Models for Measuring Induced Affective Change: Some Neglected Methodological Considerations

Abstract

The measurement of affect and other personality characteristics has usually been carried out in a predispositional, correlational framework. However, the measurement of varying levels of affect induced by different treatments during an experiment has received less attention because of the methodological problems involved. Typical weaknesses of research concern test-retest effects, test sensitization, test-wiseness, lack of control groups, inductive construct validity of measures, emotional boundedness, purging effects, and short treatment duration. A manageable design paradigm for gaining sound results on experimentally induced affect is presented.