11
Report of the Committee on Academic Responsibility

Massachusetts Institute of Technology

Summary

The Committee on Academic Responsibility was charged to:

  1. review the current situation with respect to the community values in connection with the conduct of academic research;

  2. review our existing policies and procedures in connection with the conduct of research in view of the values held by the community;

  3. compare our existing policies and procedures with guidelines and regulations of federal and private research sponsors;

  4. suggest innovative education and mentoring programs directed towards raising the consciousness of our community concerning issues associated with the conduct of research and also propose mentoring programs related to faculty career development.

The committee found widespread recognition of our dual responsibility: that of educating the next generation of scientists and scholars for their professional responsibilities and of ensuring that the research and scholarship done on our campus meet the highest standards of integrity. All of us need to have a clear appreciation of the basic values of science and scholarship, and we must articulate these values clearly to our students.

We found that principles of ethical research conduct are not often explicitly discussed during the early phases of education of young scholars. Rather, individuals are left to develop their own personalized code of behavior, based in part on personal values and in part through specific examples set by their mentors. We believe that members of the faculty must develop an enhanced level of awareness of ethical

NOTE: Dated April 15, 1992; reprinted with permission from the Massachusetts Institute of Technology, Cambridge, Mass.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II 11 Report of the Committee on Academic Responsibility Massachusetts Institute of Technology Summary The Committee on Academic Responsibility was charged to: review the current situation with respect to the community values in connection with the conduct of academic research; review our existing policies and procedures in connection with the conduct of research in view of the values held by the community; compare our existing policies and procedures with guidelines and regulations of federal and private research sponsors; suggest innovative education and mentoring programs directed towards raising the consciousness of our community concerning issues associated with the conduct of research and also propose mentoring programs related to faculty career development. The committee found widespread recognition of our dual responsibility: that of educating the next generation of scientists and scholars for their professional responsibilities and of ensuring that the research and scholarship done on our campus meet the highest standards of integrity. All of us need to have a clear appreciation of the basic values of science and scholarship, and we must articulate these values clearly to our students. We found that principles of ethical research conduct are not often explicitly discussed during the early phases of education of young scholars. Rather, individuals are left to develop their own personalized code of behavior, based in part on personal values and in part through specific examples set by their mentors. We believe that members of the faculty must develop an enhanced level of awareness of ethical NOTE: Dated April 15, 1992; reprinted with permission from the Massachusetts Institute of Technology, Cambridge, Mass.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II issues that confront scholars at all levels of experience, and provide for a more explicit and systematic discussion of these issues with their students. The responsibility to ensure systematic discussion of these issues rests with the departments, and we make recommendations for educational programs based in departments. We define three behaviors in the conduct of research that merit Institute attention. The first is research misconduct. We define research misconduct as fabrication, falsification, and plagiarism in proposing, conducting, or reporting research or other scholarly activities. Other types of misconduct that can occur in a research setting but which are not unique to research activities are differentiated from research misconduct and defined as general misconduct. In addition, there is a range of questionable or improper research practices that we do not include in either research misconduct or general misconduct, but which can negatively affect the research enterprise, compromise the responsibilities of universities, and violate ethical standards. We present a set of generic research practices and urge discussions in departments and laboratories to establish field-specific details and to determine at what thresholds deviations from these practices constitute improper or questionable research practices. We believe that discussing such research practices in research groups will contribute to our educational programs and that most disputes arising within groups about deviations from good practice should be resolved by informal discussions or mediation. We see an important role for informal mediation by faculty in departments and schools and have made recommendations to facilitate this. However, allegations of research misconduct cannot be informally resolved nor are they proper for a process of mediation. We have made recommendations on institutional response to allegations of research misconduct, placing the responsibility for initial inquiry with the department head but providing central resources to ensure proper procedures and institutional memory. We have discussed and made provisions to protect the rights of the accused to a fair, confidential, and objective process and to ensure that those who bring allegations of research misconduct responsibly and in good faith are protected from retaliation and damage to their careers. Finally, we believe that a period of stability in federal regulations is appropriate to enable universities to gain experience in the application of procedures to ensure the integrity of research done on their campuses.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II Charge and Committee Procedures The Committee on Academic Responsibility was established jointly by the president and the provost in May 1991 with the charge to: review the current situation with respect to the community values in connection with the conduct of academic research; review our existing policies and procedures in connection with the conduct of research in view of the values held by the community; compare our existing policies and procedures with guidelines and regulations of federal and private research sponsors; suggest innovative education and mentoring programs directed towards raising the consciousness of our community concerning issues associated with the conduct of research and also propose mentoring programs related to faculty career development. In this report, we set out what we believe to be the consensus of the MIT community regarding the values that must be upheld in research conduct. We make specific recommendations for programs of education in research conduct. We discuss the regulatory environment in which scientific activity must now function. We propose a definition of research misconduct and make specific recommendations for procedures to deal with allegations of research misconduct. This report is presented from a faculty community to our faculty colleagues and to the MIT administration. We present our recommendations and intend that these will be translated into policies and serve as a basis for the development of procedures. We intend that our report will serve as a basis for further discussion among members of the community and for the development of educational and mentoring programs. We believe that these actions will allow MIT to respond effectively to the rapidly changing environment. We have not discussed all details of procedures that fall within our charge nor addressed all of the federal regulations by which MIT is bound but only those which relate to important issues involving the responsibility of the Institute for research integrity, the role of the faculty in this process, and the rights of individuals caught in contentious situations. We expect that policies and procedures in this area will develop in an evolutionary manner as we gain experience. We began our deliberations in May, 1991. We met with many members of the MIT administration, faculty, graduate students, and postdoctoral fellows and associates and reviewed a substantial body of literature dealing with the issues of responsibility in the conduct of

