performance in this situation would be related to a subject’s grade in a course; in fact, the case was
generally administered in nongraded courses.
Administrative instructions were standardized (Appendix B). To date, the administration has been
conducted by 33 different people, although about 40% of the groups were run by the author. It was
administered on 91 different occasions over a period of over 5 years (February 1972 to March 1977).
The administration by different people in different settings at different times should help to compensate
for biases.
Table 3: Decisions in Panalba Case: Control Condition
Level of Social Irresponsibility Percentage of Groups (g = 57)
|
High (decision e: block FDA) |
79 |
|
Moderate (decisions b, c, d) |
21 |
|
None (decision a: remove drug) |
0 |
Results: The results address the following three questions: First, can managers be expected to act in
an irresponsible manner? Second, do differences among backgrounds of managers lend to differences in
socially irresponsible behavior? Finally, can the role perceptions of managers be modified so that they
affect decision-making?
Social Irresponsibility under the Current System: The control version of the Panalba case provided
evidence on the likelihood that managers will make socially irresponsible decisions. It should be noted
that subjects were not instructed to maximize profits. However, as noted by Larsen et al. [36] in their
study of obedience, it seems sufficient merely to put people in a mildly compelling situation. That is,
subjects in the Panalba case were expected to fee1 some pressure toward that stockholder role
because of their background. The fact that they were working in a group was expected to increase this
pressure because the stockholder role was expected to be the dominant viewpoint in the group.
The results are presented in Table 3. None of the groups removed the drug from the market. In fact,
79% of them selected the highly irresponsible decision. This was the decision that had been made by
Upjohn. According to Mintz [45], Upjohn had a judge serve an injunction on the FDA, and they also
used political pressure. Although Upjohn was eventually forced to remove Panalba from the U.S.
market in March 1970, it continued to sell Panalba in foreign markets [46].
The possibility of biases in the control version was examined. Although many researchers have
suggested that subjects help the researcher prove his hypotheses, Sigall et al. [62] reviewed the
evidence and found little empirical evidence to support such a viewpoint. Furthermore, Carlsmith et al.
[11] found no differences in their role-playing study when comparing results from two experimenters
who held contradictory hypotheses. A follow-up survey of 32 subjects in the contro1 version of the
13