Under what circumstances would you expect income to be a better predictor of consumer behavior than a composite measure of social class? When would you expect the composite social-class measure to be superior? Why do marketing researchers use objective, rather than subjective, measures of social class?