Skip navigation

Crossing the line

Academic misconduct is a reality, but what's less certain is how often it occurs and what's the best way to combat it

BY ALEX GILLIS | DEC 03 2007

Before last year’s conference of the Administrative Sciences Association of Canada, one of the submitted papers was discovered to contain plagiarized material. “The fact that there had been plagiarism was not in question by anyone, including both of the authors,” says Sunny Marche, associate dean of graduate studies at Dalhousie University and an executive of the scholarly association. The paper’s first author, a graduate student, claimed ignorance of North American practices. The second author, a faculty member, claimed not to have noticed. “These are not strong defences by way of rebuttal or explanation,” Dr. Marche wrote later on a Dalhousie blog.

The conference, taking place in Banff, Alberta, had no formal policy about what to do in a case of academic misconduct, but the solution was straightforward: “We rejected the paper and sanctioned both authors by prohibiting them from submitting work to the conference or to the academic journal connected with the community organizing the conference,” Dr. Marche recalls.

“But was that the limit of our responsibility?” he asks. “Should we have reported the individuals to the dean of their respective faculty? Should we have posted their names in some kind of on-line ‘hall of shame’? Should we report the individuals to the authors of the work which was plagiarized?”

He asked three people with experience as deans at Dalhousie what they would have expected from conference organizers if this had happened at Dalhousie. One said the paper should have been rejected – end of story. Another said both authors should have been banned from submitting for a number of years. The third said the case should have been reported to the plagiarizers’ own faculty and sanctions probably would have been applied at their university.

This lack of agreement in a single case of plagiarism may reflect the fact that no two incidents of academic misconduct are identical and that there always will be discretion in imposing sanctions. Also, the lack of agreement may point to the issue’s complexity, to insufficient knowledge about policies and procedures, or to a lack of consistency among different disciplines.

Recently, a few high-profile allegations of scientific fraud have drawn public attention to the issue of research misconduct. In a case that received national exposure in a three-part series by CBC television in 2006, Ranjit Chandra, a star professor at Memorial University, was accused of falsifying data more than once in his career – most recently for a 2001 report on how multivitamins improved memory in seniors. After Dr. Chandra resigned from Memorial, the university halted an inquiry it had initiated into Dr. Chandra’s research. In another publicized case, Eric Poehlman, a researcher at Université de Montréal, admitted in 2005 to fraudulently obtaining grants and fabricating data while working in the United States.

After the CBC series aired, Memorial released a report it had commissioned Paul Pencharz to undertake on how the university had handled the Chandra investigation. Dr. Pencharz, a staff physician and senior scientist at the Hospital for Sick Children and professor of pediatrics and nutritional sciences at the University of Toronto, found that Memorial had followed its guidelines after allegations of fraud were raised against Dr. Chandra. But he also said that Memorial should have finished its inquiry, partly to prevent a loss of public faith in the scientific process. Dr. Chandra is suing Memorial and CBC for libel.

An affair like this is a huge headache, says Nigel Lloyd, executive vice-president of the Natural Sciences and Engineering Research Council. “It takes an inordinate amount of time and money, and it damages the institution and science in general. Worse, it casts aspersions on the whole system.”

Those are some of the unfortunate outcomes when allegations of research misconduct are publicized. Still more serious are the potential repercussions when someone makes up or falsifies data in applied fields, such as health, engineering or nutrition, where public safety may be put at risk. Finally, there is the issue of justice: when researchers commit fraud, it’s not fair if they get away with it. “Why is it zero tolerance for students,” asks one observer, “and maximum understanding for professors?”

The very public Chandra case has prompted some action from federal agencies. Health Canada brought together 15 stakeholders to form the Canadian Research Integrity Committee, or CRIC. Among the members are the three major research-granting agencies and several foundations and associations, including the Association of Universities and Colleges of Canada and the Canadian Association of University Teachers.

CRIC is now identifying and assessing “potential models to improve the existing system in Canada,” says Wendy Sexsmith, acting chief scientist at Health Canada and co-chair of CRIC. (The other co-chair is Kevin Keough, former president of the Alberta Heritage Foundation for Medical Research.) Last January, CRIC invited more than 100 participants from universities, federal and provincial agencies and foundations to a workshop to review research-integrity policies in Canada and to compare them with those of other countries.

“People recognize that something has to be done,” says Christine Trauttmansdorff, corporate secretary of the Social Sciences and Humanities Research Council, who attended the workshop. “It’s the ‘how’ that gets a little more complicated.”

In addressing the issue, the Canadian agencies are part of an inter- national trend. In September, more than 300 researchers, government officials and editors from around the world met in Lisbon, Portugal, for the first World Conference on Research Integrity. NSERC’s Dr. Lloyd attended, and he says that participants had hoped to agree on the need for international guidelines on research integrity, but they didn’t get that far.

