Saturday, March 13, 1999
8:00 - 10:00 a.m.
Demand, Control, and Social Support Interaction: Effects on Sickness Absenteeism
Mark Cropley, Ph.D., Andrew Steptoe, D.Phil., and Katherine Joekes., MSc., University of London, UK.
Karasek (1979) proposed that job demand and decision latitude may interact to produce psychological strain. Demanding jobs, accompanied by low decision latitude were considered detrimental to employees’ health, and were thought to be associated with stress related disorders (e.g., cardiovascular disease and sickness absenteeism).
In 1988 Johnson and Hall redefined the job-strain model by introducing the concept of work-related social support (the Demand-Control Support Model), suggesting that supporting interpersonal relationships at work may function as a moderator in stressful jobs.
The aim of the present study was to compare the interaction effects of the three formulations of the job-strain model: the original demand by decision latitude (control with skill utilization) interaction, the demand by control interaction, and the product term including work-related social support. As part of a survey on life style and health in teachers, 733 full-time primary and secondary school teachers in South London completed a questionnaire which included measures of job-strain, work social support, and self reported sickness absenteeism.
Regression analyses were used to identify which of the three interaction terms best predicted sickness absenteeism. There were three steps in each analyses. Age, gender and marital status, was entered on the first step. School type (primary or secondary), teaching grade and length of service were entered on the second step. The relevant interaction effects were entered on the third final step.
Consistent with the job-strain model, sickness absenteeism in the past 12 months was predicted by interaction effects after controlling for demographic and work related variables. There was a significant interaction effect between decision latitude and demand (F = 10.5, p<.001), a significant interaction effect between control and demand (F = 12.4, p<.001), and a significant interaction effect between control, demand and work social support (F = 11.5, p<.001). Although supporting evidence was found for each formulation of job-strain, it is interesting that each model has similar predictive power. The inclusion of work social support and skill utilization did not increase the variance in sickness absence accounted for by the model.
It is concluded that the interaction between demand and control is the most important feature of job-strain in predicting sickness absenteeism in school teachers.
CORRESPONDING AUTHOR: Mark Cropley, Ph.D., Department of Psychology, St George’s Hospital Medical School, University of London, Hunter Wing, Cranmer Terrace, London. SW17 0RE. UK.
A Conceptual Model for a Healthy Workplace Tool
Lynda S. Robson (1), Harry S. Shannon (1,2), Michael F.D. Polanyi (1), Mickey S. Kerr (1), Joan M. Eakin (3), Ann-Sylvia Brooker (1), Donald C. Cole (1)
(1) Institute for Work & Health, Toronto (2) Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, (3) Department of Public Health Sciences, University of Toronto, Toronto
A conceptual model of a "healthy workplace" has been developed for the purpose of guiding the subsequent development of workplace assessment tools. A "healthy workplace" is defined as a workplace which promotes and protects the health and safety of its employees. Assessment tools based on such a model can be used for diagnostic, evaluative or comparative purposes.
A review of the literature relevant to a healthy workplace conceptual framework has identified relevant theory and models from various fields: occupational health and safety, health promotion, occupational and social psychology, organizational design. The present model offers a synthesis of these perspectives using a social-ecological approach.
The model can be considered to consist of two basic elements: 1) determinants of health in the workplace and 2) interventions to modify these determinants.
Determinants of health are found at the levels of the society, the workplace, the job and the individual. Factors at home also impinge on the individual. At the level of the job, both physical and psychosocial exposures are important. Influencing all of these exposures are variables at the organizational level, including policy, practices, technology, organization, and climate/culture.
Intervention can take place at any of the levels included in the model. The effectiveness of workplace interventions, in terms of promoting or protecting employee physical and mental health, also depends on variables at the organizational level.
CORRESPONDING AUTHOR: Lynda S. Robson, PhD, Institute for Work & Health, 250 Bloor St. E., 7th floor, Toronto, Ontario, M4W 1E6, Canada.