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II scientific research and scholarly inquiry. We read transcripts of congressional hearings and media coverage of developing cases. Our report benefited specifically from the ongoing study of scientific responsibility by the National Academy of Sciences. We reviewed procedures used by other universities to address allegations of research misconduct. We commissioned a study of education in research ethics that gathered educational materials of general usefulness, surveyed other universities to determine what courses and programs are in place or planned, generated a number of scenarios illustrating difficult issues that arise in the application of principles of good research practice, and produced a document containing material that may be useful to departments for their educational programs. Copies of this document are available from the committee. Members of the community were most helpful to us, generously giving of their time and sharing openly with us their perceptions and experiences as they impinge on these issues. We benefited from descriptions of activities already under way in several departments and schools to deal with the issues raised herein, and from reports of relevant experiences elsewhere and lessons learned. In August [1991] we presented an interim report that was widely distributed throughout MIT. Many individuals came before us to discuss various aspects of these issues in the light of that report. Findings and Conclusions The committee found widespread recognition of our dual responsibility: that of educating the next generation of scientists and scholars for their professional responsibilities and of ensuring that the research and scholarship done on our campus meet the highest standards of integrity. We found that principles of ethical research conduct are not often explicitly discussed during the early phases of education of young scholars. It is critical that members of the faculty, both senior and junior, develop an enhanced level of awareness of ethical issues that confront scholars at all levels of experience, and provide for a more explicit and systematic discussion of these issues with their students. Programs dealing with the ethical conduct of research are most effectively carried out in departments and research groups. We defined three behaviors in the conduct of research that merit Institute attention: research misconduct, general misconduct, and questionable or improper research practices. Each requires a unique institutional response.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II Generic research practices provide a framework for discussions in research groups about research conduct; most disputes arising within groups about deviations from good practice should be resolved by informal discussions or mediation. Faculty have an important role to play in informal mediation of disputes and in acting as advisors to individuals with concerns about research conduct. However, allegations of research misconduct cannot be informally resolved nor are they proper for a process of mediation. Effective institutional response to allegations of research misconduct in research carried out at MIT places the responsibility for initial inquiry with the department head but provides central resources to ensure proper procedures and institutional memory. We must protect the rights of the accused to a fair, confidential, and objective process and ensure that those who bring allegations of research misconduct responsibly and in good faith are protected from retaliation and damage to their careers. Finally, we believe that a period of stability in federal regulations is appropriate to enable universities to gain experience in the application of procedures for carrying through with their responsibility to ensure the integrity of research done on their campuses. Summary of Recommendations As a result of our deliberations and findings we make the following recommendations: That the MIT faculty and administration make explicit their commitment to academic integrity and to the establishment and maintenance not only of proper research conduct but also of an environment in which both research and teaching can be carried out effectively. That each department form a working group to reflect on current practices, the values they promote, and changes in practices that would improve education and research, particularly with respect to the specific research conducted by members of that department. That MIT establish a series of workshops on research conduct; that these workshops be organized at the level of departments, laboratories, or research groups and be of a size to ensure that individuals have an opportunity to speak; that these workshops be held periodically to provide new members with an opportunity to become familiar with the traditions and procedures

