Interpreting the Ethical Dimensions of Problems in the Research Setting

Ethical sensitivity involves the awareness by researchers of how their actions affect others. In addition to the ability to anticipate the reactions and feelings of colleagues, supervisors, research participants, and others, ethical sensitivity involves being aware of alternative courses of action and how each could affect the parties concerned. It also involves the ability to construct possible scenarios with knowledge of cause-consequence chains of events in the research environment. Ethical sensitivity requires empathy and role-taking skills. For individuals being socialized to the research setting, ethical sensitivity involves the ability to see things from the perspective of other individuals and groups (including other cultural and socioeconomic groups), and, more abstractly, from legal, institutional, and national perspectives. Thus, it includes learning the laws, regulations, guidelines, and norms of one's profession and recognizing when they apply. In professional settings, the focus is on ethical sensitivity, rather than the more general “moral sensitivity” described in the operational definition (Box 5-1), to signal the distinctive expectations of the researcher that derive from the norms and rules that govern research practice.

Research on ethical sensitivity in professional settings indicates that (1) ethical sensitivity can be reliably assessed, (2) students and professionals vary in their sensitivities to ethical issues, (3) ethical sensitivity can be enhanced through instruction, and (4) the sensitivity to issues is distinct from the ability to reason about issues (Bebeau, 2001). See Appendix B for a more extensive discussion of the findings from several professions that have studied ethical sensitivity in relationship to professional performance.

Developing Competence in Reasoning About the Complex Problems That Arise in the Research Setting

Ethical reasoning implies that professionals be able to critically analyze their own moral arguments and develop defensible points of view for new problems that are likely to emerge during the course of professional life. Studies of the reasoning development of individuals in a variety of professions (Rest and Narváez, 1994), including students in research training (Heitman et al., 2000), indicate that persons entering a profession are not equally able to apply moral ideals to the resolution of complex moral issues. In fact, some novices and even some experienced professionals are unable to reliably distinguish simplistic moral arguments that appeal to self-interest and the maintenance of interpersonal affiliations as guides for moral action (referred to as a “personal interest moral schema”) from arguments that appeal to societal laws and rules as a basis for deciding what should be done (a “maintaining norms moral schema”). In contrast, some professional students and research trainees are as able as persons with training in moral philosophy to work out what should be done in circumstances in which conflicts of interests exist. The mark of mature moral reasoning is the ability to figure out how to fairly modify existing rules or laws to accommodate the new moral problem that has emerged (referred to as “postconventional moral thinking”) (Rest et al., 1999).

The effects of ethics instruction on a professional's moral reasoning has been extensively studied (Rest and Narváez, 1994). In medicine, for example, Self and Baldwin (1994) have reviewed a wide range of studies that have used the Defining Issues Test (DIT) or other measures of moral judgment to assess reasoning development. They concluded that (1) a medical curriculum without an ethics curriculum tends not to enhance moral reasoning; (2) instruction can be effective, although not all interventions produce significant gains; (3) the effects of at least some interventions can be attributed to an intervention based on comparisons with control groups; (4) strategies other than discussion of a dilemma can produce change; and (5) there is a relationship between reasoning and a range of indicators of physician performance.

Although other intervention strategies can produce a change in reasoning, the most consistent effects in professional education have been achieved with a teaching and assessment strategy that incorporates the dilemma discussion technique (for example, significant change for 14 of 15 cohorts of dental students [Bebeau, 2001]). Over a 10-year period (1983 to 1993), Bebeau and colleagues tested the strategy, incrementally adding instructional elements to improve students' ability to develop well-reasoned written arguments for addressing solutions to problems that students commonly encounter. The greatest improvements were achieved when students were provided with criteria for judging the adequacy of arguments and multiple opportunities to develop well-written arguments both before and after case discussions and when they received feedback on the strengths and shortcomings of their arguments from peers as well as from the instructor (Bebeau, 1994).