Determinants of Psychological Well-being in the Canadian Working Population: A Structural Equation Modelling Approach
Cole DCA,B, *Ibrahim SAA, Shannon HSA,B, Scott FB, Eyles JB, Goel VC
A Institute for Work & Health, Toronto, ON, Canada M4W 1E6
B McMaster University, Hamilton, ON, Canada, L8N 3Z5
C University of Toronto, Toronto, ON, Canada, M5S 1A8
We analyzed data from the 1994-5 Canadian National Population Health Survey (NPHS). This multi-stage, cross-sectional survey included 8273 working adults (18-64 years) who among other items, answered an abbreviated version of the Job Content Questionnaire. We hypothesized that broad socioeconomic determinants would set a background for life stressors and work stressors to affect psychological well-being, mediated by job satisfaction (measured) and modified by non-work social support (measured). Psychological outcomes included were: distress in the last month index based on the Composite International Diagnostic Interview, Rosenberg's self-esteem score, Antonovsky's sense of coherence index and Pearlin's mastery score.
We formulated socio-economic status, work stressors and life stressors as composite variables, with each of the measure variables structured as causal. Psychological well-being was construed as a latent variable loading onto effect measures of an underlying construct. Structural equation modelling EM was done separately for men and women. The distribution of 2 of the 15 measured variables had high kurtosis for women (2.3,3.5) and for men (3.4,4.5). Despite this departure from multivariate normality we used maximum likelihood estimation methods. As suggested in MacCallum & Browne (1993), one of the loadings to the composite variables was set to one and their error variances were set to zero.
Goodness of fit was initially poor but allowing the error for measures of psychological well-being to correlate resulted in an acceptable fit. Loadings of the measured variables on the latent construct psychological well-being were adequately high. Overall the model accounted for 46%of the variance in psychological well-being among women and 51% among men. Work stressors affected psychological well-being both directly (-0.15 for women, -0.17 for men) and indirectly via job satisfaction (-0.07 for women, -0.08 for men), resulting in total effects of -0.22 for women and -0.24 for men.
Formulation of psychological well-being as a latent construct was useful in dealing with multiple relevant psychological measures. Operationalizing stressors as composite variables, permitted assessment of their influence on what are often regarded as stable traits (e.g. mastery). Similar SEM models held for women and men.
CORRESPONDING AUTHOR: Selahadin Ibrahim, Institute for Work & Health, 250 Bloor Street East, Toronto, ON, Canada, M4W 1E6.
Job Demands-Control-Support Model in Selected Korean Blue-collar Workers
Sang Jun LEE, M.D., Ph.D., Division of Industrial Environment and Health, Ministry of Labor, Republic of Korea
Background-It is expected that the deterioration of psychosocial work environment, induced by the rapid economical integration and globalization, makes job stress become the major occupational health problem in near future of Korea. The economic difficulties and a relief loan from IMF will also contribute to accelerate this change in Korea.
Objective-This study was designed to offer basic knowledge about the relationship between psychosocial work environment (skill discretion, decision authority, substantive complexity, psychological demand, physical workload, physical exertion, job insecurity, hazards of working condition, coworker support & supervisor support) and health (self-rated health, result of health examination & sickness absence) among Korean blue-collar workers.
Methods-The cross-sectional survey was conducted on 6,513 workers in 42 metal-related workplaces by self-administered questionnaire. The response rate was 63.5%(4,135 workers). But 3,402(52.2%) questionnaires from only blue-collar workers were analyzed by various statistical methods including covariance structural analysis with PC-SAS software program.
Main Results-The original measures of job insecurity and coworker support showed low reliability and validity in Korean blue-collar workers. Three dimensional model of demands, control and support of psychosocial work environment was the best fitted model compare to two alternative models, the unstructured and demands-control.
Korean blue-collar workers perceived decision authority to be dimension of social support as well as job control. On the other hand, job insecurity was recognized as dimension of job control rather than job demand.
Health of low job-strain and high supported group was better than high job-strain and low supported group. However the mutual effects of job demand, job control and social support on health were not synergistic but simply additive. Also job control had only indirect effect on health through buffering effect of job demand, while social support gave only direct effect on health without buffering effect of job demand.
Conclusion-These results showed that original indicator of job insecurity and coworker support may be inadequate for its direct assessment of psychosocial work environment from Korean blue-collar workers. Also it was suggested that the difference of social and workplace culture should be considered in job stress studies with job stress measures and models developed in foreign countries. In conclusion, Job demands-control-support model is thought to be good job stress model to be able to explain the effect of psychosocial work environment on health, especially mental health, of Korean blue-collar workers.