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II of the group; and that attendance at these workshops be encouraged. That MIT move to establish procedures for mediation as a part of its procedures for dispute resolution and that consideration be given to application of the principles of mediation in the inquiry process when appropriate. That each department designate individual faculty members to serve as advisors and informal mediators. That MIT define research misconduct as fabrication, falsification, and plagiarism in proposing, conducting or reporting research or other scholarly activities. That a single set of internal procedures including standards of proof, and rights of complainants and accused among others be used for the investigation of all allegations of research misconduct involving faculty and staff. That the responsibility for inquiring into allegations of research misconduct be vested in heads of departments and interdepartmental laboratories or comparable administrative units; that this normally be done by setting up a fact-finding panel whose report provides the basis on which the head decides what further steps are appropriate, including a recommendation to the provost that a formal investigation is warranted. That the department head submit all proposed plans and procedures for inquiries into allegations of research misconduct to the Office of the Provost for approval before the process is initiated; that the process to be followed in conducting inquiries and investigations be the responsibility of a specially designated individual(s) in the Office of the Provost; that the person(s) so designated be responsible for developing guidelines to be followed in carrying out inquiries and investigations. That MIT ensure a supportive environment for individuals who come forward with concerns about research conduct, and that specific provisions to ensure the protection of complainants who act in good faith be a part of the plan for conducting an inquiry into allegations of research misconduct and be submitted to the Office of the Provost before the inquiry is initiated. In our report, we also make many suggestions and observations that we feel will improve the environment for research and education on our campus and improve the procedures for responding to allegations of research misconduct.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II I. Introduction In our role as a teaching institution as well as a research institute, we have a dual responsibility: that of educating the next generation of scientists and scholars for their professional responsibilities and of ensuring that the research and scholarship done on our campus meet the highest standards of integrity. In our discussion with members of the MIT community we have found widespread acceptance of these responsibilities. There is agreement that we must transmit the values of science and scholarship and the specifics of good engineering and research practice to the next generation—both to the undergraduate and graduate students in our classes and to the postdoctoral fellows and junior faculty. It is widely understood that formal instruction is only a part of the educational process and that the core experience in the education of almost every scientist and scholar is to be found in the informal teaching—one-on-one, more often than not—that goes on outside the classroom and officially scheduled academic exercises. Since the atmosphere in the different research groups and the relationships among their members is central to this process, constant attention must be paid to the consequences that actions of individuals and their informal behavior may have on this informal learning process. We believe that the establishment of our committee represents an opportunity for the MIT community to engage in discussions about the shared values it holds in the conduct of research and in the education of students, and we recommend that the MIT faculty and administration make explicit their commitment to academic integrity and to the establishment and maintenance not only of proper research conduct but also of an environment in which both research and teaching can be carried out effectively. We doubt that a direct cause-and-effect relationship between the environment for research and the occurrence of research misconduct can be established. Rather we assume that occasional allegations of research misconduct will occur in a large institution with an intense research focus such as MIT, and the Institute and its faculty must be prepared to deal effectively with these difficult issues. We make recommendations about education in research conduct because it is part of our educational responsibility to our students and will improve the climate for research and scholarship on our campus. Although in our deliberations we concentrated primarily on research in science, broadly defined as the physical, biological, and social sciences and engineering, we have also had discussions with members of the Schools of Humanities and Social Science, and of Architecture and Planning, and conclude that the issues of professional conduct

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II encountered by these colleagues are not fundamentally different from those encountered by teachers and practitioners in science and engineering. In particular the values we discuss and the need for education in these values are not limited to individuals engaged in scientific research but are of crucial importance to the entire MIT community. We intend our discussions of research integrity to apply more broadly to scholarship and scholars throughout the Institute, including creative activities such as design in our definition of research. In some cases we must speak more specifically to science in responding to regulations governing the use of federal funds or in discussing research practices. II. The Changing Environment for University Research The last half century has seen the creation of a uniquely American institution, the research university, of which in many respects MIT is the prototypical example. Like universities of past generations, the modern research university pursues twin objectives: transmitting to the next generation the knowledge and understanding that mankind has gained in the course of its history, and extending the frontiers of what is known and understood. The relative importance of the latter objective has dramatically increased. In the modern research university, and in MIT in particular, innovative research is the engine that drives the entire enterprise. The spectacular successes that American science has achieved in the last half century were obtained largely through research conducted in universities. This work was performed predominantly with funds supplied by agencies of the U.S. government. Although the U.S. government had previously provided funds to universities—e.g., under the Morrill Act of 1862 and subsequent legislation—the level of government support for university research increased sharply after 1940 under a unique partnership between universities and government. The changes that have taken place in the political and economic situation of the world in the last decade—such as the collapse of the Soviet system, the emergence of Japan as the world's most dynamic economic power, the budget and banking crises, and the worsened economic conditions in the United States—have fundamentally altered the rationale that has justified the relationship between the U.S. government and the major research universities. The universities—and science in general—are perceived by many as not as central to the national interest as they were during World War II or after the launch of Sputnik, when science was seen by both the government and the

