Consistency or precision of measurements under stated sources of variation. Reliability depends on the population, conditions and intended score; it is not a permanent property of an instrument alone. Inter-rater reliability and test-retest reliability address different sources of variation. Reliability does not establish construct validity.
The degree of association among items or components of a measure under a specified model. High internal consistency can arise from overlapping content or common method variance; it does not by itself establish one substantive dimension or construct validity.
Consistency of scores obtained from the same units on different occasions, under a justified interval and conditions. A low value may reflect measurement error, genuine change or both; the intended stability of the construct matters.
Consistency or agreement among raters assessing the same units. The statistic should match the scale and whether consistency or absolute agreement matters. Repeated ratings improve information about a unit without creating additional independently sampled units.
Articles by others on the same topic
There are currently no matching articles.