Reliability - the Precision of a Measurement (Version 2.0)

Abstract

Reliability describes the precision of a measurement. This contribution begins by defining the concept of reliability and explaining why the reliability of a measurement is relevant. It then discusses the model assumptions that must be made in order to estimate the reliability of a measurement, and it presents five methods of estimating reliability: the test-retest method, the parallel-forms method, the split-half method, the internal consistency method, and the estimation of reliability using structural equation modelling. The contribution concludes with a brief outline of the commonalities and differences between classical test theory and the item response theory and the importance of these theories for the estimation of reliability.