In criterion-referenced tests, what does reliability indicate?

Reliability in criterion-referenced tests is primarily concerned with the consistency of results in classifying mastery. This means that if a student takes the test multiple times or if different forms of the test are administered, a reliable test would yield similar results each time, accurately reflecting the student's true level of understanding or skill mastery. This consistency is crucial for educators to make informed decisions about students' competencies based on the test outcomes.

The reliability of a test ensures that the classifications—such as whether a student has achieved mastery of a particular skill—are stable and unaffected by external factors unrelated to the students' actual knowledge or performance. High reliability is vital for the fairness and validity of assessments in an educational setting, as it allows instructors and stakeholders to have confidence in the results as indicative of true student performance.