; DP. Mmucrl fbr the Gwwrul Hew//k @wsrionnuirr. Windsor: NFER-Nelson, 1978) were significantly higher than the control group, and particularly so for employees in the lower levels of work hierarchy. In consonance with findings from the literature. job stress had a significant impact on mental health. with all the job stress parameters accounting for 41% of the variation in the general health. Those job stress variables that were significant predictors of health were job pressure, namely security and job tension, and interpersonal rewards. Extrinsic rewards i.e. those factors relating to income and status were not significant predictors of health. Perceived social support tended to increase at the higher levels of occupational status and at the work place it appeared to exert a positive influence on health presumably by counteracting the adverse affects of job stress. This study stresses the need for industrial organizations of the emergent nations to initiate measures which would enhance health status of workers as a cardinal factor in organizational growth.
INTRODUCTION
THE PROCESSESof industrialization and urbanization are known to induce stress prone patterns of behaviour. In today’s context stress may be evoked not only by the personal and familial lifestyles of the individual but also from the nature of their occupation and characteristics of their working environment. Although stress at work cannot be avoided, it need not necessarily be a damaging experience; in fact according to the Yerkes-Dodson law [l] stress can be both stimulating and productive though only up to an optimum level beyond which distress occurs. The experience of distress is associated with a wide range of physical, psychological and behavioural disturbances. The physical disturbances associated with distress such as hypertension, angina pectoris etc. are well defined. On the other hand the equally important psychological and behavioural disturbances are not adequately understood. In a working environment, these disturbances may manifest themselves in the form of excessive aggression, amotivation, forgetfulness, inability to organize self and take rational decisions, accident-proneness, absenteeism etc [2, 31. The adverse effects of these stress complications on personal health and organizations’ growth and productivity can be enormous, hence it is very important that appropriate preventive stress management programmes be formulated both at the individual and organization levels. Therefore the full impact of stress can be assessed only after a critical appraisal of these areas. The present study. however. limits itself to the understanding
Department of Psychiatry, College of Medicine, University of Lagos. PMB 12003, Lagos. *Author to whom correspondence should be addressed. Copies of the measuring schedules employed in this study are available from the corresponding 163
Nigeria. author.
164
.I. SHANKAK and 0. 0. FAMIJYIWA
of stress arising from perceived stressors within the working environment i.e. job stress. Beehr and Newman [2], defined job stress as a condition wherein job-related factors interact with the worker to change his psychological or physiological condition such that the person is forced to deviate from normal functioning. Although a detailed discussion on job-related stressors is out of place in this paper. it is worthwhile mentioning that conditions most frequently identified as producing stress are: (a) Adverse physical conditions of a job e.g. poor illumination, noise. heat. physical exertion. (b) Task demands, most important among them being work overload or an underload. job insecurity, career prospects and the occupation’s hierarchical level. (c) Role demands e.g. role conflict and role ambiguity. (d) Interpersonal conflict arising from status in-congruence, leadership style, group pressures. (e) Lack of intrinsic job qualities such as opportunities provided by the job for self development. self-utilization, value expression and autonomy. Job stress is implicated in the aetiology of several physiological. psychological and behavioural disturbances [3,4 61. However. there are relatively few scientific investigations on stress in the working situations and the relationship between job stress and mental health in real work settings [7, X]. The other thing of note is that the impact of stress on each individual is diffcrcnt. This is attributable to the influence of several conditioning factors which serve as moderators of the stress response [9]. One of these factors which is receiving considerable attention today due to its health promotive function is social support. It relates broadly to the assistance accrueing from interpersonal relationships and can be in one or a combination of forms namely; emotional and appraisal support [IO]. That social support has a direct impact on health has been established by several studies [I I 131, but the influence of social support in actually reducing the impact of stress in health is still poorly elucidated [14. IS]. Studies on occupational health conducted in Western countries have established that psychological disturbances are widespread among the industrial work force [X. 16. 171. Reports of such studies in the developing countries are scarce and none have been reported so far from Nigeria. although empirical observations in occupational health in the country indicate that there is some increase in the number of employees attending industrial clinics in recent times for a variety of health problems including anxiety states. However, firm inferences regarding the prevalance of psychological morbidity among them cannot be made since systematic data arc lacking. This study is exploratory in nature; its basic aim being to ascertain the extent of nature of psychological disturbance associated with occupational stress. It is cross-sectional and sets out to: (I) estimate the prevalence of minor psychological disorder in an industrial population; (2) detine the relationship between perceived job stress and mental health; and (3) ascertain whether social support at work place can moderate the impact of Job stress on health. METHOD The study was undertaken at the Tower Group of Companies branch in Lagos, capital of Nigeria. Three similar units of the brnnch. all of them engaged in the manufacture of aluminium products, and located in Lagos. wcrc selected for the purpose of the study.
