Professional Societies and Responsible Research Conduct
Mark S. Frankel
Recent disclosures of fraud and other highly questionable behavior in the conduct and reporting of scientific research have prompted scientists, their institutions, and the larger public to reexamine research practices and the present blend of formal and informal mechanisms intended to promote responsible research conduct. The traditional preference of scientists for autonomy over their own affairs as an alternative to increased public control makes it incumbent upon the scientific community to find ways to ensure that individual scientists are competent and perform according to high ethical standards. It is an effort that scientists are increasingly willing to undertake, both to ensure the integrity of science and to maintain public confidence in the scientific enterprise.
In this paper I examine the various ways by which scientific and engineering societies attempt to foster responsible research conduct. The focus is on what a sample of societies is doing, rather than on pointing to societies that have not undertaken similar kinds of activities. By documenting the former, my purpose is to identify a range of policies, procedures, and programs that might suggest approaches for other societies. I begin with a statement of why professional societies should and can play a role in fostering responsible research practices, followed by a description of the approaches adopted by the societies in implementing that role. I then report on the specific efforts undertaken by a select group of societies,1 first describing standards of conduct relating to responsible research practices and then highlighting society activities intended to reinforce those standards. I conclude with an
Mark S. Frankel is director of the Scientific Freedom, Responsibility, and Law Program at the American Association for the Advancement of Science in Washington, D.C.
assessment of the societies' efforts thus far and a discussion of additional measures that they might take to promote responsible research conduct.
THE PROFESSIONAL SOCIETY ROLE: THE RATIONALE
Although individual scientists must bear ultimate responsibility for their actions, promoting ethical conduct need not be solely the responsibility of the individual. Indeed, exclusive emphasis on the individual ignores the importance of social structures in shaping individual consciences and behavior. There is clearly a role for scientific and engineering societies to play in influencing the moral tone and ethical climate in which research is conducted.
"To be a professional is to be dedicated to a distinctive set of ideals and standards of conduct,"2 and the evolution of any profession is, in large part, characterized by its efforts to define the expected character and proper conduct of its members. Members of a scientific discipline, like other professional groups, are bound together by similar aspirations, values, and training, and as such are a community whose members "are distinguished as individuals and as a group by widely shared goals, beliefs about the value of those goals, … about the appropriate means for achieving them, and about the kinds of relations which in general should prevail among themselves, and in many cases between themselves and others."3 The scientific disciplines, then, are a prominent normative reference group, whose values and standards of appropriate research practices serve as guides by which individual scientists organize and perform their work and by which outsiders can understand and evaluate their performance.
The commitment of individual professionals to the values central to their profession is what leads society to grant the professional group as well as individual members the authority and resources to pursue their self-determined work in the public interest. The scientific community has been vested by society with the power to determine who may enter the community, what knowledge and skills must be acquired to achieve professional status as a scientist, and by what standards of conduct individual scientists will be judged. In large measure, then, a scientist is defined by his or her relationship to the group or discipline, and the professional community is charged with developing means for ensuring that individual members act responsibly. This reliance on self-regulation by the scientific group is consistent with the American tradition of limited government and has distinct advantages over the obvious alternative—public regulation.
Such regulation, as manifested, for example, in administrative rules, is typically designed to stipulate what cannot be done; it rarely prescribes what should be done. It defines the floor, not the ceiling of expected behavior. But surely we expect more from scientists as advocates for responsible research practices. By appealing to their moral consciences and their collective commitment to ensuring the integrity of science, we seek to evoke from scientists a higher standard of behavior than that which can be commanded through regulation. And when that evocation is supported by professional norms that represent a distillation of collective reflection and experience, the likelihood of ethical behavior is substantially increased.
Furthermore, there are several practices that most researchers would consider deplorable and capable of compromising the integrity of science, such as gift authorship, repetitive publication, and the selective presentation of research findings. Yet, these are not matters that ought to be subjected to the heavy hand of regulation. Rather, they are examples of practices that are more amenable to change through the process of critical self-examination that the professional community brings to bear on research practices and ethics, periodically reassessing them in the light of changing conditions and shifting perceptions of what constitutes proper behavior.
The scientific and engineering societies are distinct and easily identifiable institutions, and as visible, stable, and enduring entities, they act as the custodians of the disciplines' core values and distinctive traditions. They function as an important source of identity for individual scientists and engineers, enabling them to maintain a conception of themselves as members of a particular tradition rather than simply as technicians. And the collectivization of appropriate professional norms and their transmission by the societies to individual scientists can be an effective means of subordinating individual interests to the collective purposes of the discipline. Hence, while the profession "does not produce the next generation [of scientists] biologically, it does so socially,"4 and over time the behavior of individual members can be (and is) explained by references to it.
As publishers of major scientific journals, the societies are also well positioned to influence research publication practices directly, to serve as an influential forum for the open discussion of key ethical issues, and to educate scientists and engineers regarding acceptable research conduct.
Finally, the scientific society performs an important mediating influence between its members and outsiders. For members, the society is expected to be a strong voice in educating outsiders about the values and norms of the discipline and in securing public support for their
work. For outsiders, the scientific society is a countervailing force to the private purposes of individual members. "We rely on the group to guarantee that its members fulfill their agency obligations. … and we trust professionals because the exercise of … discretion at the individual level is governed by rules which are prescribed and enforced by the group."5 The societies are gatekeepers, whose oversight of the trust relationship between individual members and outsiders is critical to the advancement of science.
