The committee found that existing data are insufficient to enable it to draw definitive conclusions as to which elements of the research environment promote integrity. The elements discussed in Chapter 2 appear to be associated with integrity in research, but the specific contribution of each element remains poorly defined. Empirical studies evaluating the ethical climate before and after implementation of specific policies or practices are lacking; as a consequence, the decision to implement particular programs is often based on anecdotal evidence. True misconduct is rare, and statistics on misconduct are approximate. Thus, looking for a decrease in rates of misconduct is not a viable way to assess the effectiveness of measures implemented to foster integrity in research. In addition, although it is relatively easy to catalog lists of policies and procedures, it is much less straightforward to measure performance and outcomes in the research environment.

Because of the limited empirical data on factors influencing responsible conduct in the scientific environment, the committee drew on more general theory (e.g., theories of organizational behavior, ethical decision making, and adult learning) to formulate the suggestions presented in this report. The findings and conclusions are based on the committee’s collective knowledge and experience after its review of the available literature in the science and business arenas as well as its discussions with experts who presented talks at the committee’s open meetings.

On the basis of the available information, the committee has described practices that promote the responsible conduct of research (Chapter 2) and has presented a theoretical model (Chapter 3) that contains many of the key components of the research environment and their interactivity. However, this is relatively new territory that needs to be examined with greater precision. Generating specific empirical data on integrity in scientific research is essential to help institutions determine the effectiveness of their efforts to foster a research climate that promotes integrity. Such data will also aid them in the development of better programs and policies in the future.

The request for applications issued by the Office of Research Integrity (ORI) of DHHS on May 2, 2001 (Research on Research Integrity. RFA-NS-02-005), is an important first step toward this goal, as it highlights a variety of potentially productive research topics, as does the ORI website (http://www.ori.dhhs.gov/html/programs/potentialrestopics.asp). The committee believes these topics are best studied in the context of the model presented in Figures 3-1 and 3-2. In addition to the important research questions identified by ORI in its program announcements, the committee identified additional topics that warrant further study.