Stress in factory
workers
165
The group comprised 400 employees belonging to three occupational groups I+-. machine operators, foremen and superintendents. Using stratified random sampling, 110 employees from this group were invited to take part in the study such that the final sample consisted of 50 machine operators, 47 foremen and 13 superintendents. These occupational groups are distinguished by their hierarchical levels of responsibilities. Machine operators have no supervisory functions, foremen are section heads while superintendents are first level managers and each was in charge of several sections. One of the authors (J.S.) was introduced to the employees by the personnel department and they were invited to complete the questionnaires with informed consent in small groups of three to five in the presence of the author so that clarifications could be dealt with on the spot if need be. Each questionnaire took approximately two hours to complete. Confidentiality of information was assured. The self-report questionnaire package in this cross-sectional study comprised a standardized health survey, a perceived social support inventory and an occupational stress measure. The authors realized the limitations of the self report questionnaire approach and would have liked to confirm health status and job stress with clinical examinations and objective stress indices. However, it was not possible in the circumstances to do so due to budgetary constraints and lack of requisite mental health personnel.
In this study health has been treated as a dependent variable and the standard 60.item version of the General Health Questionnaire-17 was used. This standardized questionnaire (GHQ) has been extensively used to determine point prevalence of minor psychiatric illnesses in cross-sectional studies. and its usefulness and applicability to the Nigerian population ascertained [19.20]. The questions address the following areas of health: general illness and somatic complaints: sleep disturbance; social dysfunction: anxiety; mild dysphoria: and severe depression.
The instrument used here measures job stress in terms of the respondent’s perceptions of stressors. It is a slightly modified form of the one developed and used by House ef ul. [IO] in their study on occupational stress. Perceived job stress has two components t>i;: ‘perceived job pressures’ and ‘perceived job rewards’. The former includes overtime, security (both of which were assessed using single questions) and a five-item ‘job tension index’. It had an estimated internal consistency reliability coefficient alpha = 0.70. Perceived job rewards include: (a) job satisfaction measured using a three-item index with reliability coefficient alpha = 0.60; (b) intrinsic rewards, which refer to perceived opportunities for self development, self utilization, value expression, anatomy and doing interesting work, measured with a six-item index, alpha = 0.85; (c) interpersonal rewards. which refer to self-ratings of recognition and appreciation from colleagues and supervisors, measured using a three-item index, alpha = 0.67; (d) extrinsic opportunity rewards which refer to self-ratings for advancement in the organization--measured using a two-item index, alpha = 0.60; and (c) extrinsic status rewards which to self ratings of rank and income, measured using a two-item index. alpha = 0.51.
The perceived social support measure used in the study measures the extext to which an individual perceived that his needs for support are fulfilled by friends, family members and colleagues at work. Perceived social support from family (PSSFm) and friends (PSSFr) were measured using the scale developed by Procidano and Heller [2l]. Both PSSFr and PSSFm measures have been proved to be homogenous with Cronbach’s alpha of 0.88 and 0.90 respectively. Perceived social support at work WdS measured using the scale developed by House [lo]. This scale had been used in several studies in organizational stress and social support. It has clear face validity and is easy to administer. The reliability and validity of the three social support measures have been ascertained [22].
RESIJLTS
The sample population was disproportionately male (95.5%) and the mean age was 30 yr; (SD = k 3.15 yr). Ninety-three per cent were married, 85% belong to the Yoruba tribe which is the dominant community in the Western part of the country; the average length of employment with the company for the subjects was 6.7 yr (SD = f4.3 yr); the average educational level for operators was Middle School, High School with technical training in the case of Foremen, and technical Diploma or university degree in the case of superintendents. All the respondents had a satisfactory working knowledge of English.