Examining Ethnic Differences in Occupational Stress Using Karasek's Model
Lisa M. Perez, Ph.D., Bowling Green State University
This study investigated the issues of occupational stress and ethnic diversity, by examining Karasek’s demands-control model of job strain in a sample of White and Hispanic working women. Karasek’s model suggests that high levels of job demands coupled with a lack of control over relevant aspects of the work environment result in high levels of strain. The study also extended Karasek’s model to incorporate potential moderators of the model including social support and ethnicity. Additionally, this study examined the effects of stressors particularly relevant to ethnic minorities: acculturation, work group composition, availability of mentoring opportunities, and perceived discrimination. It was expected that findings would corroborate the demands-control model but that ethnicity and ethnicity-related stressors would exacerbate the negative effects of stressful work environments (i.e., high demand-low control jobs).
Surveys assessing job stressors, ethnicity-related stressors, and job strain were distributed to women belonging to various business and professional associations. Surveys were distributed by mail or at regular group meetings. A total of 107 usable surveys were returned. Forty-six percent of the final sample was Hispanic.
Results showed some support for the additive effects of demands, control, and social support in predicting job-related outcomes, such as intent-to-quit and job satisfaction. Support for the interactive effects of the model was limited.
Results also did not support the inclusion of ethnicity as an additional moderator in the demands-control framework. Nevertheless, being Hispanic was predictive of more self-reported health problems, and interacted with job demands to predict health problems. White women reported more health problems when their jobs were less demanding while Hispanic women reported more health problems when their jobs were more demanding. Also, Hispanic women reported more depressive symptoms at higher levels of job demands, while job demands were unrelated to depression among White women. Furthermore, social support appeared to buffer the relationship between job demands and some forms of strain for White women but not for Hispanics.
There was limited support for the deleterious effects of ethnicity-related stressors. However, Hispanic women who worked in high demand-high discrimination conditions reported the highest levels of job strain. Results also showed that Hispanic women who reported greater job control also reported greater job satisfaction even when they perceived high levels of workplace discrimination.
These research findings are limited by small sample size, low response rate, and a reliance on cross-sectional, self-report data.
CORRESPONDING AUTHOR: Lisa M. Perez, Ph. D., National Institute for Occupational Safety and Health, Div. of Biomedical and Behavioral Science/MS C-24, 4676 Columbia Parkway, Cincinnati, OH 45226-1998, USA.
A Cognition-Reaction-Outcome Framework for Stress Research
Jennifer R. Rosenkrans, George Mason University
The cognition-reaction-outcome framework described in this paper offers a comprehensive means of analyzing workplace stress. It expands on existing models, shifting focus from stressors and outcomes to cognitive processes and suggesting how control affects outcomes.
Most job-stress research examines the relationships between specific stressors (such as control), individual differences, and outcomes, only acknowledging the role of cognition in passing. This focus on specific elements implies that immutable worker characteristics explain stress outcomes. In contrast, the cognition-reaction-outcome framework recognizes individual differences as important mediating factors, but focuses on perceptual, attributional, and affective reactions which are more useful areas of research.
Perception of environmental problems may be affected by a wide variety of factors. For example, workers may have different tolerances for temperature or sensitivity to environmental characteristics. One person may not notice a hot room or an air quality problem until it is mentioned by a coworker. On the other hand, social or cultural conventions may dictate that overt reactions to the environment are not appropriate. Thus, a set of job-specific (individual, social, occupational, and cultural) conditions moderate the stressor-perception relationship.
Once a job characteristic is perceived as noxious, workers make attributions about the source of that stressor. These attributions are determined by salient characteristics of the environment, social conventions, and cultural beliefs, as well as workers’ own physiological states. This process is particularly implicated in extreme stress reactions such as outbreaks of mass psychogenic illness and some instances of sick building syndrome. Ability to perceive social and occupational stressors is not a factor in determining their effects; attribution is.
The assumed source of the stressor, in conjunction with social norms and roles, determines the amount of perceived control the worker has over the problem. The same headache may result in a variety of responses, depending on the perceived cause. Though the actual cause may be unrelated to the identified stressor, the attribution determines perceived control over the source of the problem, subsequent affective reactions and behavior.
