LAUSR

Purpose There is an absence of research examining job embeddedness in SMEs. Results of job embeddedness studies may not apply to SMEs, because the process of managing a SME differs from that of the large firm. The purpose of this paper is to examine relationships between on-the-job embeddedness, as well as each of its sub-dimensions, and turnover intentions; and group cohesion, on-the-job embeddedness and turnover intentions. Design/methodology/approach Data were collected from 147 employees in SMEs located in Perth, Western Australia and 350 employees from SMEs operating in four business centres in South Africa. After invariance testing, data from the two countries were combined to increase statistical power of the analysis. Findings On-the-job embeddedness and each sub-dimension were negatively related to turnover intentions. Group cohesion was positively related to composite on-the-job embeddedness. Findings suggest that while group cohesion on its own does not reduce turnover intentions, it does contribute to development of on-the-job embeddedness that, in turn, reduces turnover intentions. Research limitations/implications Future research should control for the effects of external influences on turnover intentions. Findings imply that managerial actions related to antecedents of group cohesion could foster the on-the-job embeddedness of employees. Originality/value This study is perhaps the first that tests the operation of on-the-job embeddedness in SMEs located in two countries. The conceptual arguments for links between each of the sub-dimensions of on-the-job embeddedness and turnover intentions are based on distinctive characteristics of SMEs and can serve as a theoretical foundation for future research on embeddedness in SMEs.