Generalizability Theory Analysis of CBM Maze Reliability in Third- Through Fifth-Grade Students

Abstract

Despite growing use of CBM maze in universal screening and research, little information is available regarding the number of CBM maze probes needed for reliable decisions. The current study extends existing research on the technical adequacy of CBM maze by investigating the number of probes and assessment durations (1-3 minutes) needed for reliable relative (e.g., rank-ordering students) and absolute (e.g., comparing a specific score to a cutoff) decisions. Nine CBM maze probes were administered to 272 students in third through fifth grades. Results suggested that the number of probes needed for reliable relative and absolute decisions varied by grade, with assessments in fifth grade exhibiting the highest reliability (at least two probes needed for both types of decisions). In addition, declining gains in reliability appeared to occur as assessment duration increased. Implications of the findings for universal screening and future research are discussed.