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II public as essential to our national survival. Today, science is perceived by some as yet another interest group whose claims to public funds must be severely scrutinized. Headlines we have seen in the papers during the last few years exemplify this changed attitude. However, since science is essential to the solution of many of the problems faced by the world, it is vital that the public's esteem for and trust in science be maintained. In addition to changes in the relationship between the research university and its chief sponsor, the U.S. government, the last decade has also seen major changes in social relations—in particular, relations between individuals differing in race, sex, and position in the hierarchy. Science places considerable value on the autonomy and the contributions of the individual, and therefore it is expected that individuals would continually challenge the system to ensure recognition for their contributions and to ensure the development of their future careers. Hierarchical, paternalistic structures in university research laboratories are less likely to escape challenge by today's graduate students and postdoctoral associates. Federal laws and regulations governing the treatment of personnel and the environment for career advancement affect the freedom of action of laboratory directors and individual investigators, as do MIT's own policies with respect to our responsibilities to students, faculty colleagues, and Institute staff. All of us need to understand better the changes in the environment for the conduct of research, and we need to respond effectively to these changes. The changes that have taken place during the last decade require that we modify and correct procedures and attitudes that do not respond to the new reality. All of us need to have a clear appreciation of the basic values of science and scholarship, of our responsibilities for transmitting them to the next generation, and of the many ways in which these can be compromised. We must not only articulate these values clearly but also internalize them as an essential part of our lives. III. Values in Research Research is the attempt to reveal principles or laws that govern observed phenomena. The highest standards of conduct and practice are necessary to assure the integrity of the results. Values essential in research conform to those that ideally govern behavior and activities in the general society. Among these are honesty, performing one's craft with skill and thoroughness, respect and fairness in dealing with others, and responsibility to people and institutions.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II Honesty is the foundation of scholarship. Deception in the proposing, conducting, and reporting of scientific and scholarly research subverts this enterprise. Skill and thoroughness, and other aspects of craftsmanship, are essential elements in conducting research and advancing a field. Good research requires good research practice; departure from this principle is often the cause of nonproductive scientific dispute. While it is clearly desirable to be first in reporting research results, this should not be done at the cost of ''cutting corners." Scientists must take appropriate care to ensure the integrity and accuracy of their work. An important aspect of research practice is the proper reporting of the results of one's work. Data, procedures, and controls must be fully disclosed in publications to allow the experiment to be replicated and the results and conclusions to be evaluated. Criteria used to select the data presented should be explained and defended. Such disclosures are essential to ensure the proper functioning of the system by which the priority, credit, and support for research are decided. Errata should be promptly submitted to correct errors discovered after the publication of results. While research is inherently a risky enterprise, every effort must be made to minimize error. One way to decrease the probability of error is to make the research data available to all collaborators for their review. As a minimal requirement, each coauthor should be prepared to take responsibility in his or her area of expertise for the evaluation of data and procedures as well as for the conclusions of the paper. Ideally, all authors should be able to take responsibility for and to defend the conclusions of the paper as a whole. Research data should be retained for a reasonable time after publication to allow for examination by others. Respect for and fairness to others requires that researchers be scrupulous in assigning proper credit for intellectual accomplishments. Significant research contributions by individuals in a group project must receive acknowledgment through authorship on publications, or other suitable means. While there are varied practices with regard to authorship, fairness requires that each author should have made a significant contribution to the work. Specialized contributions that do not merit authorship should be acknowledged. In addition, the published results of others used in research publications should be properly referenced. Education is the primary function of a university, and it must play a significant role in university research activities. The education and development of postdoctoral fellows and associates and graduate students in research are as important as obtaining research results. Faculty have the responsibility to communicate to the next generation of scientists the

