Abstract
In an attempt to predict behavior discrepant from measured attitudes, an instrument is developed which utilizes Parsons and Shils' pattern-variable scheme. Some 338 subjects are tested to ascertain their racial attitudes. A sample of 40 subjects is drawn, and these are forced to activate their attitudes in the face of social opposition. The instrument as a whole is unable to predict discrepant behavior, but the diffuseness-specificity variable is found to be an excellent predictor of actual situational behavior in an experimental situation which makes that dilemma paramount. Emphasis is placed on the fact that individual pattern-variable dilemmas, when measured, may offer a way to significantly increase prediction of actual behavior.

This publication has 0 references indexed in Scilit: