Logical and Empirical Oppositeness in Value Responses

Abstract

Free value responses from 555 Ss in 16 domestic and foreign samples were analyzed to see whether good and bad behaviors named by Ss would fall into logically opposite categories (about three-fifths of the responses did so) and whether logically opposite categories would be correlated across the numbers of Ss represented in them ( rs ranged from .08 to .63 in the 16 samples; the average r was .31). E concluded that free responses about good behaviors should not be mixed with those about bad behaviors, because good and bad are not necessarily empirical opposites even when they are logical opposites.