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II values that govern research practices as well as knowledge and research expertise in their fields. Although errors in science can be reduced by adherence to good research practice, their total elimination is probably not possible. Errors generally create scientific disputes and are ultimately rectified by the self-correcting mechanisms inherent in the scientific enterprise. While most fraudulent research can be expected to be corrected by these same mechanisms, research misconduct is so damaging to science and scholarship that the public record must be corrected whenever it is identified. This requires an appropriate institutional response when research misconduct is alleged. Research misconduct is a violation of the trust that society places in the scientist. In order to search for truth, the scientist is privileged to be granted resources in a compact with institutions, government, and society in general. Research misconduct is a betrayal of this compact. When trust erodes, so does support. In addition, research misconduct can have harmful practical consequences. It is wasteful of resources and time: not only the resources used by the offending scientist, but also those used by other scientists who attempt to verify or extend fraudulent results. When fraudulent results influence medical, technical, and political decisions, they can have harmful consequences to society in general. Secrecy is antithetical to the tradition of university research that basic knowledge obtained in research and scholarly endeavors should be available to all. Since the education of young scholars comes in part from participation in the debate that typically occurs in a collegial research environment as new ideas and results are described, proprietary and classified research in universities is detrimental to the objectives of education. Faculty engaging in such research are not able to divulge resulting ideas and knowledge to students and colleagues in general, eliminating this part of their efforts from the educational mission of the university and thus reducing their effectiveness as teachers and as mentors. In addition, students and postdoctoral associates participating in this type of research are not able to get appropriate credit and recognition for their work in open publications and meetings, which can be highly damaging to their careers. While we recognize that a certain degree of confidentiality might be understandable before results are published, we were concerned by reports that competition among groups and individuals has sometimes resulted in the imposition of excessive restrictions on the free exchange of information, even among faculty and students in the same department. Such informal "classification" of information in a research area cannot help but interfere materially with the effectiveness of teaching.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II because of its effects on the department. In some departments it may be impossible to select a committee without perceived bias. For such reasons, nondepartmental committees bringing the necessary expertise may be the best choice. The accused and complainant should have an opportunity to challenge the composition of the inquiry committee. The charge to the inquiry committee should be in writing and should be as specific as possible given the allegations or other evidence. We believe that, whenever possible, the charge to the committee should be limited to determining the facts and the substance of the allegations and should not charge the committee to recommend whether a further investigation should be carried out. That is, we are suggesting a separation in the role of fact-finder and adjudicator. If no evidence of research misconduct is found by the fact-finding committee, then their report can serve as a basis for mediation of the dispute should the parties involved decide to enter into such a process. If the committee find that the allegations of research misconduct have substance, the report of the fact-finding committee provides the basis upon which the department head makes a recommendation to the provost as to whether an investigation should be carried out. There are several reasons for this separation in roles. First, it limits the scope and responsibility of the inquiry committee, charging them to focus on the key elements of their task: evaluation of evidence and finding of fact. It places the judgmental role with the department head, and the provost. It does not burden the committee with recommending a particular administrative outcome. It should reduce the potential for tension between committee members and other departmental members who may not agree with the final outcome based on the limited information available to them. The committee should be briefed concerning the Institute guidelines for inquiries, including evaluation of evidence, burden and standards of proof, and the level of certainty of committee findings to be achieved. The committee is not asked for a finding of facts that misconduct occurred, since the Institute must proceed to an investigation whenever allegations are found to have substance, that is, "if there is reason to believe." The committee would gather, hold, and examine all evidence, including original data as appropriate, and would allow the accused to present evidence in writing and to meet with the committee. The evidence that such a committee would be expected to gather and evaluate includes all forms of data that faculty members have competence to evaluate: research data in its various forms, direct testimony from witnesses, publications and drafts, financial records, correspondence, logs, and other laboratory records.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II Inquiry into the possibility of research misconduct should not be conducted as an adversarial process between an accused and a complainant. The accused has the right to contest all of the assertions brought against him/her but not to challenge the particular individual who brought them. In some cases the committee might not elect to meet with the complainant. However, in some cases, because of eyewitness testimony, dispute about the interpretation of physical evidence, or other issues, the participation of the complainant would be required. We expect that in most cases the committee would take testimony from both the complainant and the accused in separate closed sessions. The accused should receive a copy of the charge to the committee and the evidence against him/her and be allowed to present evidence on every key point. We believe that given the preliminary nature of inquiries, and the many uses that MIT makes of inquiries, that attorneys should not be present at inquiries. Since the only definitive outcome of an inquiry is a finding that no misconduct has occurred, there is no finding of misconduct by the accused. We have also suggested procedures to insulate any subsequent investigation from the inquiry process to protect the rights of the accused. The role of the complainant during the inquiry and later investigation, if any, deserves careful consideration. One possibility is to have a two-branch process. In one branch, the individual who brings evidence to the department head may wish to have no further involvement with the case. If substantial evidence of misconduct is presented in the allegation on which a determination can be made without the involvement of the individual bringing an allegation, then that individual's participation is not required. In many cases, because of career pressures and fear of retaliation, this would be the preferred course. For example, a graduate student could take evidence of plagiarism to a department head who would then act on behalf of the Institute. In this case the Institute acts as the complainant, and there is no requirement that the initial complainant be made known to the accused. This individual plays no further role in the process: would not be called as a witness, would not continually furnish information, and would not be informed about the progress of the case. In the second branch the complainant becomes a principal in the case, putting forward the initial allegation, providing documentary evidence and testimony to the inquiry and investigation committees, and receiving and responding to sections of committee reports that deal with issues raised by the complainant. If the case proceeds to a formal

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II investigation, the identity of this person would become known to the accused. The latitude of the inquiry is an issue. It should be neither a freewheeling inquiry into every possible issue involving the accused nor need it be constrained to deal only with the issues originally raised in the initial allegation. If in the course of a careful examination of the evidence directly related to the initial allegation, the inquiry committee discovers evidence of possible misconduct not known by the complainant, then this becomes part of the inquiry, and the evidence and the findings of facts should be reported to the department head as part of the committee report. The accused should be kept informed of the issues being considered by the committee. The nature of inquiry into charges of research misconduct deserves careful thought. The issue is not, "Is the science correct?" at this point. Error is not misconduct; conversely, assertions that are true but made on the basis of fabricated data do constitute research misconduct. The inquiry committee is not charged with initiating repetition of the research in question, but rather with determining whether a factual basis existed at the time of submittal for the claims made in a publication. The product of the inquiry process is a written report from the committee in response to their charge, accompanied by the decision of the department head to recommend to the provost whether a formal investigation of a charge of research misconduct is warranted, using the standards prescribed bylaw and by MIT policy. The department head may decide that although there is no evidence of research misconduct, other violations of Institute policies may have occurred and may recommend to the provost that an internal investigation be initiated to deal with allegations and possible sanctions by internal procedures. If the nature of the dispute or the possible violations of Institute policies are such that mediation is an option for resolution of the dispute, then the inquiry report can serve as the basis for a mediated settlement at the request of all of the parties. In any case, the provost must be notified about the outcome and receive a copy of the report. VIII.2 Research Misconduct Investigations Federal regulations require investigation into all allegations of scientific misconduct (in research funded by NSF and NIH) that have substance. MIT uses investigative panels for a variety of purposes. It can be the final fact-finding or appeal panel in a grievance; it can be a committee set up to consider a recommendation to remove tenure; it can be a committee investigating a charge of research misconduct. MIT

