Turnover in Public Child Welfare: A Different Perspective

Purpose: Child welfare agencies across the country are experiencing a workforce crisis involving high staff turnover rates. Turnover in the child welfare workforce has strong implications for the well being of children in care (Annie E. Casey Foundation, 2003; Drake & Yadama, 1996; US General Accounting Office, 2003). The purpose of this study was 1) to determine which of the organizational and supervisory variables identified in prior research on this topic are most associated with intent to leave among employees in urban and rural public child welfare settings; and 2) to analyze how workers' career satisfaction and perceptions of the quality of their work are related to intent to leave.

Method: Four hundred and forty-seven employees in New York State responded to the survey resulting in a response rate of 68 percent. The study includes 12 rural agencies (n=229), and one large urban agency (n=218). The presence of 3 latent constructs: work satisfaction; career satisfaction; and intention to leave, were tested for. Various aspects of work satisfaction (supervision, benefits, overall satisfaction, etc.) were tested in the model. The only type of satisfaction that fit the model was satisfaction with paper work. Career satisfaction was measured by two variables: “ I would make the same career decision” and “Was casework your first choice? For intent to leave, the variables examined are “I have thought about leaving in the past year;” I have thought about leaving; “I spoke to my spouse about leaving;” “I made phone inquires;” and “I went on interviews.”

Results: The organizational and supervisory factors found significant in our earlier research and in the findings of a number of other studies were not significant when we subjected the data to structural equation modeling (SEM). Instead what emerged as most important was what we have termed career satisfaction, as well as dissatisfaction with paperwork. In previous studies on this data, “Have you considered looking for a new job within the last year” was the major outcome measure of intent to leave. In the current study, the measure intent to leave was combined with the additional items mentioned above. This may account for part of the reason why the organizational and supervisory factors factors are not significant in the SEM. The model was run separately for urban and rural counties. The hypothesis that the latent construct career satisfaction is associated to the latent construct of intention to leave is confirmed by the standardized parameter estimates (.64 and .69) between these concepts.

Implications: The results of this study suggest two important implications for child welfare administrators. One is find ways to determine during the hiring process the degree to which job applicants have a career interest in child welfare. Such a question is not usually explored in Civil Service examinations. Yet the findings of this study suggest it may be a critical factor to examine when hiring child welfare workers. Second, it is essential that administrators find ways to reduce the intolerable recording burdens on child welfare workers.