Meeting on Research Integrity
to Encourage Improved Behaviors

Recent studies show that some 15% of post docs will falsify data in order to get funding, giving urgency to a meeting planned for this weekend to address the need for a new and developing field of study: empirical research into research integrity.

This weekend, AAAS and the federal Office of Research Integrity (ORI) will host a second conference on research integrity, to consider the extent to which scientists engage in research misconduct and the factors that contribute to such behavior, and to explore ways to encourage high standards of ethical practice in the scientific community.

Mark S. Frankel, Director of the Scientific Freedom, Responsibility, and Law Program at AAAS has worked with ORI for several years on its conference program committee, helping to decide which papers should be presented. AAAS also contributes some funds that enable young researchers of modest means to attend the conference.

"Time is past due for this field," Frankel says, "and nothing could be more central to the AAAS mission of 'advancing science and serving society.'""

Whereas the first Research Conference on Research Integrity had an audience full of administrators and policy makers, the second will focus more on researchers who want to do work on research integrity. Because the field is so new, the ORI hopes that attendees will bounce ideas off each other, making new contacts and coming away with new ideas and methodologies on the best way to measure and promote research integrity.

"In order to promote research integrity, it is essential that we know more about existing efforts and attitudes about how best to foster ethical research," Frankel says. "This Research Conference on Research Integrity has that as its goal--to stimulate rigorous research in order to increase our knowledge of 'what works and what doesn't,' and how we can more effectively develop activities that can be shown to contribute to integrity in scientific research. The conference will showcase some of the best research on the topic, both nationally and internationally."

The ORI also administers a grant program (for two-year studies) for researchers to evaluate and collect data on research integrity. Seven of the grantees who received funding a year ago will present their papers at the conference this weekend; five will present empirical evidence to show how financial conflict of interest can influence research results.

With jurisdiction over all DHHS-funded institutions, the ORI has as its ultimate goal to maintain and promote public trust in the scientific enterprise, says Chris Pascal, ORI Director. The intent is not to overburden the research system, but to create a better policy regarding demands placed on institutions in order to safeguard research integrity. Possible solutions include working with scientific societies, and training researchers on responsible research conduct.

One major challenge lies in semantics: the scientific community has not decided on a clear and accepted definition for the meaning of "to commit misconduct." Pascal would like to see to have a common understanding of the term that is then made as explicit as possible.

In the next round of ORI grantees, Pascal hopes to see a greater focus on qualitative studies to answer the question of why researchers engage in research misconduct.

Although the ORI is mostly concerned with institutions that are funded or connected with the US Department of Health and Human Services, Pascal notes that the private sector could also benefit from the responsible application of ethical guidelines. He cites the impact on the scientific community and on public trust of the recent announcement that a Bell Labs researcher had falsified data on several published papers.

In this case, an awareness of the importance of research integrity might not have stopped the individual scientist from misbehaving. Pascal suggests, however, that clearly-stated, enforceable standards of research integrity might have made this individual's co-workers more vigilant.

Of the Bell Labs affair, Pascal says "many people in the scientific community believed that there was not as much research misconduct in the physical sciences, because the methodologies were more precise, data was harder to falsify, and you didn't have to be as concerned about the human error. This goes to show that someone smart can deceive people for a very long time."

Nick Steneck, ORI consultant, believes that the first step is to provide scientists with basic guidance on how to behave.

"If some researchers have poor publication practices, if they are sloppy in authorship, are they even aware that there are practices written for them?" Steneck asked. "The forthcoming conference will help to offer some reinforcement for those researchers who want to do the right thing."

—Carol Hoy

15 November 2002