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II has policies and procedures in place to deal with these issues. The charge to our committee is to review these policies and procedures as they apply to allegations of research misconduct and to review the regulatory requirements when the research in question involves the expenditure of government funds, most specifically funds provided by either NSF or NIH. In these latter cases, federal regulations govern aspects of the procedures that MIT must follow, and impose on individuals found guilty of research misconduct downstream consequences that can include criminal prosecution. Implications of these consequences necessarily affect MIT's handling of such investigations. Based upon our review of cases and procedures from other universities, we endorse MIT's current policy that the responsibility for the conduct of a formal investigation into allegations of research misconduct is vested in the provost, and that normally the provost establishes a fact-finding panel whose report provides the basis upon which the provost adjudicates the charges and determines what further steps, if any, are needed. Such a formal investigation of charges of research misconduct will be initiated by the provost, typically upon recommendation of the department head, generally following an inquiry. The Institute must at this stage notify NSF or NIH if they are involved in funding the research in question; the regulations also require notification at the allegation stage under certain circumstances. We support the separation in roles for the committee as fact-finder and the provost as adjudicator. The charge to the committee should be specific as to the finding of facts and the level of certainty to be established concerning the facts that will enable the adjudicator to decide whether research misconduct has occurred. At the investigation stage, the standard of proof increases beyond "a charge having substance," which was appropriate for the inquiry stage. Because of the implications for the career and reputation of the accused, we suggest that an appropriate standard for a determination that research misconduct has occurred is a finding of fact by "clear and convincing evidence" (this lies between the criminal standard of "beyond a reasonable doubt'' and the civil standard "by a preponderance of the evidence"). The charge to the investigation committee may contain a mixture of allegations of research misconduct and other violations of Institute policies. It is important that the charge separate these issues to aid the committee in hearing testimony and in finding fact based on which the Institute must determine the truth of each allegation. Since the implications for a finding of research misconduct differ from those for

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II violation of internal policies, the committee must keep these issues distinct. The committee would be given all of the physical evidence (laboratory notebooks, manuscripts, etc.) that the inquiry committee had gathered and would also collect additional data as appropriate. The evidence that the committee would be expected to gather and evaluate is data that faculty members have competence to evaluate: research data in its various forms, publications and drafts, direct testimony from witnesses, financial records, correspondence, logs, and other laboratory records. We believe that the committee should not gather forensic evidence that requires for its evaluation expert testimony beyond that related to the science in question, such as handwriting analysis, fingerprints, and paper or ink analysis, nor use data such as surreptitious audio or video tape recordings or other data that violate Massachusetts law, and institutional policies such as the policy on privacy. We suggest two mechanisms to insulate the investigation process from the informal inquiry: first, that no individual serve on both committees; second, that the investigation committee not be given a copy of the report of the inquiry committee. They should not interview or discuss the case with members of the inquiry committee. This insulation of the investigation should ensure that the committee focuses on its charge and the evidence. Procedural error or findings from the less formal inquiry should not influence the fact-finding aspect of the investigation. Confidentiality is essential in the conducting of inquiries and investigations. This is obvious with respect to the testimony of both the accused and the complainant. It should, but may not, be obvious with respect to all participants in the process, including particularly the members of the committee. They must be formally bound by a directive and an agreement of confidentiality. They cannot break confidentiality to respond when the principals in the case criticize their activities, impugn motive to questions asked during closed meetings, or charge favoritism, or when colleagues take sides in the case. To the extent possible, members of the investigation committee should be chosen from outside the department of the person charged as well as utilizing individuals from outside from contiguous departments to provide the necessary expertise. The accused should have an opportunity to challenge the makeup of the committee but should not have a veto. The committee should have adequate staff and budget to carry out their task. They should be briefed by a designated individual in the Office of the Provost about their charge, about rules of evidence, issues of due process, and about the standard of proof and level of certainty to be used in reaching their findings.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II The latitude of the investigation is an issue. It must not be constrained to deal only with the issues originally raised in the initial allegation or outlined in the charge. If, in the course of a careful examination of the evidence directly related to the initial charge, the investigation committee comes across serious evidence of possible misconduct that was not known by the complainant or uncovered during the inquiry, then this becomes part of the investigation. The accused must be kept informed of the issues being considered by the committee. The accused will receive a copy of the charge to the committee and must be given the opportunity to respond in writing, in meetings with the committee, and by presentation of evidence. If the accused wishes the committee to call witnesses, their names and the nature of their testimony should be given to the committee in writing. The committee would attempt to interview the witnesses suggested by the accused consistent with the developing lines of investigation. We believe that the accused should be allowed to attend all of the evidentiary hearings of the investigation committee that deal with the issue of research misconduct. One reason is that the scientific chain of reasoning that leads the committee to understand the allegations and eventually to render a finding can be long and tortuous. Fairness is served if the accused is present to understand in detail the reasoning that is being used to charge and assess culpability. The accused would not respond at that point in the proceedings unless asked by the committee, nor question witnesses, but would be able to specifically respond to the charges in writing and by testimony at a later date. The committee will thus be aided in more specifically and accurately carrying out their charge. Accurate record keeping of evidentiary hearings for the purpose of fairness to the accused puts an administrative burden on the committee. If the accused is present, the burden on the committee shifts to record keeping for the purpose of reaching and justifying their findings and in some cases communicating these to the sponsoring agency. For those portions of the investigation that deal with aspects of the case other than research misconduct, the accused need not be present but should receive an accurate summary of the testimony presented. The committee is of course free to deliberate in executive session. It is the responsibility of the chair to structure the hearings to protect the rights of both the accused and the other witnesses. The role of attorneys at this stage in the process deserves some consideration. An individual is of course always free to consult attorneys at any point in life for any reason. The issue is their participation in institutional processes. Various universities deal with this issue in various ways. Some allow attorneys to be present but do

