Content area
Full Text
Besides being one of the most ubiquitous constructs studied in organizations, job performance also plays an important role in the voluntary turnover process. Yet, despite being studied in conjunction with turnover sporadically since 1949 ([7] Giese and Ruter, 1949), job performance did not appear in prominent models of voluntary turnover until the early 1980s when [36] Steers and Mowday (1981) pointed out that performance on the job was an often overlooked factor in influencing turnover decisions. Since then, the job performance-turnover relationship has been empirically demonstrated in numerous primary studies. In fact, four separate meta-analyses have shown that job performance is negatively related to voluntary turnover (ρ =-0.25, [5] Bycio et al. , 1990; ρ =-0.17, [8] Griffeth et al. , 2000; ρ =-0.28, [25] McEvoy and Cascio, 1987; ρ =-0.31, [45] Williams and Livingstone, 1994).
However, while prior meta-analyses have found support for the relationship between performance and turnover, there has been no such meta-analysis involving job performance and intent to quit. In addition, while previous research has theorized and tested process models of the relationship between job performance and turnover (e.g. [3] Birnbaum and Somers, 1993; [17] Jackofsky, 1984; [21] Keller, 1984), none have tested a model that focused on the direct and indirect effects that job performance has on turnover intentions and behaviors using meta-analytic estimates of the relationships. Testing such a model is important because, as noted by [6] Fishbein and Ajzen (1975), intentions toward an act or behavior are distinct from that act or behavior itself. That is, even though an employee may intend to quit, they may not actually do so. In fact, based on the meta-analytic data available, intentions to quit have a stronger relationship with work-related attitudes, such as job satisfaction (-58, [39] Tett and Meyer, 1993), than between intent to quit and turnover itself (0.45, [8] Griffeth et al. , 2000). Therefore the effects of job performance on turnover decisions may not be fully mediated by turnover intentions or job satisfaction. Given that the goal of scientific research is to investigate the causal relationships among variables ([15] Hunter and Gerbing, 1982), it is important to understand how all of pertinent constructs that comprise the traditional withdrawal model of turnover are related. Thus, the aim...