Affective reactions to stress have definite indirect effects on occupational performance, and are often operationalized as stress outcomes. Realistically, affective reactions only influence performance and health through behavior and physiological changes and are therefore an indirect outcome measure at best. Defining affective reactions as components of stress confines their use to measurement of stress levels which, in turn, limits the measurement of stress outcomes to more tangible, objective problems such as poor health and productivity.
CORRESPONDING AUTHOR: Jennifer R. Rosenkrans, 2010 Colts Neck Road #22B, Reston, Virignia, 20191-2010.
Ecological Model of Occupational Stress: Application to Emergency Dispatchers
Mary K. Salazar, Ed.D., R.N., Randall Beaton, Ph.D., University of Washington School of Nursing
The ecological model described in this paper offers a convenient and comprehensive method to evaluate stressors among working populations. Ecological theory, which is based on the biological and social sciences, purports that the interrelationships between organisms and their environment are interactive and highly complex. Basic premises of ecological theory are that systems are dynamic, that change is constant, and that "everything is connected to everything else." The ecological model provides a useful approach to examine complex phenomena that are the result of multiple factors in an environment.
The ecological/occupational model includes four nested levels of stressors: 1) The microsystem which consists of the environment immediately surrounding the worker or group of workers; it includes the interactions which a worker experiences directly, the physical features of the environment, and the activities that occur there.; 2) the organizational system which is made up of the multiple structures and function’s that constitute a work organization; 3) the peri-organizational system which refers to the forces within the societal system in which the individual and organization are imbedded; and 4) the extra-organizational system which includes the cultures, societal norms, and traditions as well as government and economic policies which indirectly affect workers. These stressors lead to physical, psychological, and behavioral responses that are ameliorated by buffers in the workers’ environments. Through an examination of these various dimensions of the model, a worker’s health status can be predicted.
This paper applies this model a group of workers who experience high strain, emergency dispatchers (EDs). The model is used to examine each level of stressors among these workers. Examples of identified stressors are the need to perform tasks in a context of urgency (microsystem), an autocratic/bureaucratic management (organizational), budgetary constraints (peri-organizational), and responsiveness to prevailing cultures and norms (extra-organizational). Buffering mechanisms include individual and group coping strategies. This assessment indicated that occupational stressors experienced by EDs may lead to both acute and chronic stress responses. Psychological responses include secondary trauma symptomology and cynicism; physical stress includes headaches and gastrointestinal symptoms; and behavioral manifestations include overeating and dysomnias. This analysis suggests that occupational stressors may lead to sub-optimal levels of health in emergency dispatchers. It also suggest that more research is needed to facilitate the development of appropriate interventions and to provide a more the healthful environment for these workers.
CORRESPONDING AUTHOR: Mary K. Salazar, Ed.D., R.N., School of Nursing, University of Washington, Box 357262, Seattle, WA 98195
The Demand-Control Model, and the Demands and Controls Associated with Front-line Service Work
By Sandra M. Richmond, Ph.D. and Kristopher A. Weatherly, Ph.D.
This study had three objectives. The first objective was to replicate the demands-control model in a previously untested population of fast-food workers. The demands-control model postulates that features of job design can affect employee health and well being. Karasek and Theorell (1990) suggest that if work design is associated with cardiovascular disease and depression, then work design might also affect productivity and other workplace behavior and attitudes. The second objective was to explore and measure the customer service demands faced by fast-food workers by developing scales that measure workers' perceptions of customer behavior. The third objective was to extend the demands-control model by including construct of customer behavior demands along with a measure of management support, and by assessing the incremental effects of these constructs on work attitudes as reported by fast-food workers.
There are at least four reasons why the fast-food environment is an ideal site to test how demands-control model applies to service workers. First, many of the characteristics of fast-food work resemble traditional manufacturing work, with steady lines of customers, either on foot or in their cars, taking the place of widgets on assembly lines. Second, although a variety of occupations have been used to test the constructs of the demands-control model, many of these work groups have been predominately male (Carrere et al.,1991) or female (Carayon, 1993; Carayon, Yang, & Lim, 1995). The population of fast-food workers is characterized by a combination of male and female employees representing a diverse mix of racial and ethnic cultures and national origins. Third, there are characteristics of fast-food work that are not representative of manufacturing work. Fast-food workers must work fast in routinized jobs doing repetitive tasks (Garson, 1988; Leidner, 1993) much the same as manufacturing workers, but they must also deal with customers in a positive and polite manner (Leidner, 1993; Gutek, 1995).
