Though IRT-based reliability metrics (standard error of ability estimates, simulated test-retest reliability) only predicted a minor decrease in reliability, the empirical test-retest reliability estimates are much lower than observed in the laboratory-based Study 3, with the correlation between ability estimates across the two test waves only marginally significant after 25 items.
← all excerpts
Development and Validation of the Computerised Adaptive Beat Alignment Test (CA-BAT).
1
—
—