J. SCIANKAK and 0. 0. FAMLUIW~
I66
Hrulth
.stutu.r
Responses to the GHQ were evaluated using a scoring system that gives each column a score of 0, 0, 1, I. As suggested by the GHQ 17 those scoring 1I and below on this scale are taken as ‘normals’ and those scoring 12 and above are taken as probable ‘cases’. The two groups were compared in all variables using ‘[‘-tests of significance. It was revealed that 65% of the sample enjoyed normal health, with the remaining 35% experienced minor psychiatric disturbances (Table I). The two groups diKered significantly not only with respect to their health status, but also with respect to perceived social support and job stress levels. It appears that poorer mental health status of employees may be associated with higher levels of job stress and low perceived support
To explore the relationship between the job stress variables and health. correlation and multiple regression analyses were performed (Tables IIA. IIB). A job pressure index was created by combining the scores on overtime, security and job tension. Similarly :I job rewards index was created by combining the scores of all the individual job reward variables. It can be observed from Table Ila that many of the correlations are significant though modest. Health measure is positively associated with job pressure and inversely associated with job rewards [4X] Table IIA. All the job reward variables are inversely correlated with job pressures. Therefore increasing job pressure is associated with lower job rewards and deterioration in mental health status. All the job stress variables together account for 41% of the variation in the health variable (1. = 0.41. 1’ -C 0.001). This indicates that job stress had a significant impact on health. Of all the job stress variables examined in this study, job pressure and interpersonal rewards emerged as the best predictors of mental health (Table IIB). Analysing the job pressure variables independently, overtime did not have a significant impact on health while security has a significant impact (1. = 0.12. 11 < 0.5) so also was job tension (1. = 0.49. p < 0.01). With respect to extrinsic
**p < 0.05.
12.09 1.96 4 * -0.4788 -0.3895 0.3342
16.56 3.91 3 * -0.3372 -0.1206
0.20hl
Interpersonal rewards
Intrinsic rewards
-0.0355 -0.1566 0.247 0.1826
3.42 1.3 5
Extrinsic status rewards
Job pressures Intrinsic rewards Interpersonal rewards Job satisfaction Extrinsic opportunity rewards
Predictor variables
1.07 -0.38 -0.93 -0.48 -0.05
Regression coefficients
p < p < p < D< j
of Prrr ho,omorr< in Great Bntam.
Revearrh.
Vol
35. No. 2.3, ,,,I
lh3-17,.
,991. c
STRESS AMONG FACTORY WORKERS DEVELOPING COUNTRY J. SHANKAR and 0.0.
w22-399Y*Yl $3.00 + 00 1991 Pergmon Pm5 plc
IN A
FAMUYIWA*
Abstract-This cross-sectional study reveals that a significant proportion (35%) of the sampled population of factory workers experienced measurable health problems in form of anxiety. sleep disturbance. depression, somatic complaints and other clinical indicators of stress. Job stress and social support levels experienced by the group of workers judged ‘sick’ by the General Health Questionnaire (GOLL)BEKC; DP. Mmucrl fbr the Gwwrul Hew//k @wsrionnuirr. Windsor: NFER-Nelson, 1978) were significantly higher than the control group, and particularly so for employees in the lower levels of work hierarchy. In consonance with findings from the literature. job stress had a significant impact on mental health. with all the job stress parameters accounting for 41% of the variation in the general health. Those job stress variables that were significant predictors of health were job pressure, namely security and job tension, and interpersonal rewards. Extrinsic rewards i.e. those factors relating to income and status were not significant predictors of health. Perceived social support tended to increase at the higher levels of occupational status and at the work place it appeared to exert a positive influence on health presumably by counteracting the adverse affects of job stress. This study stresses the need for industrial organizations of the emergent nations to initiate measures which would enhance health status of workers as a cardinal factor in organizational growth.
INTRODUCTION
THE PROCESSESof industrialization and urbanization are known to induce stress prone patterns of behaviour. In today’s context stress may be evoked not only by the personal and familial lifestyles of the individual but also from the nature of their occupation and characteristics of their working environment. Although stress at work cannot be avoided, it need not necessarily be a damaging experience; in fact according to the Yerkes-Dodson law [l] stress can be both stimulating and productive though only up to an optimum level beyond which distress occurs. The experience of distress is associated with a wide range of physical, psychological and behavioural disturbances. The physical disturbances associated with distress such as hypertension, angina pectoris etc. are well defined. On the other hand the equally important psychological and behavioural disturbances are not adequately understood. In a working environment, these disturbances may manifest themselves in the form of excessive aggression, amotivation, forgetfulness, inability to organize self and take rational decisions, accident-proneness, absenteeism etc [2, 31. The adverse effects of these stress complications on personal health and organizations’ growth and productivity can be enormous, hence it is very important that appropriate preventive stress management programmes be formulated both at the individual and organization levels. Therefore the full impact of stress can be assessed only after a critical appraisal of these areas. The present study. however. limits itself to the understanding
Department of Psychiatry, College of Medicine, University of Lagos. PMB 12003, Lagos. *Author to whom correspondence should be addressed. Copies of the measuring schedules employed in this study are available from the corresponding 163
Nigeria. author.