For all of the above reasons, the scientific and engineering societies deserve recognition and support for their role in fostering responsible research practices by their members.
THE SOCIETIES' SELF-ACKNOWLEDGED ROLE AND APPROACH TO PROMOTING RESPONSIBLE RESEARCH CONDUCT
As organized, self-governing units, the scientific and engineering societies have publicly acknowledged a role for themselves in promoting ethical practices by their members, as evidenced by the wide range of society policies and activities related to research ethics described in the next section.
The specific actions undertaken by the societies generally follow one of two approaches, or some blend of them. The most common approach for those groups examined here is for the society to accept primary responsibility for the adoption, application, interpretation, and enforcement of research standards, which may include the initiation of a broad range of supporting activities. Examples of this approach will be discussed in the following section.
The other approach is for the society to promulgate guidelines for the proper conduct of research, but to defer to others for their adaptation and application, i.e., to the institutions where the research is conducted. This is the approach adopted by the Association of American Medical Colleges (AAMC). In The Maintenance of High Ethical Standards in the Conduct of Research (June 1982), the AAMC affirms its belief "that faculties and their institutions have the primary responsibility to maintain high ethical standards in research and to investigate promptly and fairly when misconduct is alleged," and offers its set of guidelines as a foundation upon which local institutions can develop programs and processes for promoting ethical research conduct.6 This same approach also applies to the AAMC's 1989 Framework for Institutional Policies and Procedures to Deal with Misconduct in Research7 and to its 1990
Guidelines for Dealing with Faculty Conflicts of Commitment and Conflicts of Interest in Research.8
The Society for Neuroscience is an example of a scientific society that combines the two approaches. Its Policy on Scientific Misconduct "supports the principle that academic institutions should develop and have in place procedures to deal with allegation of scientific misconduct."9 However, the society also acknowledges "a special responsibility and interest surrounding those scientific activities for which it is directly responsible—publication of The Journal of Neuroscience and presentations at the Annual Meeting." So although the Society for Neuroscience has chosen not to promulgate general ethical standards for research conduct for its members, it has adopted guidelines governing articles or abstracts submitted for publication to its Journal or Neuroscience Abstracts.
PROFESSIONAL SOCIETY STANDARDS OF RESPONSIBLE RESEARCH CONDUCT
The adoption of ethical standards is a visible and explicit pronouncement of professional norms, which are central to understanding what constitutes proper professional conduct as well as the expectations about the kinds of character professionals should possess. Such standards embody the collective conscience of a profession and are testimony to the group's recognition of its ethical responsibilities. Moreover, the development and periodical revision of such standards also present scientists and engineers with the opportunity for critical self-examination regarding research norms in the light of changing conditions both inside and outside the professions. It is a time for testing the profession's established norms against the experience of its members and the priorities of the larger society.
There are several different functions10 that such standards can play in promoting responsible research conduct.
First, in the absence of guidelines of ethical behavior, scientists and engineers may experience anxiety or uncertainty about the kind of behavior that is expected of them in morally ambiguous situations. Standards of conduct can help professionals to evaluate alternative courses of action and to make more informed choices based on the collective experience and distinct traditions of their discipline.
Second, standards of conduct constitute a basis for evaluating the behavior of colleagues or for the public's evaluation of professional
performance, thereby serving as a means for holding individual scientists and engineers as well as the group accountable.
Third, standards that reflect widely held professional norms contribute to the socialization of new professionals into the distinct practices and traditions of the profession, thereby securing their support of them at an early stage in their careers.
Fourth, standards can promote responsible research conduct by making it an affirmative duty for scientists and engineers to report errant colleagues, thereby creating a monitoring system in which each professional assumes a responsibility for upholding the group's integrity.
Fifth, having research standards may make is easier for scientists and engineers to resist pressures from others that might otherwise lead them to cut ethical corners.
Sixth, with established standards in place, legislative, administrative, and judicial bodies may accord them considerable weight when adjudicating allegations of misconduct. As a result, the discipline's norms for responsible research conduct may receive further support in the public arena.
Given the potential value of research standards, this section continues by examining those standards adopted by a range of scientific and engineering societies. The focus will be on guidelines that go beyond mere exhortations to be honest and open in conducting and reporting one's research. Rather, I am more interested in standards that attempt to offer substantive guidance to researchers on such matters as authorship practices, plagiarism, training and mentorship, access to and retention and sharing of data, conflict of interest, treatment of confidential or proprietary information, the reporting of research findings, and the responsibilities of scientists in addressing error or misconduct.11
Publication is the hard currency of science—it is the primary yardstick for establishing priority, the chief source of recognition from one's peers, and the standard on which advancement of science is based. As pressures for publication increase, authorship practices have come under increasing scrutiny as scientists wrestle with issues of credit and responsibility.
