any theoretical and statistical framework describing how respondents generate their answers to items on a scale or instrument and explaining associated sources of error, which in turn provides a means to determine the reliability of test scores across different testing situations. There are three major approaches: classical test theory, generalizability theory, and item response theory.