In a reexamination of intervention effects, Rest and colleagues (1997) found two different effects of an intervention. One was the acquisition of new thinking (increases in preferences for postconventional arguments— the effect that researchers have typically reported); the second effect was systematic rejection of simplistic thinking (decreases in preferences for personal-interest arguments). Rest and colleagues concluded: “From a practical educational point of view, both kinds of developmental progress are desirable: gaining more sophisticated moral thinking and also becoming clearer about what ideas should be totally rejected for their simplistic and biased solutions” (Rest et al., 1997, p. 500). As a consequence, researchers using DIT are encouraged to assess interventions in terms of moral judgment profiles (i.e., the proportion of arguments that appeal to each of the moral schemas) rather than just in terms of advances in postconventional thinking (Bebeau, 2001).

In the early 1990s, researchers at the Poynter Center designed Moral Reasoning in Scientific Research: Cases for Teaching and Assessment (described below), a series of cases for teaching and assessment that incorporated the instructional techniques first tested in dentistry. When experienced researchers review these types of cases, as well as others included in available casebooks, they may judge the cases as too simplistic and be tempted to discard them in favor of discussions around contemporary issues that present highly challenging dilemmas. A danger in limiting teaching of responsible conduct to a discussion of contemporary cases is that students may learn the rules for specific situations but not be able to generalize to other issues of immediate relevance. Moreover, by focusing entirely on contemporary cases, students may not acquire the skills needed to identify the fallacies in their own arguments or to deal with many of the future unanticipated issues. Thus, the committee encourages faculty to develop a curriculum that provides opportunities to apply the more general moral reasoning ability that develops as a result of higher education to the specific problems that arise in the research setting (i.e., to develop “ethical reasoning”). On the basis of the original work of Kohlberg (1984) that was expanded by Rest, Bebeau, and colleagues (Bebeau, 1994; Bebeau et al., 1995; Rest et al., 1986), the committee defines ethical or moral reasoning as the ability to systematically examine a situation and then choose and defend a position on the issue (Bebeau et al., 1995). Arguments are evaluated in terms of the respondent's ability to describe the following:

• ethical issues and points of conflict, including precedents, principles, rules, or values that support prioritizing one interest over another;
• the stakeholders or parties that have a vested interest in the outcome of the situation;
• the probable consequences of possible courses of actions; and
• the ethical obligations of the central characters.

There is an important distinction between the focus in the development of cases designed to promote the sensitivity just discussed and those designed to promote ethical reasoning. Unlike cases for ethical sensitivity—in which finding and understanding the conflict (i.e., becoming sensitive to the conflict) is the focus—with ethical reasoning one is presenting a conflict that is relatively easily identified and interpreted. It has been shown that instruction in ethical reasoning is effective in increasing the ability of emerging professionals to engage in such tasks (Bebeau, 2001).

Moral Motivation and Identity Formation

The third component in the Four-Component Model of Morality (Box 5-1) acknowledges that individuals have a number of legitimate concerns that may not be compatible with the moral choice. Financial and career pressures, established relationships, and idiosyncratic personal concerns, among many others, compete for the researcher's attention. Blasi (1985) notes that people differ in how deeply moral notions penetrate their self-understanding and in the kinds of moral considerations that are judged to be constitutive of the self. In other words, moral motivation varies. This requires the attention of educators. Understanding that one is responsible provides the bridge between knowing the right thing to do and doing it.

Blasi (1985) and Kegan (1982) see identity formation as a lifelong developmental process. Recent work on identity formation based on Kegan's developmental theory (Forsythe et al., in press) suggests that at least 30 percent of graduates from West Point have not achieved key transitions in identity formation that would enable them to have the broad, internalized understanding of and commitment to codes in the responsible conduct of research and other professional standards. Whereas such individuals may see codes and professional standards as guides for behavior, they are likely to conform to the guides simply to garner rewards and avoid negative consequences, without achieving an understanding of their personal responsibility. Forsythe and colleagues concluded that “[professional] development programs will not be successful in instilling desired values in less mature [preprofessionals] unless the broad educational environment in which they operate promotes identity development toward a shared perspective on professionalism” (Forsythe et al., in press). Evidence from studies of role concept development in dentistry (Bebeau, 1994) support these observations.

