Chapter 2 - Classical Test Theory, Generalizability Theory, and Item Response Perspectives on Reliability

Abstract

According to classical test theory (CTT) (or true score theory), reliability is defined as the ratio of true-score variance to observed score variance. The observed score is the score that is obtained by the measure itself. The true score is the hypothetical amount of the variable of interest that is specific to the test taker. Measurement error is the difference between the amount of true score and the observed score. The most common conceptualization of reliability is that it is the proportion of observed score variance that is attributable to true score variance.