Z:gnu-www-ja-frank--a8b1b1-Another possible explanation f/en

Another possible explanation for the economists' higher defection rates is that economists may be more likely than others to expect their partners to defect. The self-interest model, after all, encourages such an expectation, and we know from other experiments that most subjects defect if they are told that their partners are going to defect. To investigate the role of expectations, we asked students in an upper division public finance course in Cornell's economics department whether they would cooperate or defect in a one-shot prisoner's dilemma if they knew with certainty that their partner was going to cooperate. Most of these students were economics majors in their junior and senior years. Of the 31 students returning our questionnaires, 18 (58 percent) reported that they would defect, only 13 that they would cooperate. By contrast, just 34 percent (14 of 41) noneconomics Cornell undergraduates who were given the same questionnaire reported that they would defect on a partner they knew would cooperate (p&lt;.05). For the same two groups of subjects, almost all respondents (30 of 31 economics students and 36 of 41 noneconomics students) said they would defect if they knew their partner would defect. From these responses, we conclude that while expectations of partner performance do indeed play a strong role in predicting behavior, defection rates would remain significantly higher for economists than for noneconomists even if both groups held identical expectations about partner performance.