Quantifying Context in Assessment: Capturing the Effect of Base Rates on Teacher Referral and a Problem-Solving Model of Identification

Abstract

The purpose of this article was to examine the effect of base rate occurrence of race, sex, and student achievement on the accuracy of a problem-solving model of assessment and teacher referral. All students in first and second grade (n = 182) at a participating school were exposed to four screening measures. Students who performed poorly on at least one of the screening measures participated in a more thorough assessment process including individual curriculum-based assessment with individual intervention. The predictive accuracy (e.g., sensitivity, specificity) of the problem-solving model and teacher referral was examined using the more thorough assessment process as the standard for comparison. The degree to which the problem-solving model and teacher referral correctly identified male and female children, children of minority and nonminority ethnicity, and children in high-achieving and low-achieving classrooms was specified. Further, the efficiency of the problem-solving model and teacher referral was examined by comparing identification relative to base rate occurrence of problems in the screened population.