Reliability is the extent to which a test yields consistent results. A reliable instrument produces similar scores when administered under similar conditions or when its items are compared with one another.

Test-retest reliability examines stability over time. Parallel-forms reliability compares two equivalent versions of the same test. Internal consistency, often measured by Cronbach’s alpha, assesses whether items that purport to measure the same construct hang together. Inter-rater reliability evaluates agreement among different scorers. High reliability is a prerequisite for validity; an unreliable test cannot be a valid measure of anything.

Reliability coefficients range from 0 to 1.00. Values above .80 are generally considered acceptable for most research and applied purposes, though standards vary by context. Sources of unreliability include ambiguous items, inconsistent administration, scoring errors, and temporary fluctuations in the test-taker. Careful test construction and standardized procedures improve reliability.

A test that gives roughly the same answer on repeated occasions is reliable.