PREDICTION OF CORPORATE FAILURE: THE EFFECT OF FEEDBACK ON HUMAN JUDGMENTS

Previous researchers (eg. Libby, 1976 and Salamon et al., 1976) have proposed three alternative strategies for improving decisions: (1) changing the information set; (2) replacing the decision maker with a model; and (3) training the user (e.g. with the use of feedback information). The present research focusses on the role of feedback in the prediction of corporate failure. The effect of various types of feedback on human judgments is examined in the present study in two ways. The first section of the paper extends Kessler and Ashton's (1981) study which looked at the impact of several alternative types of feedback. Using 173 advanced undergraduate students as subjects, we find that, contrary to general expectations, both task properties and lens model feedback are not effective for decisions taken within a financial accounting context. Several possible explanations are given for this result. The second section of the paper examines the effect of outcome feedback within a financial accounting context. The participants in this study were 40 advanced undergraduate students. We found that outcome feedback is an effective means for promoting learning even where the task is not highly predictable.