Issues in Measuring Reliability

Abstract
In many literacy studies, it is important to establish the reliability of independent observers' judgments. Reliability most commonly is measured either by the percentage of agreement or the correlation between the observers' judgments. This article argues that the percentage of agreement measure is more difficult to interpret than are correlation measures because of the following: (a) the effects of chance agreement are not accounted for automatically by the percentage of agreement measure; and (b) rates of chance agreement are strongly influenced by the variability of the data, by “ceiling” and “floor” effects, and by the scoring of near agreement as perfect agreement. For these reasons, the authors recommend that the field of literacy research adopt correlation as the standard method for estimating the reliability of observers' judgments.