Nine of the scientific and engineering societies (American Anthropological Association [AAA], American Association of University Professors [AAUP], American Chemical Society [ACS], American
Federation for Clinical Research [AFCR], American Political Science Association [APSA], American Psychological Association [APA], American Sociological Association [ASA], Ecological Society of America [ESA], Society of Neuroscience [SN]) studied have established new or proposed standards that address authorship. The standards typically address the proper ways for determining authorship and acknowledging contributions, and they describe the responsibilities of authors. With regard to establishing legitimate coauthorship, two criteria stand out clearly—that coauthors be those who (1) have made significant scientific contributions to the work and (2) share responsibility for the results. All other "lessor contributions" should be acknowledged in a footnote or in a special acknowledgments section. Some of the guidelines offer examples of the latter—clerical assistance, advising about statistical analysis, arranging for research subjects, and modifying a computer program. Only one society (SN) explicitly refers to "honorary authorship," which it considers "not appropriate,'' and four societies (AAA, ACS, APSA, ASA) specifically obligate authors to acknowledge the work of students.
Four of the societies (AAUP, ACS, AFCR, APA) address the responsibilities of authors. They delegate responsibility to the first or submitting author of a paper for its contents and the accuracy of all primary data, for determining all legitimate coauthors, and for specifying the order in which the authors' names appear. Two sets of standards (ACS, APA) require the lead author to obtain all coauthors' assent to coauthorship. Finally, two societies (ACS, AFCR) caution against reporting research results in a fragmented manner.
The purpose of these provisions on authorship seems to be twofold: to establish ways of properly allocating publication credit and for holding scientists accountable for the content and methods of their work. As such, they seek to reinforce the ethical principle of fairness, minimize inflated achievement claims, and increase the possibility of tracing questionable research practices to their origins.
While several societies refer in general terms in their ethics guidelines to the obligation of scientists and engineers to accord proper credit to the contributions of others, two (AAUP, American Historical Association [AHA]) have adopted independent statements specifically addressing plagiarism. Both statements stress the scholar's responsibility to acknowledge every intellectual debt.
The AAUP statement emphasizes that "greatest care" must be taken not to appropriate the work of students to the scholar's benefit. The AHA stresses the importance of good work habits as a shield against plagiarism; observing the basic rules of good notetaking and good writing will help scholars avoid the sloppy work that makes it difficult to guard against plagiarism.
Paralleling the mounting concern in the scientific community with issues related to data retention, access, and sharing, several scientific societies have adopted explicit policies or guidelines governing such matters as what data ought to be accessible to whom, the timing of such release, and the factors that might affect sharing (e.g., confidentiality pledges).
Data retention and sharing are viewed by the societies as essential for assuring scientific quality and for helping to distinguish error from misconduct. Six societies (AHA, AFCR, APA, American Society for Microbiology [ASM], SN, Society of Professional Archaeologists [SOPA]) have explicitly addressed the issue of data retention with varying degrees of specificity. The responsibility for retention is typically assigned to the principal investigator. Three of the societies (AHA, ASM, SOPA) call on the investigator to deposit raw data in some central repository, where it can be accessible to others, while another cautions that provisions be made for maintaining confidentiality when storing and disposing of records. Three societies take different positions regarding the period of time for retaining the data, with one (AFCR) prescribing an indefinite period, another (APA) stipulating a minimum of five years after publication, and the third (SN) defining the time frame as being as long as there is a reasonable need to refer to them.
Data access and sharing are matters that at least seven societies (AHA, APA, American Physical Society [APS], ASM, ESA, Institute of Electrical and Electronics Engineers [IEEE], SOPA) have addressed. Where the time at which data sharing should occur is referred to at all, the obligation to share follows the point at which the original investigator has completed his or her analysis of the data and should be consistent with the researcher's prior rights to publication. One society (APA) prescribes that researchers clarify in advance with all appropriate parties the expectations for sharing data.
Two societies (APA, SOPA) refer to the credentials of requesters of data, stipulating that they be competent or qualified, while only one
(APA) addresses the responsibility of the recipients of shared data by obligating them to "obtain permission from research participants, whenever possible, to utilize the data." One society (SOPA) places access by others to data into a specific time frame, declaring that after ten years, the researcher waives the "right of primacy with respect to analysis and publication of the data," which should then "be made fully accessible for analysis and publication by other[s]."
Three societies (APSA, ASA, ASM) note that any sharing of data should not incur more than reasonable costs and that the requester may be expected to pay those costs. Six (AHA, APA, APS, APSA, ASM, IEEE) explicitly identify either privacy claims, promises of confidentiality, or national security/classification and proprietary considerations as legitimate counterclaims to data sharing, with one (AHA) of the six stipulating that researchers "must challenge unnecessary restrictions."
These provisions concerning all aspects of data maintenance reflect clear intent to establish data retention and sharing as a legitimate professional responsibility of scientists. But the precise boundaries of that responsibility can be affected by such delimiting factors as privacy rights, confidentiality pledges, proprietary concerns, national security interests, the priority rights of the original investigator, the credentials of the requester, and the costs associated with sharing. Clearly, there are a number of competing interests at play here, and the societies, while assigning considerable value to data retention and sharing in the conduct of research, have recognized the importance of other factors in advancing science, and in deciding when, how, and with whom to share.
