The purpose of this paper is to develop and test two control models (Direct and Indirect) as indicated in the graphic illustrations below. Each model includes three components (i.e., participative standard setting, standard-based incentives, and standard tightness), and job performance as the dependent variable. The direct model assumes that the control system directly affects job performance, while the indirect model assumes that the control system affects job performance indirectly through job-related stress. The main implication of the direct model is that performance can be increased by tightening the performance standard. However, the indirect model predicts the opposite effect, i.e., decreasing the standard, decreases job-related stress and subsequently increases performance.

The hypotheses shown in the graphic illustrations are as follows:

Questionnaires were distributed to 480 automobile design engineers. This produced 358 usable responses. Each item was measured with multiple indicators. For example, participative standard setting was measured with eight items. Structural equation modeling (LISREL 8) was used to estimate the parameters and to test the hypotheses (see the adaptations of Figures 1 and 2).

The results indicate that the indirect model had a better overall fit to the data based on two indices, provided a better fit to the data than the direct model in the comparison test, and supported the five hypotheses applicable to the model as indicated in the graphic above. Two of the hypotheses in the direct model were not supported.

Overall the results support the view that standard based incentives and standard tightness are influenced by the level of subordinate participation in standard setting. In addition, the control system components affect job-performance indirectly through job-related stress. The implication of the indirect model is that performance is higher when performance standards are lower.

	Web maaw.info