Below I describe the process through which I have evaluated paper. Before doing so, I wish to emphasize that I remain agnostic to the truth of allegations against Francesca Gino, and have yet to see sufficient public evidence of intentional manipulation or data fraud to warrant suspicion in any other paper on which Francesca Gino is a coauthor. The burden of proof for such accusations should be very high, and viewed entirely differently than questions about evidential value. I am hopeful that as more information is revealed from the secret HBS investigation, more definitive conclusions can be reached. Nevertheless, I am evaluating my papers carefully now for transparency reaons, to best of my abilities and available data.

This paper is a mixed methods paper that combines econometric analysis of state-wide emissions testing data with a follow-up scenario study to verify theoretical arguments and identify psychological mechanisms.  Study 1 was conducted by me using proprietary data and Stata programs. Study 2 was conducted by Francesca Gino using pencil and paper data gathering and separately analyzed by both authors from an excel file created by RAs (to the best of my memory). I have all the original emails and the excel file as sent to me in April, 2008.  I also have the original Stata do-file and log-file that I used for the analysis. We do not have the paper and pencil sheets, which were not kept past several years per standard practice and I believe IRB protocol. Thus, I infer nothing from the non-existence of those original physical data.

I note that the manipulation in Study 2 is extremely strong, asking if someone would illegally lend their parking ticket to someone based on whether or not they have a luxury car. Cars at this time were the most visible signals of wealth, particularly in Pittsburgh. I believe the effect size passes face validity evaluation. 

I have constructed an extensive replication of the main results using the excel data as sent to me. Everything replicates perfectly, with one small question about how a couple of respondents were coded as "Asian" or not as a function of national origin. I also ran extensive forensics on the data to search for anything that might indicate falsified data, a process in which I have strong expertise. I found no evidence of data manipulation or fabrication. The data do include a small percentage of repeat ID numbers, which I believe (but do not know for certain) resulted from the reuse of index cards across many experimental sessions (which were unfortunately not recorded). These repeat observations are not surprisingly more common in one condition given that repeat use of index cards would likely be session related. Regardless, these observations have no influence on the estimated effect, and there is no correlation between any two observations with the same ID. 

This paper was published in Organization Science, and since I am the Editor-in-Chief, this replication and forensic package was evaluated by an ad hoc committee by our parent organization, INFORMS, and cleared of any suspicion. The same process was followed for the one other paper involving Francesca in this journal. I note that the implementation of this process was not initiated because I believed there was sufficient reason to suspect data manipulation in these papers, but rather to protect the reputation of the journal from any inferences drawn by the public (justified or not). 

BELOW WAS EDITED 11/07/23

I originally decided not to publicly post the data or replication packet because of concerns that evaluations will not be well-adjudicated in the current public sphere.  As I noted, my intention was for scholars wishing to view and evaluate this replication packet to contact me directly for it, but many people were not satisfied with this process, and I now understand why. I should have just posted the data this weekend. Given that, I have now posted the replication packet as shared with INFORMS in July at: OSF Packet.   If this link doesn't work, let me know, because this is my first time posting on the specific OSF platform. I welcome any constructive comments, suggestions, or concerns. I spent weeks on this, and did my best to derive the right answers. I'll will work to get the other papers' packets ready and posted, but I request patience as getting this right takes time with such old projects.