Research That Matters (January 17 - 20, 2008)

Purpose: This study analyzes the differences in the causes of public child welfare workforce turnover in urban and rural contexts. Although there is an emerging understanding of the individual, organizational and supervisory factors that cause workforce turnover and retention (Dickinson & Perry, 2002; Ellett & Ellett, 2004; Lawson et al., 2005; Strolin et al., in press; Zlotnik, DePanfilis, Daining, & McDermott Lane, 2005), there is a lack of contextual understanding. In other words, there is a gap in the current literature about how the relationship between causal factors and intention to leave differs depending on child welfare agency locality. Without knowledge about system context there may be misunderstanding about which interventions will be most successful in a specific agency. For instance, is a training initiative currently being tested in rural agencies going to be effective in larger urban systems? The current study adds to the literature by comparing the causal factors of turnover in rural and urban public child welfare agencies.

Method: Eight hundred and twenty workers and supervisors from twenty-five child welfare agencies in rural and urban localities participated in a survey addressing supervisory and organizational factors related to workforce recruitment, retention, and turnover. Two types of analysis were performed. T-tests were performed to assess the mean difference between the two cohorts on clarity and coherence of practice, Work-life fit, Efficacy and job satisfaction, Job supports and commitment, Technology and training, and Salary and benefits, supervisory support, and supervisory competence. Two logistic regressions models were conducted to assess the moderating impact of agency locality on the relationship between the above organizational and supervisory factors and intention to leave.

Results: Mean scores on three of the six organizational factors found to influence turnover in the literature were significantly different in rural and urban localities. The largest difference was found in satisfaction with salary and benefits (p<.001), followed by job commitment (p<.0001), and clarity and coherence of practice (p<.03). As hypothesized, logistic regression results indicated that causes of turnover differed among locality. Burnout and lack of life work fit were causes of turnover in dissatisfaction urban agencies, while lack of commitment to the agency and with salary emerged as causes of turnover in the rural context. Job efficacy and clear job expectations were significantly correlated to intention to leave in both urban and rural agencies.

Implications for Practice: In order to develop effective interventions to decrease turnover and improve child welfare practice, a deeper understanding of how context affects causal factors is essential. Administrators in urban settings may benefit from focusing on improving life-work fit issues while rural administrators may benefit from focusing their attention on creative ways of increasing salary or improving worker commitment to the agency. The evidence emerging from this study suggests that focusing attention improving job efficacy and clarifying job expectations may benefit agencies in rural and urban contexts. Hopefully, these findings will inform the development of innovative locality-specific organizational interventions aimed at enhancing retention and improving child welfare practice.