Hochschild (1983) has identified this requirement to deal with customers in an organizationally-prescribed manner as "emotional labor." Emotional labor demands in fast food work have not been tested, and the effects of these demands associated with handling customers in this context have not been measured. Finally, while the number of fast-food employees is increasing, the problems associated with retaining these workers are also growing (Nichols, 1988). Since a key facet of the demands-control model is that of job redesign, this research may help identify fast-food job demands or controls that can be modified so employee stress can be decreased, and employee satisfaction and commitment can be increased.
In this cross-sectional field study of 203 employees (97% response rate) in 14 locations of a fast-food organization, self-report data provided by workers and interview data from managers were used to assess the effects of the work environment on worker attitudes and behavior. Job demands, worker control and management support (Karasek & Theorell, 1990) were the predictor variables in this research. An additional job-demand scale of customer behavior associated with service encounters (Gutek, 1995) was created, measured, and tested.
Each set of variables used in this study was subjected to a principal components analysis with VARIMAX rotation. The individual items were combined, averaged and analyzed for internal reliability. These results, along with means, standard deviations, alphas, and intercorrelations are presented and discussed in this paper. Two separate hierarchical regression analyses were performed. The first set was conducted to test the predictive power of the basic demands-control model. The second set of analyses was conducted to test the demands-control model with addition of management support and employee perceptions of emotional labor and customer behavior. Results of these hierarchical regression analyses are presented and discussed.
Results of this study indicate that worker control and management support are negatively associated with reported stress and positively associated with reported satisfaction and commitment.
Additionally, customer behavior and worker demands were positively associated with reported stress and customer behavior was negatively associated with reported satisfaction and commitment.
Implications regarding the impacts of worker demands and control, and the effects of emotional-labor expectations and perceptions of customer behavior are discussed. Application of the demands-control model (Karasek & Theorell, 1990) in a fast-food setting contributes useful data regarding the impacts of this work environment on fast-food workers, and offers valuable information regarding the attitudes and behaviors of this growing population of employees.
CORRESPONDING AUTHOR: Sandra M. Richmond, Ph.D. Organization Development Consultant, Walt Disney World Co., Post Office Box 10000, Lake Buena Vista, FL 32830-1000
Relating Psychological Job Demands with Measured and Self-reported Physical Demands
Kerr MS, Shannon HS, Frank JW, Institute for Work & Health, Norman RWK, Wells RP, Neumann WP, University of Waterloo and the Ontario Universities Back Pain Study (OUBPS) Group
This paper will discuss the results of post-hoc analyses to examine the relationship between the psychological demands scale, from the Karasek Job Content Questionnaire (JCQ), and both directly measured and self-reported (psychophysical) physical demands of work.
This investigation was part of a workplace-based, epidemiologic case-control study of risk factors for low-back pain that was specifically designed to assess the relative importance of work-related biomechanical and psychosocial factors. The interview-assisted study questionnaire included the core JCQ questions. Data were also collected on individual characteristics, clinical and health-related quality of life aspects, and psychophysical ratings of the physical demands of the job. Biomechanical variables, assessed by direct job measurements, included video-based computer model estimates of peak and cumulative forces on the lumbar spine, estimates of average and low-loading compression forces, as well as peak hand forces, and several posture and trunk motion variables.
In bivariable analyses, cases and controls differed statistically on the psychological demands scale, with an odds ratio of 1.7 (95% CI 1.15, 2.43), but this disappeared in the final multivariate logistic regression model when self-reported physical demands and measured physical demands were included in the model. Results of correlation and factor analyses indicated that the psychological job demands scale was strongly correlated with the self-rated perceptions of physical demands. Correlation of the psychological demands scale with the direct biomechanical measures was generally low. The factor analysis also suggested that a new "active demands" scale, comprised of three of the psychological demands scale questions, together with the two questions used in the physical exertion scale and one from the job control scale, provided a better fit for the data than the original scales.
Our results suggest that the core version of thew JCQ psychological job demands scale behaved more as a measure of self-reported physical rather than psychological demands in our study setting, an automotive assembly plant.
