Our suggestions for the Research on Research Integrity program - Pimple and Adams
My colleague, Doug Adams, and I submitted this letter yesterday in response to the Request for Information to solicit Suggested Research Topics for Future Initiatives for the Research on Research Integrity Program. The deadline for submissions is February 1, 2012. – Ken
January 25, 2012
To whom it may concern:
In response to your Request for Information, we are writing to suggest a topic for the Research on Research Integrity Program:
Research Integrity-Lessons from Criminology
Since 2004, we have encouraged the research community to draw upon the mature literature of criminology for ideas about how to understand, prevent, reduce, and otherwise deal with research misconduct and questionable research practices (Adams and Pimple 2004, 2005). The key suggestion is to apply “opportunity theories“ that emphasize situational factors that have been shown empirically to encourage or discourage criminal behavior. We believe that many of these findings can be suitably adapted to the context of research.
As a simple example, both common sense and empirical evidence support the notion that crime and other bad behaviors flourish in secrecy. The recent revelations of the massive scale of Diederik Stapel‘s fabrication of data was facilitated by the control he exerted over his data sets: “Many of Stapel’s students were simply given data to analyze and graduated without having ever run an experiment“ (Vogel 2011:579). Stapel reportedly “gathered“ all of the data himself and never showed raw data to anyone. In such a situation, it is easy to get away with fabricated or falsified data; with other people collecting data and/or examining the raw data, it is much harder.
We suggest the following sub-topics for future research:
A. The Situational Nature of Misconduct
As we have suggested previously (2005), research misconduct has historically been approached as an individual rational choice. Under this model, the primary approaches to misconduct are to prevent it through education (convincing people to make the right choices) or punish it after the fact. We advocate considering the social and/or situational factors that influence not only potential offenders‘ behavior, but even what choices present themselves as viable.
Thus, we suggest investigating under what circumstances and in what situations known instances of research misconduct have taken place and how they compare with research situations for which no misconduct has been found or alleged. Relevant data may include the size of the research group and the command structure (span of control); the area of research; the funding status of the research group; the proximity of milestones (graduation, securing the Ph.D., job search, promotion, tenure, etc.); the degree of networking among researchers (persons with small support networks are known to be more likely to commit white collar crime); particular research practices, such as sharing original data; etc.
B. Early Intervention Systems
The potential for at-risk individuals to engage in misconduct of various types, in a wide range of professions, is often addressed through early identification and training. As we have suggested (2005), the profession of law enforcement (and increasingly physician behavior in clinical settings) confront this dilemma through the use of Early Intervention Systems (Walker, 2003). It is quite likely that investigators at risk for misconduct (as demonstrated by questionable research practices (QRP), evidence of stress, a sudden increase in secretive behavior, etc.) might be identified through unobtrusive administrative measures and corrective measures initiated by the researcher’s supervisor (lab director, PI, department chair, etc.). However, the degree to which QRP and misconduct are correlated, and whether there is any causal relationship, is still empirically unclear.
C. Responses to Misconduct and QRP
Three factors are often considered when attempts are made to deter crime: Severity of the punishment for the crime, swiftness of the punishment (the sooner after the crime, the better), and the certainty that the crime will be detected and punishment will be forthcoming. Although administrators and politicians often focus on the severity of punishment, overall the probability that a person will be deterred from committing a crime is correlated with his or her perception of the certainty of being discovered and punished (Tittle and Rowe, 1974).
Too little is known about whether research institutions respond to allegations and findings of misconduct in a way that is likely to deter future misconduct. Questions to be pursued in this light include: How do research institutions deal with the aftermath of misconduct investigations? Are other researchers at the institution informed of the outcome (thus potentially deterring future misconduct)? Is there any attempt to analyze the situational and other factors that might have led to the misconduct? Are suggestions made for changes in the future in the effected research group and elsewhere to prevent future misconduct? Are policies and procedures scrutinized? Are the hard-earned lessons learned shared with other research institutions and the research community?
These questions are couched in terms of data collection and analysis; they could also be re-configured to be applied to experimental (interventional) research in which, for example, a hypothesized “best practices” regime is instituted in one research group and compared to historical or concurrent control groups. Further, the efficacy of pro-active “random audits” of researchers might be considered as well.
A finding of misconduct is generally believed to be career-ending. For young scientists (graduate students), this could be argued to be premature as they have yet to establish their own identity as researchers. For more advanced researchers (postdoctoral researchers and beyond), ending a research career could be argued to be a waste of talent, not to mention the public money spent educating and nurturing the guilty researcher. We should ask whether researchers known to have engaged in misconduct are unworthy of potential rehabilitation and reintegration into the scientific community. In the criminology literature, this issue has been addressed under the label of reintegrative shaming (Braithwaite, 1989). Perhaps components of this literature could be adapted and applied to the domain of research misconduct.
E. The Measurement of Research Misconduct and QRP
As we have suggested (2005), despite a variety of attempts to measure misconduct, we still know too little about the distribution, frequency, and severity of misconduct. Criminology addressed this issue of unknown and/or unreported offenses through the development of the National Crime Victimization Survey (NCVS). Perhaps a similar initiative would be as successful at characterizing misconduct and QRP. Such an undertaking would also allow for the measurement of situational factors.
We would be happy to talk or correspond with you on refining our ideas for a future RRI announcement. You can reach Pimple at 812-856-4986 or email@example.com.
Kenneth D. Pimple, Ph.D. Director of Teaching Research Ethics Programs, Poynter Center Indiana University Bloomington
Douglas Adams, Ph.D. Associate Professor of Sociology and Criminology University of Arkansas
Adams, Douglas, and Kenneth D. Pimple. 2004. “The flaw in the ointment: What research on research ethics can learn from research on crime.” Presented at the ORI Research Conference on Research Integrity, San Diego, California (November).
Adams, Douglas, and Kenneth D. Pimple. 2005. “Research misconduct and crime: Lessons from criminal science on preventing misconduct and promoting integrity.“ Accountability in Research 12(3):225-240.
Braithwaite, John. 1989. Crime, Shame and Reintegration. (Cambridge University Press; Cambridge, UK.)
Tittle, Charles and Alan R. Rowe. 1974. “Certainty of arrest and crime rates: A further test of the deterrence hypothesis.” Social Forces 52(4):455-462 (June).
Vogek, Gretchen. 2011. “Psychologist accused of fraud on ‘astonishing scale.’” Science 334:579 (November 4).
Walker, Samuel. 2003. Early Intervention Systems for Law Enforcement Agencies: A Management and Planning Guide. (Washington, D.C: U.S. Justice Department)