Abstract
The publication of the Jackson Personality Research Form popularized a sequential strategy of constructing psychometric scales which bolstered the hopes (at least temporarily) of constructors of personality tests. Just at the point where even the stalwart seriously considered following their colleagues in abandoning measurements of personality, the Jackson demonstration breathed new life to the belief that content scales could be developed with respectable psychometric properties of homogeneity, bipolarity, convergent, and discriminant validity. Also these could be relatively free of the stylistic response variances of acquiescence and social desirability that consistently plagued most psychometric scales of personality. While scales with purity of content are perhaps laudable their construction is not without cost, one that narrows the meaning of the general content area. This paper briefly describes a demonstration of the inverse effect of content purification on homogeneity, using procedures to separate three varieties of rejection (of self, by others, and of others) from each other. Results using an item pool of approximately 450 items on rejection and a sample of 402 individuals suggest that homogeneity and content separation are bound inversely to each other. Implications for methodology of measurement are discussed.
Get full access to this article
View all access options for this article.