Misconduct forces retraction of health behavior paper

A pair of psychology researchers at West Virginia University have lost their 2013 article in the Journal of Behavioral Medicine after one of the authors was found to have cooked the data.

The paper, “Preference for immediate reinforcement over delayed reinforcement: relation between delay discounting and health behavior,” was written by Shane Melanko and Kevin Larkin. It examined whether people who place less importance on the future were also less likely to adopt healthy behaviors, which come with delayed benefits. Melanko, then a doctoral candidate under Larkin, was evidently at one time a psychology student of some promise.

That promise might go unfulfilled. According to the retraction notice:

At the request of Kevin Larkin, this article has been retracted due to research misconduct of Shane Melanko. Mister Melanko was solely responsible for the actions that resulted in the retraction of this article, and Dr. Larkin was unaware of his actions until after the article was published. No other published work was implicated during the investigation of research misconduct, and any data collected by Mr. Melanko for other studies have been discarded and will not be analyzed or published.

That statement corrected an earlier version of the notice that appeared in October 2014, which read (pdf):

This article has been retracted based on the authors’ request. The reason for the retraction is a methodological error which resulted in inaccuracy of the results reported

Janice Stern, a Springer editor familiar with the case, told us the journal initially learned from Larkin about problems with the data — but not that misconduct was at play:

It was I (the publishing editor) who submitted the first retraction notice at the time of retraction, before it was clear that a formal investigation would take place or that intentional research misconduct had occurred. Upon learning the results of the investigation, Dr. Larkin requested that the second, more accurate retraction notice replace the first retraction notice.

Dr. Larkin first contacted the journal when it came to his attention that the integrity of the data on which the article was based could not be verified. Before the university’s research integrity investigation into Mr Melanko’s conduct was completed and the results finalized, Mr. Melanko admitted to conducting analyses for his thesis project that were reported in this article using data from two independent samples that he merged into one. It was I who called this a methodological error because it wasn’t clear to me that Mr. Melanko’s research misconduct was intentional.

The paper had received 9 citations, according to Google Scholar. We have emailed Larkin for comment but have not heard back from him.