Abstract
At the request of the Editor of Psychological Science and Sage Publications, the following article was retracted:
Gino, F., & Wiltermuth, S. S. (2014). Evil genius? How dishonesty can lead to greater creativity. Psychological Science, 25(4), 973–981. https://doi.org/10.1177/0956797614520714
The Journal Editor was contacted by the Research Integrity Office of the Harvard Business School (HBS) regarding the outcome of an internal investigation their office conducted into the data reported in this article.
An investigation by an independent forensic firm engaged by the HBS revealed discrepancies between the published data sets and what HBS referred to as the earliest known versions of the data, which were retrieved from the first author’s research records (HBS specified the earliest known and published datasets were each created and last saved in 2012, according to their Excel properties). HBS’s investigation concluded that there appear to be anomalies in the dataset used for analyses in Experiment 4:
(a) The number of participants who were classified as cheating on the experimental “problem solving” task was indicated as 31 in early versions of the original data (collectively labeled “Earlier Data,” obtained from Dr. Gino’s research records). In the file used for data analysis (“Publication Data,” also obtained from Dr. Gino’s research records), the classification of 12 of the participants was changed from “noncheaters” to “cheaters” and they were added to the 31 in the original data set, for a total of 43 “cheaters” reported in the article (p. 977).
(b) Some values for performance on the Remote Association Task (RAT; Mednick, 1962) task were manually changed within the Earlier Data. The manually entered values were transferred to the Publication Data. When the manually entered values were replaced with those that were calculated by the same Excel formula used to calculate the other values for RAT performance, several values changed. The significant relations between cheating and caring about rules, and between cheating and flexibility on the uses task became less statistically significant than originally reported, and the relation with RAT performance became nonsignificant.
(c) As a result of removal of the data anomalies, the difference in mean performance for the cheaters and noncheaters diminished and the trend in means for RAT performance inverted, such that values were higher for noncheaters, whereas they were reported in the article as higher for cheaters (Table 2, p. 978). The inverted trend in means results in a pattern of data that is contrary to interpretation of and the conclusions drawn from the research.
Counsel for the first author informed the journal that whereas Dr. Gino viewed the retraction as necessary, she disagreed with references to original data, stating that “there is no original data available” pertaining to the research reported in the article. Although to our knowledge, anomalies in the data are specific to Experiment 4 of the article (of five experiments), the Editor decided to retract the article because the experiment in question is integral to the overall conclusions of the research.
Both authors agreed to the decision to retract.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
