Establishing new relationships within a traditional voluntary turnover model: The effects of a multidimensional approach to actual ease of movement in the job market.

摘要

This study was designed to contribute to individual voluntary turnover research by adding conceptual and methodological precision to the notion of actual ease of movement in the job market (AEOM). I suggested that a three component common model of voluntary turnover has emerged across turnover research, with voluntary turnover being a function of AEOM and job satisfaction. In particular, job satisfaction's well documented negative effect on voluntary turnover was hypothesized to be moderated by AEOM, with the effect being greater when AEOM was high. However, in contrast to the typical operationalization of AEOM as a single type of unemployment rate, I proposed that incorporating individual and occupational aspects of the construct could substantially improve the common model. Data for the study consisted predominantly of responses from the National Longitudinal Survey of Youth (NLSY), although it was augmented by occupational wage data and detailed unemployment rates obtained from the Bureau of Labor Statistics. In all, 5,527 individuals employed between 1980 and 1992 were included in the final data set. Longitudinal data allowed for the conduction of proportional hazard models with repeated events and time dependent covariates. Results of the analyses supported the use of a multidimensional approach to AEOM. An occupationally driven type of unemployment rate (termed the relevant unemployment rate) derived from local and occupational rates appeared to not only moderate job dissatisfaction's effect on turnover, but also to be an approach somewhat superior to using either rate alone. Additionally, individual and occupational level AEOM indicators mirrored the relevant unemployment rate effect, suggesting a synthesis of the common model framework, whose roots lay in March and Simon's (1958) seminal work, with the human capital approach. Finally, I found evidence of hypothesized within-AEOM interactions, as relevant unemployment rate's negative effect on turnover was greater for those with lower individual and occupational AEOM. Theoretical and practical implications were discussed, as was the importance of adequate controls in occupationally heterogeneous turnover research.