CORRESPONDING AUTHOR: Michael S. Kerr, Institute for Work & Health, 250 Bloor St. E., Toronto, ON., M4J 3M9, CANADA
Epidemiological Characteristics of Workers' Sudden Death as Claims due to Work Stress
Aims: Worker sudden death can be compensated in Korea if work related stress or overwork proves to be a cause. Since no measure is officially provided for verification of stress. This study was done to identify epidemiological characteristics, risky industry and occupation.
Method: All the records of claims for workmen's compensation during 1994 other those by accidents were reviewed as for the disease categories - cerebrovascular accidents(CVA), coronary artery diseases(CAD), sudden deaths from unidentified cause(SD), and all other diseases(OD) - to compare their stress components.
Result: CVA was the largest cause of death(213 cases) claimed for compensation due to work stress, and CAD was the next(163 cases). Age level of OD and SD were lower than CAD and CVA. Among industries covered by workmen's compensation, electric and gas industry showed the highest death rates per million workers from CVA(55.9), CAD(55.9) and SD(55.9). Service workers showed the highest death rates from CAV(53.7), CAD(50.4) and SD(19.0). When work stress was divided into physical exertion, emotional excitement, prolongation of work time, change of work content and responsibility, physical exertion and psychological excitement within one day before the onset of symptoms were significantly associated with SD, CAD and CVA compared to OD. Among four disease groups of claims, CAD have shown the highest rate of work shift(36.8%). History of night work at the time of episode have shown the highest rate in SD(37.1%). Hypertension history of the group of CVA is the highest rate(71.9%) among four groups. Compared to OD, CAD and CVA showed the significantly higher rate of hypertension history.
Discussion: Based on the claims for workmen's compensation, the result can not be generalized because of its selection bias. Most of all claims already reviewed by bereaved family members and their colleagues in workplace before application for compensation. Even though this selection bias, the risky occupation and industry can be identified by the distribution of the claim-rate per total working population on the assumption of the same claim-rate among occupations and industries. Among the selected work stresses, physical exertion and emotional excitement within one day before the onset of symptoms are the most important factors in the episode of SD, CAD and CVA. This kinds of work stress can be understood as precipitating factor of such episode. In spite of night work and work shift can be suspected as a risk factor of SD, CAD and CVA, but the result of this characteristics of working type have shown the rate of under 50% in all three groups. Hypertension history can be explained as a background characteristics of CAD, CVA compared to OD and SD based on the result. Management of hypertension can be suggested as one of the strategies for prevention of death in workplace.
Conclusion: Not all the components of work stress are associated with death. The work characteristics of living controls matched for the age and workplace will be examined to further delineate the stress components.
Burnout in the Land of the Kiwi: A Multi-measure Approach
Robert A. Boudreau, Ph.D., The University of Lethbridge, Michael P. Leiter, Ph.D., Acadia University, Wilmar B.Schaufeli, Ph.D., Utrecht University, and Robert T. Golembiewski, Ph.D., The University of Georgia.
Even though burn-out has just celebrated its silver anniversary in print, many questions involving the measurement of the construct remain. By our count, over 30 different measures have been used since 1974. A perusal of recent 1997-1999 academic journal articles suggests the choice of burnout "metric" has narrowed somewhat although complete agreement remains an elusive if not an impractical goal. Most recently, two measures of burnout have emerged to dominate the burnout landscape-- the Maslach Burnout Inventory (MBI) originally developed by Christina Maslach and Susan Jackson and the Burnout Measure (BM) proposed by Ayala Pines and Elliot Aronson.
Developed from earlier work on "tedium," this latter approach centres on individual states of exhaustion and is not restricted to certain types of occupations. The BM is described as a self-diagnosis instrument and consists of 21, "feeling" and "being" experiences that a person rates on a seven-point, frequency scale. The MBI was originally developed to measure burnout in those individuals that worked in providing some service, care, or treatment of others. Maslach and Jackson described burnout along three dimensions: depersonalization, personal accomplishment, and emotional exhaustion. In the original development of the MBI, individuals were given a number (i.e., 47 or 25) of statements and asked for a rating of both frequency and intensity, using seven-point scales. Since that early work, three versions of the MBI have been published: The Human Services Survey (22 items), the Educators Survey (22 items), and the most recent, 16-item General Survey designed for use with workers outside the human service and education professions. All three versions use a seven-point, frequency scale. A fourth, modified version of the MBI (i.e., 23 items, substitutes co-worker for recipient; uses intensity scale) has also gained recent prominence.
