If you read this space, you probably know the name Brad Bushman. He studies the effects of violent video games on the people who play them. He also has just retracted his third paper, and significantly corrected another.
Although Bushman remains a prominent voice in a highly contentious field — prompting numerous media to consult him after school shootings or other violent acts — he’s retracted two papers, one following an investigation at his institution, the Ohio State University (OSU), which prompted OSU to strip his co-author of her PhD. (There’s a lot more to tell about that story, including the backlash outside critics faced for taking their concerns about the paper public. To read more, check out our in-depth piece in Motherboard.)
Bushman’s third retraction came this month; he nearly had a fourth as well, but attorneys for the publisher decided that a massive correction (to a paper which previously had been flagged with an expression of concern) would be more appropriate.
The retraction notice from Current Opinion in Psychology states the paper showed too much similarity to a 2016 paper in the same journal by Bushman and Arlin James Benjamin, based at the University of Arkansas-Fort Smith. It notes that although Bushman was the guest editor of the issue of the journal:
The peer review process for this article was handled by an independent Editor with no involvement by the Guest Editor.
Benjamin said he and Bushman were alerted in March to the duplication, and he let the journal know the same day. It didn’t take long for Benjamin to realize what went wrong:
I used the prior paper we had co-authored as a template to draft a new paper. That was a huge mistake. If I had given myself and my coauthor adequate time to really go through and do a thorough rewrite and edit, maybe all would have been okay. I did not do that, which became pretty obvious when I combed through old emails. After I got some edits back from my coauthor, I apparently did not pay much attention to what I was looking at, and thought it was good enough. Basically, I was really sloppy in drafting this manuscript and I am not sure my coauthor was aware at the time of how sloppy I was. That manuscript should have never even made it to review. I could cite any of a number of reasons for how all that could occur (including that I had taken on way too much at that period of time), but none of those excuse the fact that I took my eye off the ball. This is what happens when an author does so. It’s a mistake I promise never to repeat.
Benjamin said he considered other options besides retracting the paper:
I did look into whether or not a correction was possible out of respect for my coauthor. I know he was worried about a withdrawal occurring and wanted to prevent that if at all possible. After some communication with those responsible for that particular journal, the reasoning for a withdrawal made perfect sense to me, and I was comfortable with the decision.
We contacted Bushman to ask if he was concerned about another withdrawal; he referred us to an OSU spokesperson, who deferred all comment to Benjamin.
Even if cleared of any wrongdoing on those prior papers (which [Bushman] made sure I was well aware of), the prospect of being involved in a third was one was something he neither expected nor wanted, and that is something he conveyed to me. I think it is safe to say I wasn’t thrilled about the matter at that time either. We both would have preferred an opportunity to publish a corrected manuscript, if that could have been a possibility. As it turned out it wasn’t.
We also contacted the journal to ask why it didn’t spot the duplication — particularly since it had published the previous paper — but haven’t heard back.
Although duplication may seem like a victimless crime, it isn’t; if similar content is spread across multiple papers, a hypothesis might appear more plausible than it deserves to seem. What’s more, if the original paper is owned by a different journal, duplications can potentially violate the publisher’s copyright.
This was a tragedy of errors, and one where I shoulder a good deal of responsibility for what happened…So, lesson learned, although painfully.
Benjamin is also the first author on the newly corrected paper with Bushman, which was “significantly revised” after an outside researcher — Joseph Hilgard — discovered an error.
According to Benjamin, the original paper reported that “the mere presence of weapons influenced people to think more aggressively.” It didn’t seem to matter if study participants were shown real weapons or just images.
But when performing the analysis, Benjamin miscoded some variables, which led the researchers to draw misleading conclusions in the now-corrected paper, “Effects of weapons on aggressive thoughts, angry feelings, hostile appraisals, and aggressive behavior: A meta-analytic review of the weapons effect literature,” published by Personality and Social Psychology Review. Benjamin consulted with an outside expert to make sure they were fixing the problem appropriately, which resulted in revised findings:
We can still say that cognitive outcomes and appraisal outcomes appear to be influenced by the presence of weapons, but we can no longer conclusively do so for aggressive behavioral outcomes…we now find that the average effect size for the weapons effect is larger for images of weapons than actual weapons. We have no easy explanation for that one.
Bottom line was that we had to make a more cautious conclusions based on the available data.
Benjamin detailed some of the changes in this recent Twitter thread.
The published correction is extensive, and includes changes to two tables and the discussion. Did the authors consider retraction?
I certainly did not consider that a first choice – at most a last resort. Really once I was aware that there was a serious problem with the database and the analyses based upon it, I contacted the editorial team and let them make their judgment. My hope was that they would give us a chance to make the necessary corrections.
We contacted the journal, and heard back from Bertram Gawronski at the University of Texas at Austin, who says he was the editor in charge of the paper. Gawronski told Retraction Watch the journal did consider retracting the paper, but decided not to after “extensive discussion:”
The decision was based on an analysis by Sage’s legal team who came to the conclusion that the current case does not meet the typical criteria for a retraction. The legal team instead suggested publication of an expression of concern followed by a corrected version of the Advance Online version.
We followed up to ask why the publisher’s legal team got involved; Gawronski noted:
In this particular case, the [Committee on Publication Ethics] policies were ambiguous in that they permitted different courses of action, and the attorney provided input to resolve this ambiguity.
Gawronski said he believed the journal would have also published the revised version of the paper, if it had been the original submission:
The results of a meta-analysis on this theoretically and practically important topic are informative regardless of the outcome.
Hilgard, the researcher who first noticed the error, said he was impressed with the authors’ behavior:
Because the authors shared their data with me, I was able to find an understandable and honest mistake. The authors then explored the consequences of the mistake on their own. This is commendable; in my past experiences with other authors, I’ve had to redo all the analyses myself and argue for a correction on my own, with mixed success. This experience has shown me that post-publication correction is a lot easier when the authors work with, rather than against, their critics.
Hat tip: Rolf Degen
Like Retraction Watch? You can make a tax-deductible contribution to support our growth, follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up for an email every time there’s a new post (look for the “follow” button at the lower right part of your screen), or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at firstname.lastname@example.org.