The effect of student race/ethnicity on the diagnostic accuracy of the Social, Academic, and Emotional Behavior Risk Screener (SAEBRS)

Abstract

Universal screening has emerged as one strategy within schools for identifying students who demonstrate social emotional and behavioral (SEB) concerns to provide appropriate support. However, research has demonstrated little evidence of equity around diagnostic accuracy; that is, how students might be differentially identified as at-risk for SEB concerns based on race/ethnicity. The current study examined the diagnostic accuracy of the Social, Academic, and Emotional Behavior Risk Screener (SAEBRS) teacher screening measure as affected by student race/ethnicity, drawing from a sample of 730 students and 54 teachers from four urban, Midwestern elementary schools. Data were analyzed using nonparametric receiver operating characteristic (ROC) curve analyses, as well as ROC regression tests incorporating student race/ethnicity status as a covariate. Results indicated the SAEBRS (1) demonstrated borderline to optimal sensitivity and specificity across student racial/ethnic subgroups in predicting student SEB risk status, and (2) diagnostic accuracy did not vary across racial/ethnic subgroups to a statistically significant degree. Results highlight the need for further research on this topic.