In an attempt to compare the most popular measures of burnout, the BM, the MBI-General Survey, and the modified MBI were included as part of a larger Profile used to assess the occupational health of New Zealanders. In total, responses from 1102 kiwi workers in various occupations (e.g., manufacturing, emergency services) were obtained. This study offers both cross-national data as well as a valuable grounding for advancing our discussion on the measurement of the burnout experience.
CORRESPONDING AUTHOR: Bob Boudreau Ph.D., Faculty of Management, The University of Lethbridge, 4401 University Drive, Lethbridge, Alberta, Canada, T1K 3M4
Walking through the Burnout Jungle: A Criticism Perspective
Robert A. Boudreau , PhD., Ralph Stablein, BA, MA, PhD
In 1961, Harold Koontz wrote about the emergence of several schools or approaches (e.g., empirical, human behavior) which he collectively referred to as the management theory jungle. Some two decades later, Koontz revisited the jungle and found it had become even more "dense and impenetrable"; the number of approaches had almost doubled. Apart from any discussions of whether we have been successful in disentangling the jungle of management theory, it appears that the available approaches have continued to multiply at an alarming rate. Building on this highly stylized description, the general field of burnout seems to be experiencing a similar pattern of accelerated growth.
In this poster presentation, we offer a criticism perspective to examine the study and research on burnout. During the past 25 years, several reviews have been proffered as attempts to shape the direction of this fledgling field. These reviews attend to the recapitulation of history, construct definition, measurement, method, and theory considerations. Our criticism perspective serves as a compliment to these more traditional offerings.
Briefly, the perspective suggests that individuals and disciplines can operate on a variety of different levels. We define different levels or types of criticism (e.g., material, method, meta) which reflect research developments as well as influence our world views. Applying this perspective to burnout yields interesting insights. For example, one of the more contentious issues in burnout research these days involves the debate around which process model of burnout is the best. From the perspective to be outlined, the answer to this question will depend on the level of criticism one chooses to examine the question. At a material level of criticism, the model with the best data wins. At a meta-level of criticism, there will never be a winner. Our perspective will allow us to reconcile these seemingly disparate conclusions. In addition to providing a complete account of our criticism approach and the aforementioned model debate, several other issues will be addressing including:
the challenges of postmodern and poststructural paradigms for the field.
the implication of measurement for the representation of the burnout experience.
the challenge of cross cultural and cross-national explanation.
This meta-treatment will not only help us better understand the past, it will guide us through the ever-expanding burnout jungle.
CORRESPONDING AUTHOR: Bob Boudreau, PhD., Faculty of Management, The University of Lethbridge, 4401 University Drive, Lethbridge, AB., Canada, T1K 3M
Women's Health: Occupational and Life Experiences in Canada and New Zealand
Elizabeth Hall, MCom., University of Otago, Karran Thorpe, Ph.D, Jeanette Barsky, R.N., Robert Boudreau, Ph.D., The University of Lethbridge, Alberta, Canada.
In December 1998, 3000 members of the New Zealand Nurses’ Organisation were surveyed to gather data on selected aspects of life and health experiences of nurses as an occupational group. Over 900 responses were received to The New Zealand Occupational and Life Experiences Profile which is the first national survey of this nature. This study replicated a survey of 2,000 Canadian nurses from Alberta (n = 692). Data was gathered on nurses’ health care history, current medical, gynaecological and surgical health status. These results were correlated with respondents’ work profile, including a series of measures on burnout and job involvement.
These studies do not support the assertion that menopause causes many women to experience serious disruptions of their lives and more stress during this time of life. Specifically, there was no relationship between pre, peri or post menopausal classification and burnout phases. Relatedly, only 13% of respondents in both New Zealand and Canada stated that their menstrual cycle affected how much they accomplished in their work. For health care professionals, it seems that New Zealand nurses had a poor record of preventive health care, with 6% never having had a cervical smear test (cf 1.4% Canadian), and only a third reporting that they performed regular self-breast examination (cf 25% Canadian).
Results also indicate a low level of burnout (as measured by the Maslach inventory scale), suggesting that women working in healthcare settings may have found ways of coping with stress. Possible explanations for this are explored.