Training and Mentorship
The importance of training and mentorship has increasingly gained currency as an essential component in promoting responsible research practices. For example, the Institute of Medicine recently recommended that "scientific organizations … develop educational and training activities and materials to improve the integrity of research" and that academic institutions "monitor the supervisory and training practices of their faculty and research staff to ensure that adequate oversight is provided for young scientists."12 Also, effective July 1, 1990, all research training grant applications to the National Institutes of Health and the Alcohol, Drug Abuse, and Mental Health Administration must include a description of the types of instructional activities on the responsible conduct of research that will be incorporated into the proposed research training program.13
The scientific and engineering societies have also recognized the importance of training and mentorship by including them among the professional responsibilities of their members. Two of the societies (ASM, International Epidemiological Association [IEA]) declare that researchers should serve as exemplary role models for their trainees, to demonstrate by example their commitment to the highest possible ethical standards. The other approach, adopted by five societies (AAA, AAMC, AFCR, AHA, ASM), is to urge scientists to use the education and research settings as an opportunity to ensure that trainees and students understand the values and ethical prescriptions governing research. Three societies (AAMC, AHA, ASM) stress the value of informing students of the profession's ethical guidelines in the classroom.
A few of the societies elaborate on what is expected of mentors. One (AFCR) holds them responsible for supervising the trainee's design of experiments, reviewing all original data and overseeing their interpretation, and helping to develop reports of the results. In addition, mentors have an obligation to be sure that trainees are aware of government and institutional guidelines governing research. Two societies (AAA, ACS) obligate mentors to encourage and support students in their studies, with one (ACS) urging regular guidance, direction, and periodic evaluation of students/trainees, and help in developing initiative and independent thinking of supervises. Both societies address the responsibility of their members to advise and assist in career development. Finally, two societies (AFCR, ACS) identify obligations of students/trainees, which include maintaining honesty, integrity, and diligence in conducting research, and consulting with mentors with enough frequency so that they are kept informed of their progress or problems.
To the extent that the societies explicitly recognize professional responsibilities as part of the mentorship and training activities of their members, it is increasingly likely that resources and materials designed to help them effectively discharge those responsibilities will begin to emerge. There will be a need to test and evaluate diverse approaches and to disseminate information on particularly effective techniques for transmitting professional norms to students/trainees.
Conflict of Interest
The societies' concerns with conflict of interest, at least as reflected in their guidelines or standards, arise in the context of peer review as well as in the conduct and sponsorship of research. Twelve of the
societies examined had guidelines with provisions on conflict of interest. Four (APA, ASA, ASM, IEEE) were rather general in scope, urging scientists to recognize conflicts of interest, to avoid relationships that might precipitate such conflicts, and to disclose conflicts to affected parties. The other eight societies (Association of Academic Health Centers [AAHC], AAMC, ACS, American Institute of Professional Geologists [AIPG], APSA, ESA, IEA, Society for Epidemiologic Research [SER]) offered more detailed guidance.
In relation to peer review, there are prescriptions (ACS, APSA, ESA) that researchers decline to review the work of others where conflicts of interests are involved. One society (ACS) urges that the reviewer return the manuscript promptly, informing the editor of the conflict. Alternatively, the reviewer could return a signed review stating his or her interest in the work and deferring to the editor's discretion as to whether to accept it. Another society (ESA) prohibits members from "purposefully delay[ing] publication of another person's manuscript to gain advantage over that person."
Recognizing that conflicts of interest may bias the collection, analysis, interpretation, and reporting of data, other guidelines (ESA, IEA, SER) focus on the relationship between researcher and sponsor, stressing the importance of the researcher's independence and his or her moral obligation to hold the public interest above the narrow interests of sponsors where they seek to exert undue influence on the presentation of findings. Researchers are called upon to disclose all relevant financial, personal, or professional relationships that might lead to a conflict of interest—for themselves as well as for family members—to their institutions and in public speeches and writings, and to disclose such relationships related to the sponsor of the research (AAHC). Other provisions discourage arrangements involving confidential information that may not be shared with colleagues; prescribe policies that ensure that students and trainees are not exploited in the service of sponsored research; and address compensation arrangements in support of clinical studies, cautioning that payment not be linked solely to the enrollment of research subjects or contingent upon a specified outcome (AAHC).
One society (AIPG) addresses very specifically the relationship of its members to employers or clients. It admonishes members not to seek to profit economically from information gained without written permission of the employer or client, not to use his/her employer's or client's resources for private gain without the consent of the employer or client, and not to accept, without the employer's or client's written consent, an assignment by another if the interests of the two conflict.
Overall, the conflict of interest provisions of the societies examined reflect an attempt to balance the value of sponsored research against the
risks of conflicts of interest—real or apparent. They also seek to minimize the effects of bias on the peer review process. They have succeeded in identifying key areas of concern, but as yet have not sought to formulate more detailed criteria for recognizing a potential conflict and determining when disclosure is required.
Reporting of Research Findings
The expectation of publication based on research underlies most of the provisions pertaining to the reporting on one's results. Nevertheless, at least three societies (AAA, AAHC, ASA) have legitimized restrictions or delays in publication based on such factors as proprietary rights, patent preparation, or potential harm to clients, collaborators, or participants. For the most part, deciding when or how to publish—or when to trigger those factors that might justify delays or other restrictions—is the sole responsibility of the investigator.
Two societies (APA, ASA) proscribe their members from suppressing disconfirming or other significant data in their reports, while another society (SOPA) proscribes its members from entering into contracts that prohibit the scientist from including his or her interpretations and conclusions in the contract report. One society (APA) prescribes that scientists acknowledge the existence of alternative hypotheses; one (APSA) that members disclose any "material condition" imposed by sponsors or others on their research and publication; and another (ASA) that members state all significant qualifications on their findings in reporting their research; and yet another (IEEE) declares that members be "realistic in stating claims" based on available data. At least three of the statements (APA, APSA, ASA) prescribe that scientists acknowledge the sources of funding for their research in their public reports; another (AAA) simply prescribes "candor concerning sponsorship." One society (ASM) emphasizes the responsibility of scientists for the timely release of research reports, while another (SOPA) establishes a ten-year time frame for publication, after which the researcher waives his or her right of primacy with respect to publication.
