Refining the evidence-based practice attitude scale: An alternative confirmatory factor analysis

Barriers to adopting evidence-based practices into real-world mental health organizations have received considerable attention and study. One particular attempt is Aarons's Evidence-Based Practice Attitude Scale (EBPAS), which measures a worker's attitudes toward adopting new treatments, interventions, and practices. This study follows Aarons's work by conducting a confirmatory factor analysis of the EBPAS administered in a large child and family human service agency in New York state (N = 1,273). Replicating Aarons et al.'s four-factor model of the EBPAS, the authors found that, within the model, the pattern of factor loadings that was apparent in previous investigations held for their data as well. That is, the factor loadings of items within the Divergence subscale were larger for items 5 and 7 and smaller for items 3 and 6. The authors found that both of their alternative models, one that added a residual covariance to items in the Divergence factor and a five-factor model that divided the Divergence factor into two factors, fit their data better than Aarons et al.'s model. They also investigated measurement and structural invariance for workers in community-based and in residential programs using a multiple group analysis. Measurement invariance was supported but factor means and correlations differed.