Recent work on integrity in research is directing attention to the need for more formal efforts to socialize trainees and beginning researchers to professional expectations and values. For example, Braxton and Baird highlight the need to socialize researchers to the role of self-regulation, arguing that “doctoral study can be configured so that future scientists are prepared to participate in the deterrence, detection, and sanctioning of scientific wrongdoing” (Braxton and Baird, 2001, p. 593). The responsibility for self-regulation would be addressed as part of identity formation, whereas the actual skills would be taught as part of survival skill education (see below). The need for such socialization is further confirmed by Anderson's (2001) study of doctoral students' conceptions of science and its norms. She concludes: “The theme of individual, independent work that runs through these interviews suggests that students might not be subject to as much osmotic group socialization as many faculty assume. It is also clear that the channels by which socialization to the normative aspects of academic life are communicated are primarily informal. Calls for more formal, more deliberate approaches to normative socialization find support in the vagueness with which students conceptualize the norms that underlie academic research” (Anderson, 2001, p. 6).

Anderson's study is continuing and is expected to demonstrate the extent to which these norms develop in a academic setting.

Developing Self-Regulation, Self-Efficacy, and Implementation Abilities Necessary for Effective and Responsible Research Practice

Fundamental to responsible conduct in any profession is the ability to perform the complex tasks of the discipline with integrity, i.e., to have acquired survival skills. When the committee defined integrity in research (Chapter 2), it defined an aspect of moral character and experience. The fourth component in the Four-Component Model of Morality (Box 5-1) attends to the importance of character to the effective and responsible conduct of research. Integrity, ego strength, perseverance, backbone, toughness, strength of conviction, and courage are also qualities required for effectiveness as a researcher. A researcher may be ethically sensitive, may make good ethical judgments, and may place a high priority on professional values; but if he or she wilts under pressure, is easily distracted or discouraged, or is weak willed, a moral failure may occur because of a deficiency in character and competence.

Professional educational programs assist individuals in understanding the broad fundamentals of their disciplines; gaining some depth in the details of a particular subarea; and obtaining practical experience in research, including experimental design, methodology, data analysis, and other practices detailed in the definition of integrity in research. Fischer and Zigmond (1998) point out that although graduate education has as its purpose the development of a range of practices relevant to the specialty, such programs often lack an essential dimension (Bloom, 1992; Widnall, 1991), the development of a set of general professional skills (Fischer and Zigmond, 1998; Magner, 2000; NAS, 1995b). For example, scientists should be able to present their results at scientific meetings, defend their chosen methodologies and interpretations of data, and prepare written reports. They need to be able to learn from critical comments and suggestions from their professional peers both at oral presentations and through peer review of manuscripts. They may need to obtain grants to fund their research, hire and supervise technical staff, and teach classes and advise individual students. Moreover, with the decrease in tenured positions in academia, science apprentices should learn how to find information on and prepare themselves for other types of careers (NAS, 1995b; Varmus, 1995). Morever, within this broader range of careers, skills other than those specific to the collection and analysis of data may become all the more essential (Cordes, 1997; Greenwood and Kovacs-North, 1999; NAS, 1995b).

Principles of Adult Learning

This section briefly describes six learning principles that should be considered when developing an educational program on the responsible conduct of research.

1.

Education is best provided by individuals who have a deep understanding of their subject matter and whose teaching reflects that they care about and value the material being taught (Wlodkowski, 1999). Teaching of the responsible conduct in research presents a special challenge because it requires a synthesis of ethics and science. When scientists and ethicists collaborate in the design and implementation of learning experiences, students come to appreciate the complexity of problems that arise in the practice of science. Furthermore, when instruction requires the application of norms (and the ethical theories that support them), values, and rules and regulations to the practical problems that arise in the day-to-day practice of science, learning is more likely to last and to transfer to new situations. It follows, then, that instruction in the responsible conduct of research by a team of faculty—or by a faculty member with expertise in both ethics and science—is optimal.

