Women heard another person state attitudes that were either in high agreement or high disagreement with their own attitudes. The potency of an approach-avoidance dependent variable was compared with traditional dependent variables for this situation, ratings of inter-personal attraction. Eighty-five percent of those hearing high agreement volunteered to return to the laboratory to continue participation in the study at a later time. Nobody who heard high disagreement volunteered to return. This difference between the two treatment conditions was considerably greater than the difference for ratings of attractiveness. It was concluded that the increased potency of the approach-avoidance measure came from its importance to the subjects and that approach-avoidance would be a good measure to use in comparing different predictions from the several theoretical positions that attempt to account for the effects of agreement and disagreement.