Predictor Importance of Job Attributes to Burnout in Child Welfare Worker

Background: Child welfare workforce researchers have studied the concept of burnout for over 40 years and identified many factors contributing to work-related burnout. Common factors include caseload size; the difficulty of cases; time pressure; role ambiguity within agencies; logistical stressors; supervisory support and satisfaction; and workers’ demographic factors. This study used a dominance analysis to test the predictor importance of such job characteristics on work-related burnout in frontline child welfare workers who had been employed for approximately 18 months.

Methods: This study used data from the Florida Study of Professionals for Safe Families, a longitudinal cohort study of newly-hired child welfare workers (N=1,500), recruited between September 2015 and December 2016. Baseline survey data was collected during pre-service training with follow-up surveys every 6 months for 3.5 years. For the current study, eligible participants included frontline child welfare workers at Wave 4 (n=618). Independent variables included validated scales of job attributes commonly associated with burnout (e.g., pay satisfaction, time pressure, caseload difficulty, supervisor support, organizational support). Caseworker demographics and role (investigations or dependency) were included as control variables. The first step of the dominance analysis required ordinary least squares (OLS) modeling to determine model significance. Next, the dominance analysis compared all pairs of predictors across all subsets of the predictors in the model to determine each predictor’s unique contribution to the model.

Results: Most workers (n=531, 85.8%) reported at least one symptom of work-related burnout. The OLS model explained 74.0% of the variance (R²=0.74; F_{(17, 601)}=104.5, p=.000) in work-related burnout. When controlling for worker demographics and role, pay satisfaction (β=0.49, t=8.34, p<.001), time pressure (β=0.28, t=7.37, p<.001), proportion of difficult cases on a caseload (β=-0.07, t=-2.98, p<.01), role ambiguity (β=0.36, t=2.09, p<.05), organizational support (β=-0.41, t=-2.34, p<.05), supervisor inattention (β=0.07, t=2.03, p<.05), psychological distress (β=0.15, t=2.18, p<.05), stress (β=-0.24, t=-2.66, p<.05), and secondary traumatic stress avoidance symptoms (β=0.37, t=2.19, p<.05) were statistically significant predictors of work-related burnout. The dominance analysis results produced the rank order of predictor importance across 17 model subsets, based on the average additional contribution of each predictor to the variance in burnout. The rank order and R² associated values were: 1.pay satisfaction (0.10), 2.time pressure (0.08), 3.coworker support (0.07), 4.supervisor support (0.07), 5.stress (0.07), 6.psychological distress (0.06), 7.role ambiguity (0.06), 8.organizational support (0.06), 9.supervisor inattention (0.04), 10.secondary traumatic stress symptoms (0.04), and 11.percentage of difficult cases (0.02).

Conclusions and Implications: Results indicated that child welfare workers perceptions of pay satisfaction was ranked the most important predictor of work-related burnout compared to the other job attributes included in the model. Thus, workers’ pay satisfaction may serve as a protective factor for workers’ feelings of burnout in relation to their other job demands. Agency leadership should consider equitable and fair compensation models for their frontline staff to demonstrate valuing of frontline workers’ symptoms of burnout and increase worker retention.