Classical Test Theory Reliability

Abstract

Reliability is the degree to which assessments are free from errors of measurement. There are many types of measurement error, including stability over time, consistency across forms, consistency in ratings, and internal consistency. The different sources of error lead to different estimates of reliability: test–retest, alternate form, interrater, split half, and internal consistency. The context of the test use determines the type of reliability that should be used. While the reliability coefficient provides an overall index of the consistency of assessments, the standard error of measurement provides an index that is useful for examining the consistency of individual scores.