Author(s): Podeschi Denise M, Wyer Robert S
88 college students (judges) indicated their acceptance of generalizations of the form "A's verb B's" on the basis of information about whether the relation held between individual instances of A and B (inductive evidence) or whether it held between more general classes of elements to which A and B belonged (deductive evidence). Evidence was specific to either the subject or the object of the proposition to be judged. Generalizations were more readily accepted on the basis of inductive than deductive evidence. The acceptance of generalization based on inductive evidence appeared to depend in large part on judges' assumptions about the implicit quantifiers (e.g., "all" and "some") of the categories described in the these propositions, whereas the acceptance of generalizations based on deductive evidence depended on whether the definitions of these categories were interpreted as disjunctive or conjunctive. In some instances, the acceptance of generalizations about affective relations appeared to be based on different criteria than was the acceptance of generalizations about overt behavior. Specifically, these criteria appeared similar to those used as a basis for making evaluations of objects in an impression formation task. A general positivity bias was also detected. Results are generally consistent with those obtained in an earlier study by R. P. Abelson and D. E. Kanouse (1966); however, they clarified the reasons for these earlier findings and, in some instances, suggested contingencies in their occurrence.