The

Relationship Between Job Security and Employee Health

KARL W. KUHNERT

University of Georgia RONALD R. SIMS College of William and Mary MARY ANNE LAHEY

University of Georgia The

relationship between perceived job security and employee health was examined in two manufacturing organizations. A significant relationship was found between health and job security using global measures of both constructs. Closer examination of the correlations among the subscales of the measures of health and perceived job security revealed different patterns of subscale relationships in two organizations. These results suggest that job security is an important determinant of employee health, and that inclusion of this construct in general models of stress and well-being may help to improve our understanding of work performance.

Since Kornhauser’s (1965) pioneering book, Mental Health of the Industriall Worker, there has been a slowly growing list of researchers who have studied the deleterious effects of employment conditions on the mental and physical health of workers (e.g., Beehr & Newman, 1978; Frese, 1985; Ivancevich, 1986; Ivancevich & Matteson, 1980; Warr, 1987; Weitz, 1970). By far the most popular approach employed to date is to identify features of the work environment that serve as potential stressors, and to relate those features to one or more measures of employee health (e.g., Beehr & Newman, 1978; Cooper & Marshall, 1976). Factors such as shift work, role ambiguity, poor communications, conflicts between work groups, and technological change have all been pinpointed as potential stressors, and have been related to measures of employees’ physiological health (e.g., coronary heart disease and arteriosclerosis), psychological health (e.g., depression and anxiety), and indices of societal health (e.g., alcohol and drug abuse; Kahn, 1981). One construct that is gaining increasing importance to organizations and its members, but for the most part has been ignored by social scientists, is job security (Greenhalgh & Rosenblatt, 1984). Layoffs due to increases in Group & Organizalion Studies, Vol © 1989 Sage Publications, Inc

14 No

4, December 1989 399-4100

399

400

automation, plant closures, and corporate takeovers have had unsettling effects on U.S. workers and have created anxieties about the future of workers’ jobs. Perhaps at greatest risk in an insecure workplace is the blue-collar worker. Manufacturing, the leading blue-collar stronghold, continues to show declines in its work force, and it is clear that those workers must confront and cope with an insecure work life. Although research on the consequences of job insecurity is sparse, the research that has been done suggests that there are ill-health effects when workers perceive their jobs are in jeopardy. Kornhauser’s (1965) study, for example, examined the psychological condition of workers in modern society and reported that concern about job security is &dquo;expressed by equal proportion of men at all skill levels,&dquo; and that feelings of job security are &dquo;related to the mental health of workers&dquo; (p. 94). More recently, London, Crandall, and Seals (1977) found a positive relationship between workers’ satisfaction with the amount of security in their job and their overall quality of life. In addition, Jick (1980) reported that workers experiencing high levels of job insecurity were more likely to experience severe psychosomatic reactions such as depression, anxiety, and irritation than their more secure colleagues. While the research reported suggests a link between job security and health, there is no in-depth examination of the ways in which job security affects health. To our knowledge, no studies have taken a systematic approach to the study of job security and employee health, to some extent due to the absence of an adequate measure of the job security construct.

The study reported here examined the impact of perceived job security on employee health among workers in two manufacturing organizations, using a newly-developed multidimensional scale of perceived job security (Lahey & Kuhnert, 1988). It was hypothesized that perceived job security would be directly related to self-report of health symptoms. Based on previous research on job security (Jick, 1980), it was predicted that anxiety and depression, two dimensions of the health measure employed here, would account for significant variation in the job security construct.

METHOD SUBJECTS A total of 201 employees in two manufacturing organizations in the northeastern United States participated in this study. Organization 1 (n = 98)

401

is

a manufacturer of aluminum siding and aluminum storage buildings, and Organization 2 (n 104) is a manufacturer of ball bearings and chains for motorcycles and other small machinery. The majority of respondents were line workers, and approximately half of the respondents from both organiza=

tions were female. MEASURES

Job security. The Job Security Survey (JSS, Lahey, 1984; Lahey & Kuhnert, 1988) used here is a 44-item measure comprising five subscales: (a) company concern for the individual, (b) job permanence, (c) job perfor-

(d) company growth and stability, and (d) long-range perspective. Subscale definitions and sample items appear in the Appendix. Items are rated on a 5-point scale with higher values being associated with increased security. A total JSS score is obtained by summing scores on the subscales. Health. Employee health was measured using the SCL-90-R (Derogatis, Lipman, & Covi, 1973) designed to measure specific symptom dimensions. This measure has been widely used in epidemiological research and is sensitive to low levels of psychological distress in the general population (Davidson & Baum, 1986). Four of the nine subscales from the SCL-90-R somatization, depression, anxiety, and hostility-were used in this study (see Appendix). Symptom responses are made on a 5-point scale ranging from 1 (not at all bothered by the symptoms) to 5 (extremely bothered by the symptoms). Thus higher scores on each subscale indicate poorer health of the participants. A total health score is obtained by summing items on the subscales. mance,

RESULTS The

reliability (coefficient alpha) estimates for the JSS and SCL-90-R quite high, .90 and .95, respectively, indicating good internal consistency in the measures. The correlation between the job security and health total scores was significant (-.25 p < .01), indicating that employees who reported lower levels of job security also reported more symptoms of ill health. Additional analyses to clarify this relationship were performed. ANOVA comparison of the JSS and SCL-90-R total scores (Table 1) indicated that there were no significant differences in overall ratings of health and job security in the two organizations. Although the total scores were not different, MANOVAs were used to test for organizational differences on the measures were

402

TABLE 1 Means and Standard Deviations for Job

*p