Workers' Intention to Leave: Influence of Location and Organizational Factors In Child Welfare

Purpose: This study expands on existing research regarding location and organizational factors' impact on child welfare workers' intention to leave their jobs. Retention of child welfare workers is a critical issue in providing services to families and children (Westbrook, Ellis, & Ellett, 2006). Research has shown that organizational factors, such as workgroup support (Glisson & James, 2002) and job satisfaction (Mor Barak, Nissly, and Levin, 2001), impact workers' decision to leave their jobs. In addition, child welfare workers' location has been shown to contribute to intention to leave. Strolin-Goltzman, Auerbach, McGowan, & McCarthy (2008) found that suburban workers had less reported intention to leave than their rural and urban counterparts.

Methods: Data used in this analysis was from the 2005 Survey of Organizational Excellence that is administered to Texas state agencies on a biennial basis. Data was analyzed from the Child Protective Services division of the Texas Department of Family and Protective Services (N=3,316). Employees working at the statewide hotline and/or the state administrative office were excluded because they do not work directly with clients (n=3,191); 288 participants were excluded due to missing data on key variables (n=2,903).

The dependent variable in the analysis was worker intention to leave the agency within the next two years. The primary independent variables examined were organizational features: work environment, work group support, fair compensation, and job satisfaction. County population was used as a moderating variable. The researchers classified participants into counties based on the predominant location of the participants' residential ZIP codes. Counties were then designated as metropolitan, micropolitan, or non-core as defined by the Office of Management and Budget. Results of this classification process found that 2,480 participants resided in metropolitan counties containing at least one urban area of at least 50,000 people; another 251 participants lived in micropolitan counties, which contain at least one urban cluster of 10,000 but no more than 50,000 people; finally, 172 participants resided in non-core counties, which have no urban clusters.

Findings: Results of an ANOVA analysis revealed no significant differences among child welfare workers residing in metropolitan, micropolitan and non-core counties with regards to their intention to leave. However, results of logistic regression models showed factors predicting intention to leave varied among the groups. While all organizational factors were significant predictors of intention to leave for employees residing in metropolitan counties only job satisfaction was found to be significant for participants living in non-core counties.

Discussion: The findings suggest that caseworkers residing in non-core counties are more likely those in metropolitan and micropolitan counties to remain in their position until they become dissatisfied with their jobs. Unlike caseworkers in more urban counties, organizational factors do not seem to impact the non-core county caseworkers' decision to stay in their job. One likely explanation is the decreased number of career opportunities in their communities. Future research should seek to clarify the role location plays in impacting workers' decisions to leave child welfare organizations and to develop locale-specific interventions to foster organizational factors which promote worker retention.