CORRESPONDING AUTHOR; Liz Hall, Lecturer in Management, Department of Management, University of Otago, Box 56, Dunedin, New Zealand.
Burnout in Law Enforcement: Pilot Data from a Burnout Inventory
Byron E. Greenberg, M.S., M.A., Matt Riggs, Ph.D., Loma Linda University.
Work stress is a complex problem drawing the attention of researchers form fields as diverse as business management, chiropractic medicine, sociology and nursing. Within the issue of job-stress and perhaps at its core is the concept of burnout. This syndrome has been linked to turnover, absenteeism, moral problems (Maslach & Jackson, 1979) as well as psychosomatic, emotional, and physical conditions, poor decision making, poor impulse control and other symptoms related to low work performance (Burke & Greenglass, 1988; Hills & Norvell, 1991; Cherniss, 1980; Maslach, 1993).
Among the high stress professions, law enforcement is listed as second only to mining on the Cooper Occupational Stress Rating (Cooper, Cooper and Eaker, 1988). As a high stress, fast paced profession, it is important to have tools to quickly and effectively assess burnout, the sources of that burnout and appropriate preventative measures to lower the incidence of burnout. Building on the work of Cherniss (1980), Maslach (1981), and Burke (1987, 1997) and benefiting from findings such as those of Walkey &Green (1992) the authors created a burnout inventory for law enforcement professionals. Recognizing that burnout occurs in those who are "on fire" when they begin their profession, the authors intentionally chose subjects at the beginning of their careers. The following study tests a pilot instrument on law enforcement cadets from two academies in southern California.
Exploratory principal axis factor analysis was conducted on responses from 194 subjects. Promax oblique rotation yielded simple structure for a six-item emotional exhaustion scale and an eight-item emotional withdrawal scale. Respective scale alphas were .85 and .75. Factors were significantly correlated (r= .46). Preliminary evidence of external validity includes the finding that emotional exhaustion is somewhat more predictive of aggressive tendencies than is emotional withdrawal.
CORRESPONDING AUTHOR: Byron Greenberg, M.S., M.A., Department of Psychology, Loma Linda University, 11130 Anderson Street, Loma Linda, California 92350.
Work Stress and Resources in the "Process Model of Burnout"
André Buessing*, Ph.D., and Juergen Glaser, Ph.D., Chair of Psychology, Technical University Muenchen, Germany
Recent models about the burnout process (e.g. Leiter, 1993) postulate that work stress triggers physical and emotional exhaustion and that emotional exhaustion mediates work stress on depersonalisation. We performed a study with a sample of 482 nurses from three general hospitals in Germany. It was the primary focus of our study to examine the role of work stressors in the process of burnout more theory-driven. For this purpose we referred to the theory of action regulation and the related concept of regulation problems. Regulation problems are objective stressful conditions of the work situation causing observable negative consequences like additional effort, increased effort and risky actions which in turn can lead to subjective strain like burnout.
The enhanced and modified process model was analysed and strengthened in three respects. First, we found evidence for correlations of work load factors with emotional exhaustion; however, in the centre are not the well-known work load factors, but social and interactional stressors. Second, we found mediating effects. Results of structural equation analyses not only support Leiter's model with respect to the order of emotional exhaustion and depersonalisation in the burnout process but also with regard to two mediating effects: the mediating function of additional, increased effort/risky actions between work stressors and emotional exhaustion, and furthermore the mediating effect of emotional exhaustion between additional, increased effort/risky actions and depersonalisation. Third, we inspected the process model of burnout with regard to the postulate that depersonalisation is largely determined by resources instead of work load and emotional exhaustion. In contrast to the assumption in Leiter's model that autonomy and participation at work directly influence the components of burnout, our results add evidence to the modified model which proposes that activity latitudes exert a direct impact on work stressors and their negative consequences. Thus, a high degree of activity latitudes obviously gives a chance to more easily overcome regulation problems (barriers, obstacles) in the work situation, and therefore to prevent their negative consequences. Moreover, two categories of resources could be distinguished: Work-related resources (e.g. activity latitudes) show a direct impact on work stressors but not on burnout, whereas person-related resources exert a direct impact on burnout.
CORRESPONDENCE AUTHOR: André Buessing, Ph.D., Chair of Psychology, Technical University Muenchen, Lothstr. 17, D-80335 Muenchen, Germany .