The Reactive Effect of Pretesting in Attitude Change Research: General or Specific?

Abstract

In an earlier study it was found that a pretest containing positive statements counteracted the development of negative feelings toward a treat ment. The present experiment sought both to repli cate the earlier study and to answer an additional question: Is the biasing effect general or is it re stricted to pretest content? Again, it appeared that a positive pretest acts as a device to counteract the development of negative sentiments. The effect size, small in the original experiment, was even smaller in the present experiment. Moreover, it was found that the sensitization effect operated only when the pretest and the posttest were identical. Using dis similar pretest and posttest measures eliminated the biasing effect.

It is widely recognized that the standard pre test-treatment-posttest design raises the problem of a pretest sensitization effect. Nevertheless, measures prior to treatment are frequently used in attitude change research. There are a number of reasons for this preference of pretesting over a randomized design without a pretest (e.g., Chris tensen, 1977; Lana, 1969).