A confirmatory structural equations analysis of the job characteristics model

Abstract

K. H. Roberts and W. Glick (1981, Journal of Applied Psychology, 66, 193–217) noted that research on J. R. Hackman and G. R. Oldham's (1976, Organizational Behavior and Human Performance, 16, 250–279) job characteristics model has lacked consistency along a number of dimensions, including the nature and direction of the relationships hypothesized and the variables studied. Further, no variation on the model has been tested as a whole. This study was designed with two objectives: (1) to compare competing interpretations of the job characteristics model and (2) to test the job characteristics model more comprehensively than has been done in previous research. The question of which variation on the model is most useful has important implications, given the widespread use of job characteristics theory to guide productivity and satisfaction improvement efforts. Structural equations analysis can yield information to compare competing variations of the job characteristics model, and allow assessment of each interpretation's ability to describe data. Such analyses were performed on data collected by survey from 208 white-collar government employees. Findings indicate that a single-factor model, with that factor interpretable either as common method variance or generalized affect, explains these data at least as well as any traditional variation on job characteristics theory. The results are discussed in terms of further research on job characteristics, satisfaction, and performance.