Participants agreed, however, that any new guidelines have to go beyond fabrication, falsification and plagiarism and that agencies should investigate other questionable practices, which are also “quite significant,” says Dr. Lloyd. Other problems include omitting data, not declaring conflicts of interest, not crediting people as authors, and falsely crediting people who shouldn’t be listed as authors. At the CRIC workshop, Bernie Bressler, assistant dean of research at University of British Columbia’s faculty of medicine and executive director of the Vancouver Coastal Health Research Institute, gave some concrete examples of questionable research practices, including:

  • a transplant patient enrolled in a clinical trial even though he didn’t meet the inclusion criteria, which could result in false data;
  • a doctor who held shares in the company conducting clinical trials;
  • a disagreement over who was principal investigator that delayed release of important data on a large study on arthritis.

What is the incidence of research misconduct? At the CRIC workshop, participants were told, “About five or six studies in total have estimated that the rate of misconduct is somewhere between one in 100 and one in 1,000.” But participants also heard that in one U.S. study, 50 percent of researchers answered “yes” when asked if they knew someone who had committed misconduct.

There aren’t any reliable estimates for Canada, although most observers believe Canada’s experience wouldn’t be very different from that of the United States. The three major research-granting agencies award thousands of grants each year, but they field relatively few reports of misconduct allegations:

  • NSERC fielded 77 misconduct allegations between 2003 and mid-2007, with 23 resulting in full investigations and six in misconduct findings.
  • CIHR opened reports on 54 allegations of misconduct between 2000 and 2006, and 21 were found to be in violation of CIHR policies.
  • SSHRC, in the past five years, forwarded 13 allegations to institutions for investigation, one of which resulted in a finding of misconduct.

To put this in perspective, NSERC hands out annually close to 2,500 multi-year grants to individual researchers under its major, individual-grants program, and SSHRC and CIHR each distribute about 900 multi-year grants to individual researchers under comparable programs.

The granting councils are not the sole players in handling cases of research misconduct, but their Tri-Council Policy Statement on Integrity in Research and Scholarship is the most important policy statement on the topic in Canada. The agencies developed it more than a decade ago after a tragic case that included allegations of academic misconduct. Valery Fabrikant accused his engineering faculty at Concordia University of tolerating academic misconduct and claimed that the situation drove him to murder four staff members in 1992. Harry Arthurs, president emeritus at York University, looked into Fabrikant’s allegations of academic misconduct and concluded that some of them had merit.

Every institution whose researchers receive funds from a granting council must sign a memorandum of understanding that makes the Tri-Council statement binding. It says that researchers shouldn’t fabricate, plagiarize, hide conflicts of interest and engage in other misconduct. It places the responsibility for ensuring integrity on those that administer research funds, Canada’s universities. The latter are responsible for promoting integrity, investigating possible misconduct, imposing sanctions, and informing the appropriate council of the actions taken.

“We’re a funding body, pure and simple,” says Burleigh Trevor-Deutsch, director of ethics at the Canadian Institutes of Health Research. “We’re not a regulator and we’re not mandated to be a tribunal or anything like that.” Enforcement takes place only at universities, he says.

Many believe that Canada’s institution-based system for investigating allegations of misconduct is the appropriate system. “I can’t speak for every university in the country,” says Ted Hewitt, vice-president, research, at the University of Western Ontario, “but I know that at Western it seems to work well. We have a policy; it’s on the web. We very occasionally will receive a complaint, and they’ve been handled expeditiously and usually in a manner satisfactory to the parties, but where there continue to be disagreements, we have measures in place to deal with those also.”

Most if not all Canadian universities post on their websites their academic integrity policy and the process to be followed if an allegation of misconduct is made. At universities with faculty unions, such as Western, the procedures are often included in the collective agreement.

Peer review committees and scholarly journals also have responsibility in how they treat submissions when editors suspect plagiarized or falsified data. An international forum for editors of peer-reviewed journals, the Committee on Publications Ethics, encourages editors to report, catalogue and initiate investigations into misconduct in the publication process. Scholarly journals often detect cheating first. For example, in 2000 the British Medical Journal concluded that a Chandra submission (about his multivitamins improving seniors’ memories) sounded fabricated and rejected it. BMJ requires that researchers hand over data when asked, and Dr. Chandra didn’t supply the data.

The consequences of an accusation are grave. An allegation of misconduct “is a potential career killer,” notes a CRIC workbook. “The consequences of being convicted or even accused are serious, [so] due process and clear rules are required.”

A paper prepared for CRIC says the strengths of the Canadian system are that it’s non-bureaucratic, flexible, self-regulating and relatively inexpensive to operate. It says the weaknesses include insufficient protection for whistleblowers at institutions, insufficient information about misconduct occurrences, and no clear system for correcting the research record. It notes “inconsistencies in policy, procedures and practice across institutions … [with] no single standard definition of academic misconduct across Canada.”

