ABSTRACT
Reliability is how much repeatability, consistency, and precision a measure's scores have. There are multiple aspects of reliability, and the extent to which a given aspect of reliability is important depends on the construct of interest. Measures should show strong inter-rater, intra-rater, immediate parallel-forms, and internal consistency reliability. However, the extent to which our measures should show test–retest and delayed parallel-forms reliability depends on the stability of the construct and the time lag. Moreover, although measures should show strong internal consistency reliability, the internal consistency reliability of a measure can be too high, and for some constructs (i.e., formative constructs), internal consistency would not be expected.
Reliability is not a characteristic that resides in a test. The reliability of a measure's scores reflects an interaction of the properties of the test with the population for whom it is designed and the sample, situation, and context in which it is administered. Thus, it is important to adequately describe the aspects of reliability that have been considered and the population, sample, and context in which the measure is assessed.