While the societies view the reporting of research results as an integral part of the research process and a professional responsibility of scientists, at least in some cases they are prepared to accept constraints on the discharging of that responsibility. They have explicitly recognized a limited set of considerations that might justify publication delays or restrictions, leaving to the investigator the responsibility to determine when such considerations outweigh the prescription to publish.
Treatment of Confidential or Proprietary Information
Several societies have recognized that working with confidential or proprietary information creates special responsibilities on the part of scientists and engineers. There is a general disposition toward prescribing that confidential or proprietary information (including manuscripts under review) not be used or reported without permission from the persons (or their legal representative) from whom it was obtained. However, at least two societies (APA, ESA) identify explicit exceptions to this general prescription: where withholding the information would present a clear danger to others or where it is appropriate to comply with a legal requirement (APA), and where confidentiality would contribute to "unnecessary or significant degradation of the environment" as well as jeopardize public health or safety (ESA).
One society (AHA) calls on members to clarify the conditions of confidentiality prior to beginning one's work, to press for changes in the confidentiality requirements when they are "unsatisfactory," and to inform the readers of their publications of the rules of confidentiality that governed their work. Another society (APSA) imposes an obligation on researchers to seek changes in the law so that the confidentiality of sources "may be safeguarded." Yet another (APA) establishes a professional obligation to inform people at the outset of the limits of confidentiality that will affect their professional relationship. Finally, two societies (AIPG, SOPA) proscribe members from using confidential information obtained during the course of work for an employer or client in any way that adversely affects their interests, except with their consent or when disclosure is required bylaw.
The provisions referred to above are an attempt by the scientific and engineering societies to balance the traditional patterns of free exchange in science with promises to withhold certain information from public view. There is an assumption at work here that finds confidentiality agreements essential for some types of research to proceed. They are accepted without necessarily being encouraging. Having entered into such agreements, scientists are obligated, with a few exceptions, to honor them. To do otherwise is presumed likely to do more damage to science than would occur by the withholding of information.
Addressing Error or Misconduct
In addition to promulgating standards of conduct to promote responsible research conduct, scientific and engineering societies also address the researcher's responsibilities when he or she confronts abuses or errors, or the potential of such, by others. For two societies (AFCR, ESA), being responsible means acknowledging and correcting error when it is detected.
Another approach endorsed by some of the societies to promote responsible research practices and the integrity of science is for scientists to take seriously their self-policing responsibilities and speak out against improper practices and violations of research norms. One society (AHA) describes as "troubling" the reluctance of scholars to speak out about their suspicions of misconduct; one may infer from this a duty of scholars to be more forthcoming in reporting their suspicions. Six other societies (AAUP, AFCR, AIPG, APA, ASM, ESA) acknowledge more explicitly these responsibilities on the part of researchers. In the case of one society (APA), if the violation does not lend itself to an informal solution or is of a serious nature, then the scientist should bring it to the attention of appropriate committees of the profession. Another society (AFCR) encourages investigators to alert their laboratory chiefs or institutional officials when they know of a violation of the profession's standards. A third (ASM) obligates researchers to bring to public attention premature, false, misleading, or exaggerated statements, and to protect and cooperate with others who identify such misconduct. One society (SOPA) makes it a professional responsibility to report violations of its Code of Ethics "to proper authorities," while three societies (AAUP, AIPG, ESA) declare that suspected misconduct should be brought to the attention of the appropriate body within the profession, with two of them (AAUP, ESA) urging that affected parties also be notified.
By including such provisions as part of their conduct standards, the societies are encouraging scientists to view the reporting of misconduct as a positive step for maintaining the quality of science, rather than as an uncollegial act. They are acknowledging that dishonest work ultimately damages all of science and that it is in the enlightened self-interest, indeed an affirmative duty, of individual scientists as well as the community of scientists to voice their disapproval of scientific misconduct and to pursue such allegations conscientiously.
REINFORCING PROFESSIONAL STANDARDS
Many of the societies examined here recognize that the adoption of standards for responsible research is an important but not sufficient step toward fostering proper research practices. A set of standards must be viewed as only one part of a larger effort intended to promote responsible conduct. In seeking ways to reinforce their prescriptive role, the societies have at their disposal a range of activities relating to education, recognition, and enforcement.
The educational activities of the societies are intended to inform members of their standards and what the societies are doing to enforce them, to offer guidance to members in interpreting them, and to educate members in a more general way about the ethical issues confronting researchers and how the research community and others are responding.
Informing scientists about society standards takes many forms. The AHA plans to print its statements on ethics as a pamphlet for distribution to all academic departments of history. SOPA requires that all archaeologists accepting certification sign a pledge to abide by its Code of Ethics and Standards of Research Performance. A frequently used mechanism for alerting members to standards is the societies' journals and newsletters, where they will publish drafts of new standards or revisions for which member approval will be sought.
