Reliability may be assessed mathematically. Internal consistency may be assessed by correlating performance on two halves of a test (splithalf reliability); the value of the Pearson productmoment correlation coefficient is adjusted with the SpearmanBrown prediction formula[?] to correspond to the correlation between two fulllength tests. A commonly used measure is Cronbach's α, which is equivalent to the mean of all possible splithalf coefficients. Stability over repeated measures is assessed with the Pearson coefficient, as is the equivalence of different versions of the same measure (different forms of an intelligence test, for example). Other measures are also used.
Reliability may be improved without formal psychometric analysis. For example, increasing the number of significant figures[?] in a reliable measure will make it more reliable.
Search Encyclopedia

Featured Article
