This dissertation explored the job embeddedness model of turnover in a collectivistic country (India).The job embeddedness model (JE) by Mitchell and Lee (2001) has 6 original dimensions - organization links and community links (individual connections with people in the organization and community), organization fit and community fit (individual perception of fit within an organization and community), and organization sacrifice and community sacrifice (what the individual gives up when leaving the organization or community). JE has been found to explain variance in turnover above the most significant predictors, such as job satisfaction and job alternatives in the US, but has not been explored in collectivistic cultures.

This dissertation took a two-step approach to testing and extending the JE model to India. First, I explored the generalizability of the JE model in India and applied the individualism-collectivism framework to posit differences in how strongly each dimensions of JE relates to turnover in the US and in India. I suggested that organization links, community links, and organization fit are more important predictors of turnover in India than in the US, while community fit is a more important predictor of turnover in the US. In addition, I examined fit with job and suggested that perception of job fit is a more important predictor of turnover in the US than in India. Second, I expanded the job embeddedness model to include a family factor by creating three new dimensions, family links, family fit, and family sacrifice, and suggested that this factor would predict turnover in both countries.

Data were collected from call center employees in the US (n = 323) and in India (n = 474). Multi-group confirmatory factor analysis supported the three-factor structure of job embeddedness (organization, community, and family factors) in both cultures. As hypothesized, organization embeddedness and family embeddedness predicted turnover in both countries. Community embeddedness did not predict turnover in either country. In addition, organization fit, organization links, and community links interacted with country in the hypothesized direction such that they were more important in predicting turnover in India, while job fit was more important in predicting turnover in the US.