164
.I. SHANKAK and 0. 0. FAMIJYIWA
of stress arising from perceived stressors within the working environment i.e. job stress. Beehr and Newman [2], defined job stress as a condition wherein job-related factors interact with the worker to change his psychological or physiological condition such that the person is forced to deviate from normal functioning. Although a detailed discussion on job-related stressors is out of place in this paper. it is worthwhile mentioning that conditions most frequently identified as producing stress are: (a) Adverse physical conditions of a job e.g. poor illumination, noise. heat. physical exertion. (b) Task demands, most important among them being work overload or an underload. job insecurity, career prospects and the occupation’s hierarchical level. (c) Role demands e.g. role conflict and role ambiguity. (d) Interpersonal conflict arising from status in-congruence, leadership style, group pressures. (e) Lack of intrinsic job qualities such as opportunities provided by the job for self development. self-utilization, value expression and autonomy. Job stress is implicated in the aetiology of several physiological. psychological and behavioural disturbances [3,4 61. However. there are relatively few scientific investigations on stress in the working situations and the relationship between job stress and mental health in real work settings [7, X]. The other thing of note is that the impact of stress on each individual is diffcrcnt. This is attributable to the influence of several conditioning factors which serve as moderators of the stress response [9]. One of these factors which is receiving considerable attention today due to its health promotive function is social support. It relates broadly to the assistance accrueing from interpersonal relationships and can be in one or a combination of forms namely; emotional and appraisal support [IO]. That social support has a direct impact on health has been established by several studies [I I 131, but the influence of social support in actually reducing the impact of stress in health is still poorly elucidated [14. IS]. Studies on occupational health conducted in Western countries have established that psychological disturbances are widespread among the industrial work force [X. 16. 171. Reports of such studies in the developing countries are scarce and none have been reported so far from Nigeria. although empirical observations in occupational health in the country indicate that there is some increase in the number of employees attending industrial clinics in recent times for a variety of health problems including anxiety states. However, firm inferences regarding the prevalance of psychological morbidity among them cannot be made since systematic data arc lacking. This study is exploratory in nature; its basic aim being to ascertain the extent of nature of psychological disturbance associated with occupational stress. It is cross-sectional and sets out to: (I) estimate the prevalence of minor psychological disorder in an industrial population; (2) detine the relationship between perceived job stress and mental health; and (3) ascertain whether social support at work place can moderate the impact of Job stress on health. METHOD The study was undertaken at the Tower Group of Companies branch in Lagos, capital of Nigeria. Three similar units of the brnnch. all of them engaged in the manufacture of aluminium products, and located in Lagos. wcrc selected for the purpose of the study.
Stress in factory
workers
165
The group comprised 400 employees belonging to three occupational groups I+-. machine operators, foremen and superintendents. Using stratified random sampling, 110 employees from this group were invited to take part in the study such that the final sample consisted of 50 machine operators, 47 foremen and 13 superintendents. These occupational groups are distinguished by their hierarchical levels of responsibilities. Machine operators have no supervisory functions, foremen are section heads while superintendents are first level managers and each was in charge of several sections. One of the authors (J.S.) was introduced to the employees by the personnel department and they were invited to complete the questionnaires with informed consent in small groups of three to five in the presence of the author so that clarifications could be dealt with on the spot if need be. Each questionnaire took approximately two hours to complete. Confidentiality of information was assured. The self-report questionnaire package in this cross-sectional study comprised a standardized health survey, a perceived social support inventory and an occupational stress measure. The authors realized the limitations of the self report questionnaire approach and would have liked to confirm health status and job stress with clinical examinations and objective stress indices. However, it was not possible in the circumstances to do so due to budgetary constraints and lack of requisite mental health personnel.
In this study health has been treated as a dependent variable and the standard 60.item version of the General Health Questionnaire-17 was used. This standardized questionnaire (GHQ) has been extensively used to determine point prevalence of minor psychiatric illnesses in cross-sectional studies. and its usefulness and applicability to the Nigerian population ascertained [19.20]. The questions address the following areas of health: general illness and somatic complaints: sleep disturbance; social dysfunction: anxiety; mild dysphoria: and severe depression.