Society publications are also used to inform members of enforcement efforts. The APA, for example, requires its Ethics Committee to publish an annual report in its journal, American Psychologist, on the types of complaints investigated, cases adjudicated, and their disposition. And the AHA intends to begin publishing soon in its newsletter semiannual reports on cases of misconduct reviewed by the association.
The APA has published its "Rules and Procedures" for investigating, adjudicating, and reporting on alleged violations of its Ethical Principles in its journal, American Psychologist. Six of the societies examined for this study included such rules and procedures as a companion document to their standards, while others have incorporated them into their bylaws.
Some societies have taken steps to elaborate on their basic principles of research conduct as a means of helping members to interpret their application in specific situations. One approach is to publish more detailed, supplementary guidelines in particular areas of research. Prime examples of this are the APA's "Guidelines for Ethical Conduct in the
Care and Use of Animals," "Guidelines for the Use of Drugs in Research by Psychologists," and "Ethical Issues in Psychological Research on AIDS." Another approach is to publish illustrative cases, describing how they were handled by the society in light of its prescribed standards. This is the approach recently undertaken by the AHA and one that the APA has also pursued for several years through a series of casebooks, the latest of which was published in 1987.
Many of the societies examined for this study have sought to better inform and educate their members about the ethical issues associated with research by encouraging coverage in their society journals and newsletters and by sponsoring open forums at national or regional meetings. Others, such as the American Association for the Advancement of Science (AAAS) and the AAMC, have organized invitational workshops that have examined those issues in a rigorous fashion.
Some societies have issued special publications exploring in some depth one or more components of responsible research conduct. Examples include Sigma Xi's Honor in Science14 and ACS's Trade Secrets … Ethics and Law.15 The APS has distributed On Being a Scientist, a publication of the National Academy of Sciences,16 to every student member. Only one society contacted for this study (ACS) indicated that it was currently planning to develop educational materials dealing with science and responsibility.
One way to foster attention to the value that the research community attaches to responsible research conduct is to bestow public recognition on those scientists and engineers who exemplify model conduct in their own research, who display qualities of leadership in promoting responsible research practices among scientists and engineers, or who responsibly speak out against research misconduct. Although there appears to be no such recognition specifically intended for these purposes, there is at least one case where it has been accorded, and the potential for it to be done elsewhere.
The annual AAAS Scientific Freedom and Responsibility Award recognizes exemplary responsible behavior in science and engineering across a wide range of conduct, and in 1989 the AAAS selected Robert Sprague as co-recipient of the award. Sprague was the researcher who first suspected research fabrication on the part of psychologist Stephen Bruening and was cited for "his courage and persistence in reaffirming the highest standards of scientific integrity by initiating the censure of
a research colleague who fabricated data ….'' In 1990, the American Institute of Chemists (AIC) established a new "Ethics Award" to recognize outstanding contributors to ethics in the chemical profession. In addition, several other scientific and engineering societies have established awards for exemplary service to the profession or the public, or for contributions to the ethics of the profession, all of which could include efforts by researchers who, in one way or another, contribute in critical ways to ensuring the integrity of research.
A number of societies have designed procedures for disciplining members who have violated their ethical standards and, in fewer cases, for supporting members who are placed at risk by their efforts to live up to those standards. Besides disciplining or supporting particular individuals, these procedures can perform other functions as well. They send a message to all members as well as the institutions in which they work that the society considers deviation from the standards of responsible research a serious matter. This will put potential wrongdoers on notice that there will be a price to pay for misconduct, and it will increase the confidence of others in the integrity of their discipline's research base. Examples of disciplinary and support actions may also serve an interpretive function if the society's enforcement procedures allow for "opinions" to be issued.
For those societies with enforcement procedures, three types of approaches can be identified. One approach, adopted by the AAAS, does not link the procedures to any specific set of ethical standards adopted by the association. Rather, the AAAS uses the National Science Foundation's definition of scientific misconduct, and its procedures apply only to AAAS staff and their collaborators engaged in research or publication ventures.
A second approach is for societies to adopt enforcement procedures applicable to publication practices or the submission of abstracts for meetings and to implement them through journal editors, or publication or program committees. The ASM has published its procedures in its newsletter and instructions for authors; the SN has published its procedures as part of its Policy on Scientific Misconduct.
The most common approach used by the societies examined here takes the form of procedures to enforce their ethics guidelines or research standards. Nine societies have developed detailed enforcement procedures, only some aspects of which are briefly described here. In the case of one society (ESA), no enforcement mechanism applies to
regular members, but those members who seek certification renewal may have it denied if the society's Code of Ethics is violated. Seven societies assign specially constituted standing committees with responsibility for implementing their enforcement procedures, and the range of sanctions that may be applied include the following: private or public reprimand, censure, probation, suspension, denial of recertification, stipulated resignation, imposed rehabilitation or educational training, required supervision, and expulsion from the society.
Three societies (ASA, AHA, APSA) include the option of trying to mediate a disputed matter, while two (AHA and APA) will consider referring the matter to other organizations with a request for arbitration or resolution. Two societies (AHA, APSA) state that they will not normally pursue a complaint if it is under litigation, while another (APA) explicitly declares that litigation will not be a bar to its consideration of complaints. The other societies are silent on this matter.