When faculty take time from their scholarly work to provide practical instruction that draws on expertise from related fields, they demonstrate the importance of this educational task and its relevance to the practice of science. Only faculty with a deep understanding of the complexities of the related disciplines can answer questions with concrete examples, avoiding the mistake that many so-called experts make in instructing adults: that simply knowing something is enough to teach it effectively (Wlodkowski, 1999). Turning over a course in the responsible conduct of research either to an ethicist with no understanding of the current practice of science or to a scientist who has not taken the time to educate himself or herself about the various processes involved in ethical decision making conveys to students that the subject matter under discussion is peripheral to the current realities in the field.

2.

Educational programs in responsible conduct of research should occur over an extended period; indeed, they should occur throughout a trainee's tenure at an institution. It is clear from educational research that “spaced review and practice” (learning that occurs over an extended period, with frequent opportunities for practice and feedback) leads to greater retention of information and longer-term behavior change than “massed learning” (learning that occurs intensively during a brief period). Furthermore, some learners take a great deal more time to learn something than others do. Academic learning time—that is, time spent actively and successfully involved in learning—is strongly related to achievement (Fisher et al., 1980). Such findings suggest that programs limited to several hours of instruction during a single day, or even spread over a few weeks, are unlikely to have much of a long-term impact.

3.

Active participation in problem-oriented learning is an important component of effective educational programs. This principle argues for learning that is experiential and contextual. The teaching of rules and regulations and the testing of knowledge outside of the context in which they are applied not only violates this principle but also is devalued by self-directed learners who are well aware of the limits of learning that is disconnected from its application. Experiential learning enables participants to immediately apply and test their new skills, resulting in a more rewarding and effective learning experience (Darkenwald and Merriam, 1982; Dickenson and Clarke, 1975; Knowles, 1970; Tight, 1996).

Researchers have also found that adults are self-directed in their educational objectives (Cross, 1979; Even, 1981) and seek learning experiences that are directly applicable to their lives (Burgess, 1971; Carp et al., 1974). Moreover, giving students an opportunity to use language, orally and in writing, facilitates learning, both because of the active involvement inherent in such assignments and because it helps learners link prior knowledge to the new information that they are learning (Lemke, 1995). This principle argues for teaching to take place in small, interactive groups, and it further suggests the importance of problem-oriented assignments that require application of learning.

4.

Programs will be more effective if educators help students assess their prior knowledge and integrate new material with familiar ideas. People bring different knowledge to the learning situation (von Glasersfeld, 1989), and individual learning involves transfer of that prior knowledge to the new situation (Tobin and Tippins, 1993). To encourage new learning, a teacher needs to help students assess their prior knowledge and integrate new material with familiar material. Teachers who learn what their students know, and who know how their students learn, are in a position to teach for conceptual change (Loucks-Horsley et al., 1998; Strike and Posner, 1992). Effective learning requires that trainees be knowledgeable about themselves, and that they have frequent opportunities for assessment and feedback from instructors who are knowledgeable about their trainees.. This typically requires the use of small groups or, even better, individual conferences. In the context of the responsible conduct of research, it requires that trainees have opportunities to assess their strengths and shortcomings with respect to the broad abilities that contribute to competence in ethical decision making.

5.

Students should be encouraged to share their own experiences with others in the class. This not only seems to improve participant satisfaction but also provides a richer learning experience for all participants. Each student comes to the learning experience with factual knowledge as well as experiential learning connected with social roles, such as those associated with sex, race, class, and other affiliations. Course participants should be seen as de facto instructors as well as students. In this way, additional issues, information, and perspectives will be incorporated into the curriculum (Bruffee, 1993; Even, 1981; Merriam and Caffarella, 1999). This principle demands that some portion of a program of education in the responsible conduct of research involve interactions with and among the participants in a relatively small group.

6.

