a measure of the consistency and freedom from error of a test, as indicated by a correlation coefficient obtained from responses to two or more alternate forms of the test. Also called comparable-forms reliability; equivalent-forms reliability; parallel-forms reliability.