The Benchmarking Capacity of a General Outcome Measure of Academic Language in Science and Social Studies

Abstract

The amount of research evaluating the technical merits of general outcome measures of science and social studies achievement is growing. This study targeted criterion validity for critical content monitoring. Questions addressed the concurrent criterion validity of alternate presentation formats of critical content monitoring and the measure’s predictive validity. Participants were fifth-grade students (N = 51) who completed five different forms of critical content monitoring probes as well as oral reading fluency and maze probes over three benchmarking periods. Criterion measures were the science and social studies subtests of the online abbreviated Stanford Achievement Test–10th edition. Concurrent correlation magnitudes for critical content monitoring ranged from .47 to .60. Predictive correlations for fall and winter ranged from .23 to .64. In three of four cases, commonality analyses findings favored critical content monitoring over oral reading fluency and maze as benchmarking choices. Study limitations and benchmark assessment framework implications are discussed.