A physiology journal has retracted a pair of papers from a group in Australia after learning that the flawed work was the subject of an institutional investigation.
The articles, both of which were published last year in The Journal of Strength and Conditioning Research, came from a group at the Murdoch Applied Sports Science Laboratory, part of Murdoch University. The first author on both papers was Liam J. Hughes, a PhD student at Murdoch who was terminated as a result of the misconduct.
James Steele, an exercise researcher at the ukactive Research Institute in England — part of a group of data sleuths who have been responsible for identifying unlikely data in several other studies in his field — told us that he became of aware of the retractions inadvertently:
We first noticed the retraction because we are in the midst of a systematic review and meta-analysis which included one of the studies. We were on a second round of independent search and screening to check we were turning up the same papers when this time round we noticed it had been retracted. However, the Journal of Strength and Conditioning Research didn’t provide any information as to why.
Steele — who recently retracted one of his own papers — continued:
Anyway, I was aware of the lead author, Liam Hughes who was a PhD student, having seen them quite active on researchgate and twitter previously. So I thought I would contact them to ask. But when I did I noticed all their accounts had been disabled.
Steele said he then reached out to Brendan Scott, the last author on the articles and Hughes’ supervisor, who confirmed that he and the third co-author had grown suspicious about the data and launched a formal inquiry into the papers.
I also asked about another paper which Liam was a co-author on and led by another student, but Brendan confirmed that this one was fine and that he had actually assisted more directly with data processing in that particular study.
Scott confirmed Steele’s account, telling us by email:
We were made aware late last year that some members of our lab had concerns that Liam had fabricated data. A/Prof [Jeremiah] Peiffer and I began looking into this, and unfortunately we also became concerned, so I lodged a formal process with our university whereby a formal investigation was undertaken (i.e. not including Jeremiah and me, but handled by an impartial investigator).
During this investigation, Liam confirmed the allegations that he had fabricated data without my or Jeremiah’s knowledge. As a result of this, the university contacted the journals where Liam had published these papers to request that they be retracted on our behalf.
According to the retraction notice — which has yet to appear on the journal’s website, but which we received from the editor-in-chief, Nicholas Ratamess, of The College of New Jersey:
This is to notify our readers that the Editorial Office of The Journal of Strength and Conditioning Research has retracted the articles entitled “The reliability and validity of using the Push Band v2.0 to measure repetition velocity in free-weight and Smith machine exercises”1 and “Estimating repetition in reserve in four commonly used resistance exercises”2 by Hughes LJ, Peiffer JJ and Scott BR, both published ahead of print in The Journal of Strength and Conditioning Research.
We are hereby retracting these articles due to errors in data collection by the authors which invalidated the data. The Editorial Office of The Journal of Strength and Conditioning Research was informed of the errors by the corresponding author’s university after the manuscript was already published ahead of print.
Hughes did not respond to a request for comment. Scott added:
This has been a very sad situation for all of us involved. Liam seemed to be a bright young researcher, with a good grasp of his field. As co-authors, Jeremiah and I were shocked to realise that some of Liam’s research was based upon fabricated data – he had presented findings to us which seemed appropriate and never caused us to question the integrity of his work. Nevertheless, we see this as a good learning experience to be more aware of the potential for research misconduct in the future.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, 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 [email protected].
I’m ready to admit that something has eluded me here but… how does the “fabricated data” issue identified by the supervisor translate into “errors in data collection” in the retraction notice???
Hi Jack; It sounds like some data may not have been collected, and instead these data may have been “fabricated”. Maybe?
Is there a list of retractions in Australia?
I also found this odd, particularly since the student confessed to fabricating the data and the university concluded that this happened, so there shouldn’t be a legal issue in saying so. Maybe they were just trying to be ‘kind’ to a young scholar? Or just trying to minimize attention?
How does the lab head or supervisor not notice the data when it is originally collected and recorded in the lab notebook? Where’s the oversight?
Good question, Ray! I have two perspectives on this. First, yes, every PI should closely supervise all data collection & processing going on in her/his lab. After all, hypothesis testing and datal collection are our “core business”. I am saying this in general and explicitly do not pass judgment on the particular supervisor involved here. I simply know too little about what he did or did not do. And even if he checked the lab books and data, it’s not out of the question that a particularly clever individual will know how to fake these trails of evidence, too.
Second, to the extent that lack of oversight results from incentivizing big labs with lots of grant money and lots of staff, but little incentives for being careful and meticulous, we need to have a serious conversation in science how to deal with this kind of pressure. As long as the mantra of cash-strapped higher-ed institutions is “more, more, more” and we behave like entrepreneurs who try to invest maximally in their production facilities, but care little about the product itself or whether the yield is even remotely commensurate to the investment in terms of quality (not quantity!), scientists will fight a losing battle. More scandals will ensue and public trust in science will be eroded. That’s bad, because it will undermine one of the pillars of open, democratic societies and will lead to less funding for science in the long run.
We need to start incentivizing good, careful science again, high-quality productivity that really gets us somewhere in terms of understanding our world.
Though I am unaware of the details of this case, I am uncomfortable with the fact that the PhD student was dismissed but without any consequences for the supervisor. All authors should be responsible for a research paper unless the paper explicitly states otherwise.
Also, Universities should disclose research issues (such as retractions, alterations to already published research etc) for everybody’s consumption.
We have no idea whether or not there were consequences for the supervisor. Certainly not dismissal (which would have been highly unwarranted) but there could be other consequences which would never been made public.
Obviously fabricating data is unacceptable and dismissal might have been appropriate or not, depending on the corcumstances but on a personal level I would hope that someone has reached out to the student to check on their wellbeing.