ON TESTING THE COMMONALITY OF CONSTRUCTS IN SUPPLIED GRIDS

Abstract

The repertory grid technique has long been the empirical tool of choice for research based on personal construct theory (Kelly, 1955). In some situations, respondents are asked to evaluate a set of elements with respect to a fixed or supplied repertoire of constructs, and an assumption is made that respondents will employ these constructs similarly. This situation is thus equivalent to the semantic differential technique. In the repertory grid context it is justified through the commonality of construing corollary of personal construct theory. This is an assumption that can be tested, although it seems to have attracted little attention in the past. Here a criterion for commonality of a construct is proposed, namely that of the data conforming to a unidimensional scaling model, and a simple method of testing this is illustrated.