In February we reported on the case of Fred Walumbwa, a leadership scholar at Florida International University who was poised to lose five papers in the Leadership Quarterly for reasons not entirely clear but which appeared to involve problems with the data.
Now we see a sixth retraction for Walumbwa, this one in the Journal of Organizational Behavior. The article, titled (ironically enough), “Authentically leading groups: The mediating role of collective psychological capital and trust,” had appeared in September 2009. Per the abstract:
Although there have been recent theoretical advances in what is increasingly being recognized as authentic leadership, research testing possible mediating processes and the impact on group-level outcomes has not received attention. To help address this need, this study examined at the group level of analysis the role that collective psychological capital and trust may play in the relationship between authentic leadership and work groups’ desired outcomes. Utilizing 146 intact groups from a large financial institution, the results indicated a significant relationship between both their collective psychological capital and trust with their group-level performance and citizenship behavior. These two variables were also found to mediate the relationship between authentic leadership and the desired group outcomes, even when controlling for transformational leadership. Implications for future research and practice conclude the paper.
According to the retraction notice:
The following article from the Journal of Organizational Behavior, Authentically leading groups: The mediating role of collective psychological capital and trust by Fred O. Walumbwa, Fred Luthans, James B. Avey and Adegoke Oke, published online on 1st September 2009 in Wiley Online Library (www.wileyonlinelibrary.com) and in Volume 32, Issue 1, January 2011, Pages 4-24, has been retracted by the journal’s Editor in Chief, Neal M. Ashkanasy, and John Wiley & Sons Ltd. The retraction is on the grounds of the authors’ advice that they made an error in relation to the level of analysis used. As a result of this error, the authors incorrectly calculated key fit statistics. When correctly estimated, the fit statistics do not provide an acceptable level of support for the hypothesized model, rendering the authors’ conclusions, as stated in the article, unsustainable.
The paper has been cited 11 times, according to Thomson Scientific’s Web of Knowledge.
Hat tip: Rolf Degen