Application of Algebra Curriculum-Based Measurements for Decision Making in Middle and High School

Abstract

This article reports the results of a study examining the utility of curriculum-based measurement (CBM) in algebra for predicting performance on a state math assessment and informing instructional placement decisions for students in seventh, eighth, and tenth grades. Students completed six Basic Skills algebra probes across different time schedules and took their respective grade-level state math assessment. Using a dichotomous pass/fail variable based on whether students met grade-level proficiency, logistic regression and receiver operating characteristic (ROC) curve analyses were conducted to assess the CBM’s predictive validity. Rates of improvement (ROIs) were computed using the Tukey method to collect initial information on whether these measures would help monitor student progress. Finally, to determine whether these measures might inform instructional programming, a MANOVA using course assignment as the grouping variable and CBM, ROI, and state assessment performance as the dependent variables was conducted for the tenth-grade students. Results indicate that these measures show promise to predict which students might not meet grade-level proficiency standards as well as to determine whether students are progressing in the curriculum. Implications for practice are discussed.