There is wide variation in the position the societies take with regard to notifying other parties of the outcome of their investigations. One society (IEEE) leaves notification of the membership to the discretion of its Board of Directors; no reference is made to any other parties who might be notified. In the case of expulsion or stipulated resignation, the ASM identifies the member and the sanction in its newsletter. If the charges are dismissed, the accused is given the option of whether the decision is published. If the AHA decides that the matter is "indicative of a larger problem," it may publish an advisory opinion or guideline in the association's newsletter. SOPA provides for publication of disciplinary action, and its Board of Directors is given discretion as to whether to inform other "individuals, corporations, government agencies, and the media" of the results of disciplinary proceedings. Finally, the APA gives both its Ethics Committee and Board of Directors some discretion in notifying others of the outcome of its cases. Where the committee has imposed probation or suspension or has stipulated resignation of a member, it may inform members and other individuals or organizations (several are specifically listed) in order to "maintain the highest level of ethical behavior by members or to protect the public." The APA board is required to ''report annually and in confidence" to members the names of those who have been expelled or dropped from membership and the ethical principles involved. It is required to notify other parties if it "deems it necessary" to protect the public or to maintain the association's standards.
The IEEE's enforcement procedures explicitly offer assistance to members who strive to adhere to the institute's Code of Ethics, whose
livelihood is jeopardized by such efforts, or whose ability to discharge their professional responsibilities is compromised, or when the situation can be detrimental to the IEEE or to the engineering profession. The only type of assistance specifically offered in the documents reviewed is the submission of an amicus curiae brief in court proceedings. The IEEE Board of Directors may, at its discretion, publish findings or other comments in support of members and take whatever further action it deems appropriate. The APSA "promises to do all that it can within its resources to protect political scientists from unjustifiable abuses," relying on "persuasion and vigorous protest" as its primary means for supporting members.
Finally, the guidelines on research misconduct and conflict of interest issued by the AAMC and AAHC are accompanied by recommendations that mechanisms and procedures be established at research institutions to handle allegations of impropriety. The AAMC guidelines on conflict of interest suggest a process for disclosing and reviewing conflicts, and the framework for dealing with research misconduct offers lists of suggested sanctions and of parties that might be notified of the outcome of cases.
IMPROVING THE PROFESSIONAL SOCIETY RESPONSE
The scientific and engineering societies vary with respect to their history, their power and influence, their relationship to members, and their resources. Nevertheless, to some degree they all function as an important source of identity for individual members, and they are legitimate custodians of the disciplines' values and traditions. They are well positioned, then, to articulate ethical standards for professional conduct, and several sets of those standards were described earlier.
In almost every case the standards examined had been adopted within the past three to five years. They are an attempt to keep pace with recent changes in the practice of science that have permeated scientific research and publication and with changes in legal and regulatory requirements. They remain to be tested in the world of experience, but in principle they address critical issues facing scientists and engineers and offer prescriptions of what is expected of them.
But the adoption of such ethical standards does not guarantee their usefulness when caught in the cross-pressures of contemporary research. Such standards should be viewed as only one part, albeit an important one, of a larger system intended to promote responsible research conduct. One complementary strategy involves keeping the standards visible and relevant in the eyes of researchers. The societies can
accomplish this in a variety of ways, but the most common method employed is through discussions of research ethics at professional meetings and in the pages of their publications. While such efforts undoubtedly impart information, and perhaps insight as well, to members and ought to be encouraged, they are subject to the vagaries of editorial and program decisions about what ultimately reaches a larger audience. A useful complement to such activities would be a routine and systematic interpretative function.
The ethical prescriptions embodied in the standards are like blunt instruments; they must be sharpened by interpretation if they are to function as useful guides to responsible conduct. To accomplish this, the societies might publicize decisions rendered in cases of violations of their standards—a procedure now employed by only a few of the societies examined—with a detailed description of the reasoning used by the society in reaching its verdict. Published regularly, over time these decisions and commentaries will come to constitute a type of "case law" that breathes life into the society's ethical prescriptions and alerts members to the cumulative wisdom of the profession in applying the standards to various real-world dilemmas.
Another approach to interpretation would be to develop case materials—based on real or hypothetical incidents—designed for the education of practicing scientists as well as those in training. In pursuing this strategy for the latter group, the societies should work with their members based in colleges and universities in order to increase the likelihood that such materials will be developed.
Whether or not a society chooses to enforce its standards with a system to investigate and rule on allegations of misconduct and to levy sanctions is a decision that every society makes—either consciously or by default—as it considers the kind of relationship that it wants to develop with its members. But once committed to a system of enforcement, a society should mobilize the machinery necessary to carry out this function in an efficient and fair manner in order to earn the trust and respect of both members and outsiders. For members, this means designing a set of procedures that not only identifies violators but that also protects those who are falsely accused. For outsiders, the system must assure them that the society is prepared to acknowledge the possibility of scientific misconduct, to investigate allegations thoroughly, to hold researchers accountable, and to protect the public from the adverse consequences of improper research conduct. The ethical standards adopted by the societies not only define the boundaries of responsible research conduct; they also embody the virtues that researchers are expected to possess. Scientists and engineers are not only expected to act in a particular way; they are also expected to
exhibit a certain type of character. Hence, such standards reflect an ethics of character or virtue as well as an ethics of action. Recognizing such virtue in particular scientists and engineers offers the societies an opportunity to reward virtue, to call attention to the importance that the profession attaches to such character, and to publicly identify role models. The societies should seriously consider doing much more than is now done to confer recognition on scientists and engineers who exemplify the ideals of their discipline.
