A contentious case over whether a fired ecologist deserves whistleblower protection is playing out in Kansas, and the National Science Foundation (NSF) has once again weighed in.
For the second time, the NSF has told the researcher, Joseph Craine, that he does not qualify for protection as a whistleblower after he was fired from Kansas State University (KSU) for sharing misconduct allegations with a journal editor.
Craine initially asked the NSF for whistleblower protection status in 2014, arguing that he had been retaliated against for making misconduct allegations to the journal Ecology. The NSF denied Craine’s claim, but Craine appealed to federal court, which found the NSF’s reasoning opaque and remanded the case back to the agency. On June 24, the NSF’s General Counsel Lawrence Rudolph issued a new 11-page letter that lays out the basis for its decision:
The nearly thousand-page record in this matter is more than sufficient to support a finding by NSF that you were fired because of baseless allegations of academic fraud made to the Ecology editors, and not because of any protected disclosures. No KSU employee ever suggested that you should be terminated because you had uncovered, or were likely to uncover, misconduct. Accordingly, I conclude that your employment was not terminated as a result of a reprisal.
(As a side note: The verdict suggests pausing before passing your misconduct allegations onto a journal editor. For tips on how to report suspected misconduct, see our recent advice in Lab Times.)
The convoluted case has a lot of backstory, which we previously covered. Craine’s court battle with NSF and KSU has also brought to light new documents that highlight conflicts in the peer-review process and provide a cautionary story to people who allege misconduct outside of protected forums.
In the fall of 2013, a group of Craine’s colleagues submitted a manuscript to Ecology about research conducted at the Konza Prairie Biological Station, an NSF Long-Term Ecological Research (LTER) site. “Abrupt transition of mesic grassland to shrubland: evidence for thresholds, alternative attractors, and regime shifts,” led by Zak Ratajczak, reported that a greater frequency of fires over time could trigger an abrupt transition of the grasslands into shrublands. Fire management policies remain an important, and sometimes contentious issue, among ranchers who graze cattle on these lands, environmentalists who would like to preserve grassland ecosystems, and citizens who may be impacted by smoke levels.
When Craine heard about the manuscript, he emailed the journal to issue a “confidential” warning on Oct. 19, 2013:
It pains me to say this, but I think the paper is fraudulent.
I think you can understand that it would be better for me to address this discretely during the review process. I would prefer not to force a retraction publicly.
Craine already had some differences of opinion with these colleagues and has openly challenged decades of KSU advice on fire management practices. He thought that the abruptness of the transition was an artifact of a change in the person collecting the data.
Ecology editor Debra Peters reached out to one of those data collectors, Scott Collins of the University of New Mexico — a member of Ratajczak’s dissertation committee — to ask his informal opinion. According to emails Craine obtained from a freedom of information request and provided to Retraction Watch, Collins wrote that he had a low opinion of Craine:
Suggesting fraud is rather inflammatory. If that term is coming from Joe Craine, then he should be chastised. He is a genuine poison pill who has published several papers that are probably fraudulent.
Peters decided to invite Craine to review the manuscript, and he recommended that it be rejected. A second reviewer expressed separate concerns about whether the data really indicated an abrupt transition, and Peters asked the authors to sharpen their analysis. In the cover letter for their revised submission, Ratajczak and coauthors admitted having made mistakes:
We regret this honest error and are incredibly grateful it has been identified. Fortunately, analyses we had completed before suggest that non-linear shifts in shrub cover are not artifacts of investigator changes . . . While this observer bias may be possible, it is not well-supported by the data.
Collins reiterated his enthusiasm for the manuscript and the data, and Peters ultimately decided to accept it. “If I do accept it, then Craine has threatened to go public with his accusations,” she said in an email to Collins. “How do we keep this issue in-house at KSU and out of the ESA journals?” According to emails, Peters added Craine to Ecology’s secret blacklist of reviewers to never use, while Collins passed on information about Craine’s allegations to his KSU colleagues.
After the revised manuscript was posted online in January, Craine sent another email to Ecology, noting he “may need to involve NSF OIG [Office of Inspector General].”
Collins passed these emails along to John Blair, a collaborator at KSU, and the university followed protocol by launching an inquiry into possible research misconduct by the authors. It concluded that the allegations were frivolous and pointed out that the article had been:
rigorously peer reviewed, further demonstrating that Dr. Craine’s issues with regard to the article represented only his opinion of the data, and not actual fraud.
The KSU panel launched a follow-up investigation under an unusual policy specified in the University Handbook. Employees were prohibited from disclosing misconduct allegations to individuals outside of the University without first going through university channels.
Fearing that he would be fired, Craine sought federal whistleblower protection through the NSF. The agency denied his claim in 2014 because he had communicated his concerns to a journal editor rather than a “qualifying person,” such as an NSF employee or a law enforcement agency; the next month, KSU fired him for violating its policies and making “misrepresentations that are malicious, and at best, frivolous.”
Last fall, Craine filed a petition for review with the Tenth Circuit Court of Appeals, arguing that he had never been given an opportunity to respond to the NSF’s decision and that the general counsel relied more heavily on a one-sided account provided by KSU rather than its own investigation. He also argued that disclosure to journal editors should be covered under the whistleblowing statute:
Journals are effectively contractors of the University that are provided monetary compensation for the review and publication of federally funded research conducted by university employees. . . . Not only are editors management officials, but editors investigate, discover, and address misconduct for universities.
In the new decision letter, the NSF denied Craine whistleblower protection and reiterated that “disclosure to an academic journal is not a protected disclosure.” So, that’s some food for thought before passing your misconduct allegations on to a journal editor.
We’ve reached out to Ratajczak and Peters for their perspective on the events and the Ecology peer review process, and will update this post if they respond. In an email to Retraction Watch, Scott Collins wrote:
To be a whistleblower [there] has to be a crime [or] a potential crime. Neither happened.
Meanwhile the saga continues: This week, Craine filed a new petition for review in federal court.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.