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II not allow them to speak. Others do not permit their attendance at any phase of university proceedings. By custom and tradition, MIT has not permitted attorneys to participate in Institute proceedings for either students, faculty, or employees. However, special circumstances apply when the investigation concerns research funded by the federal government. In this case, a finding of scientific misconduct may give rise to criminal charges being filed against the accused. In this case, there is an issue of "self-incrimination" during the Institute procedures. In cases involving students, MIT has decided to hold in abeyance a student discipline case when the student was also under possible indictment by a court for the same incident. In the case of scientific misconduct, we are not free to do this because of the time limits set by the agencies and our responsibilities to carry through the federally mandated process. Thus, we may be asking the accused to participate in an Institute procedure where there exists some possibility of self-incrimination. Therefore, we suggest provisions be made for the accused to bring an attorney for counsel when testifying, if desired. In this case, the role of the attorney is restricted to that of a confidential advisor to the accused. The attorney would not be present during the testimony of other witnesses, nor raise questions or objections with the committee. At any time in the proceedings, the chair may rule that the presence of the attorney is interfering with the committee procedure and may refuse permission of the accused's attorney to attend the hearing. In this case provisions should be made for the accused to have access to the attorney outside the hearing room or by telephone as the hearing progresses. We believe that members of the community have an obligation that is inherent in their positions as MIT faculty, staff, or students to participate in Institute administrative processes such as those discussed herein. If the accused refuses to participate, the committee will proceed as best they can and base their findings on the evidence presented. In the case of research funded by NIH and NSF, if the committee cannot make a finding because of the refusal of the accused to participate, MIT may have no choice but to refer the case to the agency for investigation. The outcome of the investigation is a written report containing a summary of the evidence and a finding of fact to the standard of certainty outlined in the charge. The accused receives a copy of this report and may append a response. The complainant receives a copy of those portions relevant to the complainant's allegations and may append a response. The provost receives the report plus the appended responses, adjudicates the case, and decides on an appropriate action within the framework of MIT policy and procedures. In all subsequent

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II Institute proceedings, including appeals and hearings to remove tenure, we recommend that in the absence of new and significant evidence, the facts not be refound but used as the basis for further procedures. If the sponsoring agency is NIH or NSF, the outcome of the investigation and the actions taken must be reported; the agency may take additional action. IX. Protection of Complainants MIT must ensure that individuals who raise allegations in good faith do not experience retaliation by any supervisor. We suggest that this concern be dealt with early in the process by appropriate means, such as by making alternative arrangements to have the individual's work supervised and evaluated, and by ensuring fair and objective letters of recommendation. Part of the setting up of an inquiry should include a plan to ensure the protection of the complainant. Alternatively, individuals who raise allegations maliciously may be guilty of general misconduct. We recommend that MIT ensure a supportive environment for individuals who come forward with concerns about research conduct, and that specific provisions to ensure the protection of complainants who act in good faith be a part of the plan for conducting an inquiry into allegations of research misconduct and be submitted to the Office of the Provost before the inquiry is initiated. X. Rights of the Accused Great sensitivity is required toward protecting the rights of the accused, who is after all a colleague and member of our scholarly community and who is presumed innocent of the allegations until the investigation is complete. There is a natural imbalance between the institution and the individual. In this process their interests will collide. The Institute will have legal resources and will carry through the required processes to fulfill its responsibilities. The individual will feel isolated and may lack resources to fully protect his or her rights. The rights of the accused during the proceedings described above are adequate notice of the charges, and an opportunity to respond in an impartial, fair, timely, and objective process. We have outlined procedures to provide adequate notice: receiving the charge; attendance at evidentiary hearings during the investigation; and the opportunity to

