Differences between new and established industrial workers: an interactional model

Abstract

The main purpose of this study was to investigate a model which incorporated personality/organizational (esteem, job involvement, rigidity, support, structure and complexity), "psychological climate" (ambiguity, challenge and evaluation) and outcome measures (satisfaction, anxiety and performance). Under this model it was hypothesized that the psychological climate variables would mediate between personality/organizational and outcome measures. Additionally, it was expected that differences in regression weights would be found between new and established workers. Finally, it was predicted that the results would generalize from one sample to another. The original sample consisted of four hundred and twenty-three manufacturing and insurance workers (66 new and 367 established workers), whereas the cross-validation sample contained 220 mental health workers (72 new and 148 established workers). Hierarchical regression analyses were used to find the regression weights for new and established workers. A subgroup moderator analysis was employed to test differences in regression weights between new and established workers. LISREL and a comparison of variances in the climate and outcome variables served as a test of the equivalence of findings for the two samples. Among the three outcome variables (satisfaction, anxiety and performance), the results supported the mediational hypothesis only for satisfaction. For anxiety, the results for the new workers were confounded by "suppressor" effects. For established workers, it appeared that the personality/organizational and climate variables acted independently in their influence upon anxiety. Strong direct effects due to the personality and organizational variables were evident for the performance measure. No differences in regression weights were found between new and established workers for the two samples. The lack of differences was attributed in part to the small number of new workers in either sample. Finally, the results generalized from one sample to the other. It appears that the "psychological climate" model is operative for a "cognitive" variable such as satisfaction. In the case of anxiety and performance, it was conjectured that the determinants of these outcome were of such a nature to preclude the mediational effects by the climate variables. Although this study was exploratory in nature, the generalization of the results to another sample speaks to its strength.