The instrument used here measures job stress in terms of the respondent’s perceptions of stressors. It is a slightly modified form of the one developed and used by House ef ul. [IO] in their study on occupational stress. Perceived job stress has two components t>i;: ‘perceived job pressures’ and ‘perceived job rewards’. The former includes overtime, security (both of which were assessed using single questions) and a five-item ‘job tension index’. It had an estimated internal consistency reliability coefficient alpha = 0.70. Perceived job rewards include: (a) job satisfaction measured using a three-item index with reliability coefficient alpha = 0.60; (b) intrinsic rewards, which refer to perceived opportunities for self development, self utilization, value expression, anatomy and doing interesting work, measured with a six-item index, alpha = 0.85; (c) interpersonal rewards. which refer to self-ratings of recognition and appreciation from colleagues and supervisors, measured using a three-item index, alpha = 0.67; (d) extrinsic opportunity rewards which refer to self-ratings for advancement in the organization--measured using a two-item index, alpha = 0.60; and (c) extrinsic status rewards which to self ratings of rank and income, measured using a two-item index. alpha = 0.51.
The perceived social support measure used in the study measures the extext to which an individual perceived that his needs for support are fulfilled by friends, family members and colleagues at work. Perceived social support from family (PSSFm) and friends (PSSFr) were measured using the scale developed by Procidano and Heller [2l]. Both PSSFr and PSSFm measures have been proved to be homogenous with Cronbach’s alpha of 0.88 and 0.90 respectively. Perceived social support at work WdS measured using the scale developed by House [lo]. This scale had been used in several studies in organizational stress and social support. It has clear face validity and is easy to administer. The reliability and validity of the three social support measures have been ascertained [22].
RESIJLTS
The sample population was disproportionately male (95.5%) and the mean age was 30 yr; (SD = k 3.15 yr). Ninety-three per cent were married, 85% belong to the Yoruba tribe which is the dominant community in the Western part of the country; the average length of employment with the company for the subjects was 6.7 yr (SD = f4.3 yr); the average educational level for operators was Middle School, High School with technical training in the case of Foremen, and technical Diploma or university degree in the case of superintendents. All the respondents had a satisfactory working knowledge of English.
J. SCIANKAK and 0. 0. FAMLUIW~
I66
Hrulth
.stutu.r
Responses to the GHQ were evaluated using a scoring system that gives each column a score of 0, 0, 1, I. As suggested by the GHQ 17 those scoring 1I and below on this scale are taken as ‘normals’ and those scoring 12 and above are taken as probable ‘cases’. The two groups were compared in all variables using ‘[‘-tests of significance. It was revealed that 65% of the sample enjoyed normal health, with the remaining 35% experienced minor psychiatric disturbances (Table I). The two groups diKered significantly not only with respect to their health status, but also with respect to perceived social support and job stress levels. It appears that poorer mental health status of employees may be associated with higher levels of job stress and low perceived support
To explore the relationship between the job stress variables and health. correlation and multiple regression analyses were performed (Tables IIA. IIB). A job pressure index was created by combining the scores on overtime, security and job tension. Similarly :I job rewards index was created by combining the scores of all the individual job reward variables. It can be observed from Table Ila that many of the correlations are significant though modest. Health measure is positively associated with job pressure and inversely associated with job rewards [4X] Table IIA. All the job reward variables are inversely correlated with job pressures. Therefore increasing job pressure is associated with lower job rewards and deterioration in mental health status. All the job stress variables together account for 41% of the variation in the health variable (1. = 0.41. 1’ -C 0.001). This indicates that job stress had a significant impact on health. Of all the job stress variables examined in this study, job pressure and interpersonal rewards emerged as the best predictors of mental health (Table IIB). Analysing the job pressure variables independently, overtime did not have a significant impact on health while security has a significant impact (1. = 0.12. 11 < 0.5) so also was job tension (1. = 0.49. p < 0.01). With respect to extrinsic
**p < 0.05.
12.09 1.96 4 * -0.4788 -0.3895 0.3342
16.56 3.91 3 * -0.3372 -0.1206
0.20hl
Interpersonal rewards
Intrinsic rewards
-0.0355 -0.1566 0.247 0.1826
3.42 1.3 5
Extrinsic status rewards
Job pressures Intrinsic rewards Interpersonal rewards Job satisfaction Extrinsic opportunity rewards
Predictor variables
1.07 -0.38 -0.93 -0.48 -0.05
Regression coefficients
p < p < p < D< j