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II receive the committee reports. We have outlined procedures to provide the opportunity to respond, including having an impartial committee; an opportunity to present witnesses; and an opportunity to respond to the committee report. The accused has a right to avoid self-incrimination related to a potential criminal proceeding, and we have recommended the option of the accused, if he or she testifies, having the right to consult an attorney—but not otherwise having lawyers participate. The accused has the right to a confidential proceeding; individuals who disclose facts concerning the case to individuals without a need to know may violate MIT policy and may risk civil suits. Current MIT policy also grants the right to be accompanied to MIT proceedings by an MIT advisor. Procedures to ensure these rights differ between inquiry and investigation. Since only an investigation can result in a finding of misconduct and lead to sanctions as well as public disclosure, the procedures are necessarily more formal. Our suggestions for procedures to safeguard the rights of the accused, in this and in previous sections, are based in part on our perceptions of the unwritten covenant between faculty and administration about the values inherent in our relationship. The suggestions we have made are directed toward providing protections for faculty, students, and staff who are accused of what in scholarship is a capital crime. XI. Institutional Memory Because the process of inquiry and investigation into allegations of research misconduct is carried out with a high degree of confidentiality, there is little opportunity for the MIT community to learn about how to respond effectively to new cases as they arise. And yet, because of the importance of these issues to the Institute and its faculty, staff, and students, we must effectively deal with such cases. The thrust of our procedural recommendations is to ensure that possibly serious cases immediately come to the attention of senior officials who can ensure that proper procedures are followed. We also believe that there is a need to establish a formal mechanism to ensure institutional memory for these issues. Some of the important functions requiring such institutional memory are to provide assistance and advice to a department head concerning the selection of and the charge to a committee of inquiry; to foster consistency of procedures and standards across departments; to brief committees of inquiry and investigation as to their charge, evaluation of evidence, standards of proof, and fair process requirements; to ensure

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II that plans are made to protect complainants who act in good faith; and to make available knowledgeable advisors in the event that the inquiry or investigation takes an unexpected turn. Some universities have established a standing faculty committee to provide these functions and to ensure that past experiences are used to guide future actions. We believe that this function can be more effectively provided by centralizing the activity within the Office of the Provost. The provost will be the adjudicator after the investigation (if any) is completed, and should not be involved at this stage in the process of developing evidence. Rather, the provost should identify individuals within MIT who can ensure that proper processes are initiated in response to allegations and who can advise committees on procedural issues and charge such individuals with carrying out these functions. Therefore, we have recommended above that MIT establish a function within the Office of the Provost to guide the processes of responding to allegations of research misconduct. The earlier section of our report on responding to allegations can be interpreted as setting up procedures for these processes. However, such procedures must be continually updated to respond to changing regulations and legislation. Part of the responsibilities of this individual would be the development of procedures for inquiries and investigations and the continual review and update of these, both to respond to changes in federal regulation and to improve their effectiveness. XII. Interactions with the Federal Government For research supported by NIH and NSF (currently), the end of the Institute's investigation begins the response of these federal agencies. MIT is required to furnish to these sponsoring agencies the evidence, the findings and the conclusions of its investigation, and the actions taken in sufficient detail to permit a thorough evaluation of the outcome and basis for the Institute's findings and to allow the agency to repeat the investigation if it wishes. At this point, actions of the accused, the Institute, and the individuals who participated in the process, as complainants, as members of faculty committees, or as Institute officials, may be subject to further scrutiny. The accused may be censured or debarred from future federal funding. The Institute may be criticized for its handling of a case. Individuals involved may be accused of conflict of interest, of making false accusations, or of negligence for their roles in carrying out an inquiry or investigation. It is thus imperative that the Institute give full attention to these matters.

OCR for page 159
Responsible Science: Ensuring the Integrity of the Research Process, Volume II The scientific community has expressed its concern about the vagueness and the inconsistencies between agencies in the definition of scientific misconduct as well as concerns about failures of due process and confidentiality on the part of federal agencies. A dialogue is ongoing that, it is hoped, will resolve some of these issues, thus enabling universities and the federal agencies to fulfill their responsibilities while protecting the rights and reputations of the individuals involved and ensuring the productivity and creativity of the scientific enterprise. We endorse MIT's efforts to join with other universities, professional societies, and individual members of the scientific community in working cooperatively with federal agencies to improve procedures for the federal response to allegations of scientific misconduct. The past few years have seen considerable turmoil surrounding the issue of institutional response to allegations of scientific misconduct. During these past few years, universities have put in place federally mandated procedures to deal with such allegations that occur on their campuses. The National Academy of Sciences … established a Committee on Scientific Responsibility and the Conduct of Research … [that reported in April 1992]. We urge a period of stability with respect to new federal regulations to give universities and the scientific community an opportunity to gain experience with these procedures.