Finally, in considering how the scientific and engineering societies can more effectively promote responsible research conduct, one must keep in mind that such efforts incur costs in the form of time, energy, and resources committed to developing ethical standards, disseminating information, educating, registering disapproval, and conferring recognition. Consequently, the societies must be sensitive to what they can reasonably undertake at any particular point in time. This caution should not be interpreted as a prescription for inaction. Rather, it reflects a belief that costs, as well as more intellectual and professional factors, must be factored into the evaluation of alternative courses of action under consideration by the societies.
Twenty societies representing diverse areas of research were contacted by the author for information on their policies/standards and activities related to research ethics. This draft includes information from the 13 societies that have responded. (See Appendix A for a complete listing of the societies contacted.) In addition, the sections reviewing professional societies' standards draw from materials describing standards adopted or drafted by several other societies. (See Appendix B for a list of all standards and guidelines referred to in the paper.)
Jennings, B., D. Callahan, and S. M. Wolf, 1987, "The professions: public interest and common good," Hastings Center Report 17(special supplement), p. 5.
Camenisch, P. F., 1983, Grounding Professional Ethics in a Pluralistic Society, Haven, New York, p. 48.
Pavalko, R. M., 1971, Sociology of Occupations and Professions, F. E. Peacock, Itasca, Ill., p. 25.
Tuohy, C. J., and A. D. Wolfson, 1977, "The political economy of professionalism: a perspective," in Four Aspects of Professionalism, Consumer Research Council, Ottawa, p. 67.
Association of American Medical Colleges (AAMC), 1982, The Maintenance of High Ethical Standards in the Conduct of Research, AAMC Ad Hoc Committee on the Maintenance of High Ethical Standards in the Conduct of Research, AAMC, Washington, D.C.
Association of American Medical Colleges (AAMC), 1989, Framework for Institutional Policies and Procedures to Deal with Misconduct in Research, AAMC, Washington, D.C.
APPENDIX A—SOCIETIES CONTACTED FOR DATA ON RESEARCH ETHICS ACTIVITIES
American Academy of Forensic Sciences (AAFS)
American Anthropological Association (AAA)
American Association for the Advancement of Science (AAAS)
American Chemical Society (ACS)
American Historical Association (AHA)
American Institute of Chemists (ACS)
American Institute of Professional Geologists (AIPG)
American Political Science Association (APSA)
American Psychological Association (APA)
American Society for Microbiology (ASM)
American Sociological Association (ASA)
Association of Academic Health Centers (AAHC)
Association of American Medical Colleges (AAMC)
Ecological Society of America (ESA)
Institute of Electrical and Electronics Engineers (IEEE)
International Epidemiological Association (IEA)
Society for Epidemiologic Research (SER)
Society for Neuroscience (SN)
Society of Professional Archaeologists (SOPA)
Society for the Scientific Study of Sex (SSSS)
APPENDIX B—SOCIETY STANDARDS AND GUIDELINES CONSULTED
American Anthropological Association
Principles of Professional Responsibility, 1990
American Association of University Professors
Statement on Plagiarism, July 1989
Statement on Multiple Authorship, June 1990
American Chemical Society
Academic Professional Guidelines (Draft), April 1990
Ethical Guidelines to Publication of Chemical Research, 1986
American Federation for Clinical Research
Guidelines for the Responsible Conduct of Research, 1989
American Historical Association
Statement on Interviewing for Historical Documentation, May 1989
Statement on Plagiarism, May 1986; amended May 1990
Statement on Standards of Professional Conduct, May 1987; amended May 1990
American Institute of Professional Geologists
Code of Ethics, December 1989
American Physical Society
Resolution on Freedom of Scientific Communication, November 1983
American Political Science Association
Rules of Conduct, 1968
Advisory Opinions of the Committee on Professional Ethics, Rights and Freedoms
American Psychological Association
Ethical Principles of Psychologists, 1981; amended June 1989
Ethical Principles Revised (Draft), June 1990
Publication Manual, Third Edition, May 1986
Guidelines for Ethical Conduct in the Care and Use of Animals
Guidelines for the Use of Drugs in Research by Psychologists
Ethical Issues in Psychological Research on AIDS
American Society for Microbiology
Code of Ethics, 1987; amended 1988
Instructions to Authors for All ASM Journals, 1990
American Sociological Association
Code of Ethics, August 1989
Association of Academic Health Centers
Conflicts of Interest in Academic Health Centers, 1990
Association of American Medical Colleges
Framework for Institutional Policies and Procedures to Deal with Misconduct in Research, March 1989
Guidelines for Dealing with Faculty Conflicts of Commitment and Conflicts of Interest in Research, February 1990
The Maintenance of High Ethical Standards in the Conduct of Research, June 1982
Ecological Society of America
Code of Ethics, August 1990
Institute of Electrical and Electronics Engineers
Code of Ethics, August 1990
International Epidemiologic Association
Ethics Guidelines for Epidemiologists (Draft), 1990
Society for Epidemiologic Research
Statement on Ethical Issues Involving Conflicts of Interest for Epidemiologic Investigators (Draft), May 1989
Society for Neuroscience
Policy on Scientific Misconduct, 1989
Society of Professional Archaeologists
Code of Ethics
Standards of Research Performance, 1976