Instructional programs that attend to developmental differences and individual learning preferences are more likely to be effective. Individuals entering graduate education differ in their levels of intellectual development (King and Kitchener, 1994), moral development (Kohlberg, 1984; Rest, 1983), and identity formation (Blasi, 1985; Kegan, 1982). Developmental psychologists have shown how to construct learning experiences that promote developmental progress, and they have shown the negative consequences of learning and learner perceptions when instructional strategies are not appropriate to the learners' developmental levels.

Although developmental differences are of prime importance for promoting the intellectual and moral dimensions of integrity in research, other individual differences also warrant consideration when designing educational programs. Furthermore, individuals vary in their preferred learning styles (Burgess, 1971; Dickenson and Clark, 1975; Knox, 1978). Many educators believe that learning is likely to be enhanced by accommodating different learning styles and different levels of intelligence through the use of a variety of teaching formats, including lectures and discussions, individual as well as group exercises, and both oral and visual stimuli (Parker and Rennie, 1998). Thus, a “one-size-fits-all” approach to education in the responsible conduct of research is not likely to be adequate.

Need for an Integrated Approach

Learning relies on the interconnection of four learning environments: the learner-centered, knowledge-centered, assessment-centered, and community-centered environments (Bransford et al., 1999; The Cognition and Technology Group at Vanderbilt, 2000). Learner-centered environments build on what the students bring to the classroom, their strengths, and their prior learning. Knowledge-centered environments help students construct new knowledge by providing organized disciplinary knowledge and the skills needed to use that knowledge. In assessment-centered environments, both students and teachers set goals, ask for feedback, and make revisions as needed. Community-centered environments establish the normative behaviors for individuals in the learning institutions and other aspects of their professional lives, as well as in the other communities of which they are a part.

Each of these learning environments influences the others (The Cognition and Technology Group at Vanderbilt, 2000; Mentkowski, 2000). Thus, an effective program of education in the responsible conduct of research requires a broad, multicomponent approach. The following section describes the key elements of this approach.

Incorporating Current Practices in an Effective Curriculum

The background section of this chapter outlined the committee's rationale for recommending that education in the responsible conduct of research be taught in the context of the overall educational program.

Formal education in the responsible conduct of research can be provided in a wide variety of ways. Some of the most common approaches are discussed here.

Integration of Relevant Ethical Issues within the Core Curriculum

The committee believes that education in the responsible conduct of research should be provided within the core curriculum of a discipline, with ethics cases strategically selected to promote development of each of the abilities that will enable responsible conduct. There are two reasons for this. First, if responsible conduct is an integral part of conducting research, as argued above, then it should be infused through the educational program for new researchers. Second, many topics do not fit logically into a more general context of the responsible conduct of research but nevertheless deserve attention. Depending on the discipline, such topics might include informed consent (for the use of human subjects in research), the use of animals in research, data management, storage and retrieval of data, and ethical issues related to developing technologies (e.g., human cloning, gene therapy, and reproductive technologies).

Issues related to the responsible conduct of research can be integrated into a core course in two complementary ways. First, individuals teaching the courses can include comments on the ethical dimensions of a subject as part of the lectures. Thus, a faculty member discussing genetic markers for disease might comment on the ethics of genetic testing, whereas a faculty member teaching a course in anthropology might mention the possible adverse impact of fieldwork on the lives of indigenous people. This can occur without any special fanfare but as a natural component of the discussion of a given topic.

Second, time can be set aside to discuss an ethics case of particular relevance to the focus of the course. This discussion might be included as part of the series of breakout groups that are often organized for core courses to focus on a particular research article. Faculty teaching in one or the other formats can provide opportunities to learn to lead a discussion that promotes ethical reasoning or to lead a group activity that promotes ethical implementation (several texts with case studies are listed in Appendix D in the section Responsible Scientific Conduct).

