Education Commons^™

Using Reading Cbm To Predict Performance On Smarter Balanced Assessment, Jonathan Wesley Shank

Theses, Dissertations and Capstones

This study examined the relationship between AIMSweb oral reading fluency (R-CBM) and reading comprehension (MAZE) curriculum-based measures and performance on the English language arts/literacy (ELA/L) component of the Smarter Balanced Assessment (SBA) using a sample of students in third through fifth grade (N = 499). Pearson correlations between R-CBM, MAZE, and SBA were moderate to high, with R-CBM generally demonstrating the strongest relationships with coefficients ranging from .73 to .75. Results from hierarchical multiple regression models indicated that R-CBM provided strong predictive validity for SBA performance among third grade students (63.4% variance explained, p<.001), while the addition of MAZE to the equation was negligible (1.4% additional variance explained, p<.001). Similar findings resulted from the fourth and fifth grade multiple regression models. The predictive value of R-CBM and MAZE each decreased as grade level increased. Results support continued use of CBM to predict success on the Smarter Balanced Assessment, although CBM using cloze passages explained little variance in high-stakes test scores beyond that of oral reading fluency alone.