|
Article Excerpt
The study of occupational stress has become one of the major areas of organizational research. The majority of research in the area utilizes single-source, incumbent reports that limit conclusions about the objective environment. Most studies that have utilized additional (non-incumbent) sources of data have relied on co-worker, observer, or supervisor reports, but these sources are not independent of the incumbent. Thus, it is possible that the co-workers, observers, or supervisors were influenced not only by the objective features of jobs, but by the apparent levels of strains in the incumbents being rated (Spector & Jex, 1991).
Some researchers have attempted to circumvent the problems of non-independence by assessing job conditions objectively with existing job analysis databases, such as the US Department of Labor's (1977) Dictionary of Occupational Titles (DOT; Adelmann, 1987; Murphy, Thornton, & Prue, 1991; Spector & Jex, 1991) and the McCormick, Jeanneret, and Mecham's (1972) Positional Analysis Questionnaire (PAQ; Murphy et al., 1991; Shaw & Riskind, 1983; Voicu & Nereuta, 1985). In these studies, participants' job titles are matched to corresponding job categories in the database. This study represents an attempt to use the new Occupational Information Network (O*NET; US Department of Labor, 1998) as a source of objective data in job stress study, reanalysing data that was used in Spector and Jex's (1991) paper. The O*NET database provides new objective information on job control not available when the original study was conducted. We linked these data to employee reports of autonomy, as well as health and well-being.
Job stressors and control
Job control refers to individual's ability to choose his/her own actions from two or more options (Ganster & Fusilier, 1989). According to Spector (1998), control in the workplace ranges from autonomy (control over the individual's own immediate scheduling and tasks), to participation in decision-making process (control over the organizational decision-making process). Autonomy is an important aspect of the broader construct of control. With autonomous jobs, incumbents can determine the order and pacing of job tasks, specific procedures for accomplishing those tasks, scheduling, and coordination with other employees and other conditions of work (Spector, 1986). A number of studies and theories link job control to both psychological well-being, physical health, and behaviour, although most studies utilized self-reports of control (Evans & Carrere, 1991; Ganster & Fusilier, 1989; Spector, 1986, 1998).
Empirical studies using self-report job control
A link between perceived control and employees' psychological, physical, and behavioural outcomes has been established. Spector (1986) conducted a meta-analysis on 101 control papers to study the effect of perceived autonomy on 19 outcome variables. He found that perceived autonomy was positively associated with job satisfaction (overall and individual facets), commitment, involvement, performance and motivation, and negatively associated with physical symptoms, emotional distress, role stress, absenteeism, turnover intention, and actual turnover. In addition, studies have revealed that perceived control is negatively associated with depression (Karasek, 1979), anxiety (Kohn, 1969; Kohn & Schooler, 1973; Kornhauser, 1965; Miller, Schooler, Kohn, & Miller, 1979), neurotic illness (Kornhauser, 1965), stomach trouble, sleep problems, headache, and heart symptoms (Hurrell & Lindstrom, 1992).
Empirical studies using objective assessments of job control
Because of the problems introduced by relying entirely on self-report data, a number of researchers have discussed the need to assess job stressors objectively (e.g. Aldag, Barr, & Brief, 1981; Loher, Noe, Moeller, & Fitzgerald, 1985; Roberts & Glick, 1981; Spector & Jex, 1991). These include obtaining job stressor data from supervisor-reports, peer-reports, rater-reports, and job analysis databases. However, results are inconsistent and the picture is less clear.
Spector and Jex (1991) studied job autonomy from three independent sources: incumbents, job raters, and the DOT database. They found that incumbent-autonomy was significantly related to job satisfaction, frustration, anxiety, physical symptoms, doctor visits and intent to quit. However, the two objectively measured autonomy (rater-autonomy and DOT-control), were not associated with most of the employee outcomes except for the relation between rater-autonomy and doctor visits. Stansfeld, North, White, and Marmot (1995) investigated the relations between work characteristics and psychiatric disorders, well-being, and global satisfaction in civil servants in London. Similarly, they obtained both subjective--(self-reports) and objective-control (personnel managers' assessments). While high levels of perceived control were related to less psychiatric disorder, greater well-being, and greater global satisfaction in both men and women, the personnel managers' assessments of control were not associated with most of the outcomes except that greater control overwork was significantly associated with lower general health in men and higher global satisfaction in women (Stansfeld et al., 1995).
Shaw and Riskind (1983), using the PAQ dimension of making decisions, found a significant relation with job satisfaction and illness, and Adelmann (1987) using the DOT derived job control score found relations with general happiness. Some researchers...
|