Although the committee wholeheartedly supports the concept of integrating education in the responsible conduct of research into the core curriculum, the committee also recognizes that accomplishing this is no simple matter. One or two faculty members can mount their own educational program in the responsible conduct of research. But influencing the courses that other faculty members teach demands practical and time-consuming diplomacy. Moreover, it is an initiative that requires continual maintenance. There is often constant pressure on a course director to find time for new topics. In addition, the faculty and directors for a given course may change frequently. In both instances, issues of the responsible conduct of research can quickly disappear from the curriculum.

Florida State University, for example, offers an interdisciplinary science course (for graduate or undergraduate credit) that integrates the responsible conduct of research in science into each of its sections (Gilmer, 1995, 1998; Gilmer and Rashotte, Dec. 1989/Jan. 1990). The course uses historical examples, including the development of the atomic bomb, the Tuskegee syphilis trials, and scientific freedom and responsibility, to highlight the importance of science and the profound influence that science has on society. Current examples of integrity in science are seen within such a historical context. Assessment is a critical aspect of such a course, and writing is highlighted, with students reviewing and critiquing each other's ideas on a course website. Students use electronic portfolios to document their learning in the course (Gilmer, 1995, 2002). Students are given the option of selecting for one of their collaborative group projects a service activity that fits into the goals of the course. This way of bringing the consideration of integrity in science into the curriculum incorporates the four learning environments: the knowledge-centered, student-centered, assessment-centered, and community-centered environments (Bransford et al., 1999; The Cognition and Technology Group at Vanderbilt, 2000).

Ethics in the Context of Education in Professional Survival Skills

Earlier in this chapter, instruction in survival skills was described as one of the keys to the development of an environment that promotes integrity in research. The committee made this choice for three major reasons. First, virtually every aspect of any curriculum has an ethical dimension, and, as already noted, the committee believes that these issues are best identified and addressed in context. Professional skills, like more traditional aspects of a curriculum, have ethical dimensions. For example, when one is teaching about writing research articles, discussions might include issues of plagiarism, honorary authorship, data selection, and graphic design, while a workshop on grantsmanship might include a discussion of the importance of not overstating the sensitivities of methods or the quality of pilot data, exaggerating the assistance that one obtains from colleagues, or promising more than can be accomplished. As in the case of education in the responsible conduct of research within the core curriculum, the teaching of responsible conduct within a program of professional development can and should be done in two ways: through the inclusion of material in the lectures and through the active discussion of ethics cases.

Second, many of the ethical issues discussed in survival skills workshops are simply not likely to arise as topics either in the core curriculum or in traditional courses on the responsible conduct of research. Examples of such issues include the responsibility to publish worthwhile data in reports so that others may benefit from the work (particularly if the work was supported with public funds); the importance of acknowledging the contributions of others in oral presentations as well as in written work; and the responsibility to ensure that other researchers can replicate published results, by providing a complete and accurate presentation of methods and by being willing to share all reagents not commercially available.

The third reason stems from the fact that institutional climate appears to be an important determinant of responsible conduct in science (see Chapter 3). This may reflect, in part, the fact that most people are likely to learn less from what a faculty member or an institution offers as formal instruction than from the actual behavior that is observed. Offering instruction in survival skills is a clear indication that the faculty and the institution accept some responsibility for the professional advancement of their students. However, programs in professional skills should not replace or reduce the efforts of individual advisers to provide individual mentoring.

Other Venues

This brief discussion has not exhausted the ways in which instruction on the responsible conduct of research can be made an integral part of conducting scientific research. For example, authors should include issues related to integrity in science (including cases) in the textbooks of scientific disciplines (e.g., see Kovac, 1995; Tobin and Dusheck, 2001; and Zigmond et al., 1999) education directors should make ethical issues a component of annual retreats. Responsible conduct of research also should be a subject for online forums in areas of research (e.g., Fischer and Zigmond, 2001a; Zigmond and Fischer, 1995), and discussions of issues related to the responsible conduct of research should appear as part of the programs of professional meetings (for example, they are standard features of the meetings of the Society for Neuroscience and the American Association for the Advancement of Science). The objective is straightforward: to ensure that teaching of the responsible conduct of research exists side by side with discussions of all other aspects of science.