A Confirmatory Factor Analysis of the Evidence-Based Practice Attitudes Scale in Child Welfare

Background and Purpose: Despite the emphasis on evidence-based practice (EBP) in child welfare services (CWS) as an emerging and inevitable trend, implementation factors such as child welfare worker attitudes towards implementing EBP have received little attention in CWS.  Aaron’s (2004) Evidence Based Practice Attitudes Scale (EBPAS) has been used in multiple studies across mental health and healthcare settings to assess the attitudes of workers towards using EBP. However, it has rarely been used in CWS. The purpose of this study is to examine the EBPAS applied in a large urban child welfare services setting using a confirmatory factor analysis. 

Methods: A-Mid Atlantic State implemented the EBPAS between July 2009 and March 2010 as part of a cross-sectional population survey to all child welfare jurisdictions to measure workers’ attitudes and performance. The response rate for the survey was 44% (n = 848). From the larger statewide survey, this study selected a sample from a large urban CWS jurisdiction (N = 146).  The typical participant was female (79%), African American (64%), between 45 and 54 years in age (45%), a caseworker (50%) and working in foster care (49%).  Based on previous literature, a confirmatory factor analysis on the EBPAS was used to test three models:  a four-factor model, a five-factor model, and a second order four-factor model. Analysis was conducted using Mplus version 7.31. 

Results:  The second order model provided the best fit to the data when considering the fit indices and previous literature suggesting the second order model represents global attitudes of front line staff towards EBP.  Although the χ² value was 173.861 and had a significant p value, it had the lowest score on the RMSEA (0.084) and it also had the highest TLI (0.983) and CFI (0.980). Factor loadings of the second order model were similar to the four and five factor model for requirements, appeal, and openness, and the four-factor model for divergence.  Three of the first order factors (appeal, openness, and requirements) loaded significantly on the second order attitudes factor.  However, divergence did not load significantly on the second order attitudes factor. 

Conclusions and Implications: The findings suggesting the EBPAS can measure workers’ global attitudes, as well as four attitude types, towards EBP in a CWS setting. This study validates the use of the EBPAS with child welfare workers and addresses the gap in knowledge validating the EBPAS with child welfare workers in a large urban CWS under a consent decree.  The EBPAS has the potential to be a useful tool in understanding attitudes of workers in CWS and how they perceive implementation of EBP.  The EBPAS may be an effective tool for CWS administrators to assess and incorporate child welfare worker attitudes into the implementation process.  Future research is needed to understand the complex interaction of policy, organizational factors, and practice when implementing EBP in CWS.  Implementation science can be used in CWS to help navigate the multiple levels of influence on child welfare workers’ attitudes towards EBP.