Three basic measurement characteristics found in a good assessment tool are reliability, validity, and responsiveness.47-49 Reliability is primarily concerned with the stability of items within a test and the uniformity between test scores over time. Two common forms of test reliability are internal consistency (how well items "hang together") and test-retest reliability (the stability of scores over repeated measurements). Test validity deals with the degree to which an instrument accurately measures what it claims to measure. Several types of validity evidence can be examined, such as face validity (degree to which the scale appears to measure the intended domain), content validity (how well test items qualitatively represent the actual content area of study), and criterion validity (how well an instrument's scores correlate with a "gold standard"). Two types of criterion validity are concurrent and predictive (or known-groups) validity. Another type of validity is construct validity (how well test items reflect the latent variable(s) in question), which can be measured through convergent or discriminant associations with other variables. Finally, responsiveness exists when a measure detects QOL changes as a result of disease or
Was this article helpful?