Some people cite potential conflict of interest during investigations by universities. “Canada does not have a national agency that does this kind of investigation,” says Jack Strawbridge, who was Memorial’s director of faculty relations during the Chandra case and is now retired. “If a university was faced with an allegation against one of its own and it did an investigation, people are not as likely to believe the outcome of that investigation, because of the conflict of interest.” NSERC’s Dr. Lloyd suggests, “The universities should be concerned about protecting their image because they could be perceived to be in a conflict of interest. There’s a tendency to cover up the problems.”

Others disagree. “The implications of not dealing properly with a case of academic misconduct are so much more severe than the possible embarrassment of having a dishonest researcher,” says John Hepburn, vice-president, research, at the University of British Columbia. “I don’t see how it reflects badly on an institution that someone who’s done something wrong is caught and censured.”

One possible step in clarifying Canadian rules and responsibilities on academic misconduct would be updating the Tri-Council Policy Statement on Integrity in Research and Scholarship, say some CRIC members. In addition, three other avenues for action have been suggested.

The first is a compulsory, raw-data repository, one of the main recommendations of the Pencharz report done for Memorial on the Chandra affair. SSHRC requires that researchers be able to hand over data, although this policy has not been enforced. While a data repository might work for scientific data, some wonder what would happen with qualitative data (such as interview transcripts from social scientists, including historians and journalists). Also, a database could contravene other policies, such as the Tri-Council policy on research involving humans. Dr. Hepburn says that central storage of raw data “would just be an enormous bureaucracy” for a university like UBC that does $400 million a year of sponsored research. “I can’t imagine storing all of the data associated with all of that research in any way that would be retrievable,” he says.

A second avenue is whistleblower protection, a particularly crucial issue for graduate students. Ms. Trauttmansdorff of SSHRC says that the policy must ensure that students who work with researchers are “able to raise issues and problems without any fear of retribution or punishment.” The Canadian Federation of Students recently started a Whistleblowers Campaign on its website, with examples of ethical breaches in the private sector and what was done about them.

The third possibility is the most controversial: setting up a national body that would take over from the universities the role of enforcing policy and investigating allegations of misconduct. This kind of body exists in the U.S. for publicly funded health research (the Office of Research Integrity) and in Denmark. The United Kingdom, Finland, Sweden and Australia recently established national agencies or are exploring doing so, says Dr. Lloyd.

However, national programs have their drawbacks – for example, the U.S. office can investigate allegations of cheating only when the allegations concern researchers who used public money and not allegations about researchers who received funds from corporations or foundations.

“It’s way too early in the process to tell if a national body is needed,” says Ms. Trauttmansdorff. “It’s an area that covers federal, provincial and private interests, and how do you impose a system on those entities constitutionally or practically?”

A national body in Canada would have to negotiate with institutions that have various collective agreements and privacy rules. “You’d need a national body working for each discipline and for each collective agreement at each university,” argues Jack Lightstone, president of Brock University, who, as a faculty member at Concordia, was involved in drawing up a research-ethics policy for Quebec universities after the Fabrikant affair. A complicating factor is different scholarly traditions. “For example, supervisors in the humanities rarely claim second authorship, but in the sciences and medical sciences, it’s commonplace. … Enforcement would be impossible for a national body,” Dr. Lightstone says.

“The solution may not be centralized control or accreditation or central bodies undertaking this, but very, very good guidelines that we can follow and put into place at our institutions,” Western’s Dr. Hewitt says. “But the key word is ‘guidelines’ because every institution is different. … We work very closely with our faculty association and we have a good record. It works here. I don’t want to mess with that.”

It’s no wonder, then, that granting agencies and universities are talking more about education than enforcement. When it comes to preventing cheating, Brock does what many other institutions are doing: it educates faculty. “We run information sessions for new faculty. Most Canadian universities do,” Dr. Lightstone says. He’d also like to see “ongoing, voluntary training for faculty and obligatory teaching seminars for teaching assistants.” Queen’s University, for its part, will soon be requiring its graduate students to complete a non-credit ethics course in either principles of academic integrity or research involving humans.

But in the U.S., some disquieting studies show that education programs prove ineffective later in a career and may even promote misconduct early in one’s career. At the Lisbon conference, Melissa S. Anderson, director of the Postsecondary Education Research Institute at the University of Minnesota, presented her studies of mid-career and early-career researchers, showing that education has little relationship with subsequent behaviour among mid-career scientists and has both positive and negative associations with early-career researchers’ misbehaviour. For example, when the education comes from mentors who advise on how to survive in an academic field, the mentoring is associated with higher rates of misbehaviour and misconduct.

Fifteen years after the Fabrikant case, Dr. Pencharz wonders why it is taking so long for Canada to address research misconduct. He says that after he tabled his report for Memorial, “I thought nothing more would be done.” But now that the agencies and institutions are getting together under CRIC, he says, “it looks like maybe something will be. We Canadians move slowly.”

PUBLISHED BY
Alex Gillis
Missing author information
COMMENTS
Post a comment
University Affairs moderates all comments according to the following guidelines. If approved, comments generally appear within one business day. We may republish particularly insightful remarks in our print edition or elsewhere.

Your email address will not be published.