Abstract
To examine the construct validity of measures of reflection-impulsivity (R-I), multitrait-multimethod matrix procedures suggested by Campbell and Fiske (1959) were applied to data collected from an educable mentally retarded sample. Specifically, the two R-I dimensions of time and error were assessed by three different methods of measurement at two different age levels. Although the correlations between the dimensions based on two somewhat similar methods were consistent with expectations at the younger age level, the data generally demonstrated minimal and weak construct validity. These results were discussed in terms of the inherent differences between methods and the need to specify the measurement paradigm as well as the nature of the content employed in the instruments.
Get full access to this article
View all access options for this article.