Predictive Validity of Mathematics-Curriculum Based Measurement

Although mathematical competency is associated with life skills and economic success, research in student mathematical performance offers less substantial evidence when compared to investigations of student performance in reading. Empirical evidence to support the predictive validity of Mathematics-Curriculum Based Measurement (M-CBM) to identify accurately learners at-risk for low academic performance is a critical research area. This study analyzed correlations between M-CBM benchmark scores and student performance on the Tennessee Comprehensive Assessment Program (TCAP), a state mandated high-stakes test (i.e., related to federal and local funding, student placement decisions, teacher tenure, etc.). Participants were 1,732 students enrolled in grades 3-8 in a rural southeastern school system. Linear regression models were used to investigate the research questions. Specifically, this study sought to determine to what degree the Fall, Winter, and Spring M-CBM scores are correlated with the TCAP results. Additionally, the authors sought to assess the validity of M-CBM benchmarks from three time points for predicting TCAP performance at the various grade levels. Lastly, this study sought to identify any temporal differences in M-CBM and TCAP correlations based upon the time of the benchmarking (i.e., Fall, Winter, and Spring).