Reliability is another important criterion for evaluating measurement quality. Unreliability in assessment results is caused by random error(s) in the assessment situation. The common sources of random error are unpredictable changes in the assessment environment, the persons assessed, the assessors/observers, the occasion of the assessment, or the form of the assessment tool used. In classical test theories, quantitative estimates of reliability are obtained through two related statistics: the Standard Error of Measurement (SEM) and the Reliability Coefficient.

Page 5