STUDENTS AS RESEARCH SUBJECTS

Why does it matter if research subjects are students? Three answers surface immediately: first, students might be children; second, students might be in school; and third, students might be engaged in learning. None of these answers is always true, but they are true often enough to deserve consideration in research plans involving students as research subjects. Research involving students, be they minors or adults, should be conducted in accord with ethical principles and applicable regulations to protect students from potential coercion and harm. Paradoxically, these regulations and ethical codes send conflicting signals: The regulations carve out certain kinds of education research as being exempt from the umbrella of regulatory protection, while various ethical statements, disciplinary codes, and guidance—most notably from the American Educational Research Association—single out students as deserving careful treatment (Office for Human Research Protections; Strike, Anderson, Curren, et al.; American Psychological Association; American Sociological Association).

Primary and Secondary Education Students

The vast majority of students in primary and secondary schools have yet to reach adult maturity, legally and developmentally. Given the ethical principle of respect for persons, from which arises the practice of informed consent, research involving young students raises the question of students' abilities to make voluntary, competent and informed decisions about whether or not to participate in proposed research. Human development and experience affect the level of student understanding of research participation. Language and literacy skills affect students' ability to receive and interpret relevant information, much less appreciate the implications of what the consequences of participation could be. Children are generally less familiar than adults with key concepts relevant to research participation and risk, such as confidentiality, experimental trials, or the estimated probability of a particular outcome. Such cognitive tools are essential to grasping a specific research activity and the involvement of research subjects. As they develop and learn, students gradually become more like adults in their capacity to truly understand what involvement in research entails (Bruzzese and Fisher). Some education systems set standards regarding what students should know and be able to do in science at various grade levels, as illustrated by the American Association for the Advancement of Science and the National Research Council. Such standards may provide useful guidance regarding what prospective young research subjects should be expected to understand about participation in research.

Young students also vary significantly from adults in their perceptions and assessments of risks and benefits. Their abilities to make practical judgments are less well developed than those of adults. They may attach very different values to specific harms and benefits, and most concern themselves with short-term consequences. A typical sixth grader views the sacrifice involved in giving up math class to participate in research differently from how an adult would understand it. Children and adolescents also have their own views about common forms of research compensation such as money or material goods. Young people's faculties of moral judgment, including the reasons they use to justify practical decisions, also vary from adults in patterned ways (Bebeau, Rest, and Narvaez).

Students' voluntary decision-making is also shaped by various influences. Very young children are strongly affected by their parents and other significant adults, while adolescents become more susceptible to their friends' and peers' value orientations and pressures as adult influences wane (Steinberg, Brown, and Dornbusch) These influences obviously bear upon how researchers should construct the circumstances in which students are asked to participate in research.

Parents, Guardians, and School Officials

These developmental considerations lead ethicists and some federal agency regulations to view the agreeable young person as providing assent, that is, an affirmative expression of willingness to participate voluntarily in research. U. S. federal regulations generally require assent from minors and supplement that requirement with the permission of a parent or guardian. Requirements are less rigidly established in other countries, and practices vary widely. Permission is construed according to standard criteria for informed consent with respect to the parent or guardian's decision on behalf of the student.

Permission generally provides an appropriate mechanism for protecting the autonomy, interests and welfare of the young student, but it may also present challenges. If permission comes from the parents, the research team needs twice the number of affirmative responses to its request for participation. Parents often are not as easy to contact as students, and research suggests that some parents do not give permission for their children to participate in research not because they object, but simply because they don't get around to signing and returning a consent form (Singer). Other parents have reservations about a research team's overtures that are unrelated to any concerns about their children's welfare; they may be embarrassed to reveal their lack of literacy skills, or believe that signing a form reflects a legal concession, or even fear that their signature somehow puts them in jeopardy. Some parents have interests contrary to the students whose welfare they are expected to protect; they may worry that their child's responses to a research survey may embarrass or incriminate the parents. On the other side, students' hesitation about participation may stem from apprehensions about what their parents may find out about sensitive survey questions or how they answered them.

Other adults have a role in protecting prospective student research subjects, which may lead to tension over who has authority to permit student participation in research. School officials are responsible for students' welfare and directly supervise students' school day activities. Since proposed research activities may disrupt normal school life, researchers are generally obliged to obtain school officials' permission to carry out research with students as research subjects. Researchers might also view school officials as appropriate sources for permission to involve students in research. This view is contestable, however, as conflicts arise about parents having control over, or at least a say in, what happens to their children in schools. In the United States, federal laws—most notably the Family Educational Rights and Privacy Act and the Protection of Pupil Rights Amendment—and state and local laws and policies reflect efforts to prescribe when parents must be consulted before researchers may collect data about students through school records, surveys, or other means. Since federal regulations do allow for waiver or alteration of elements of consent or documentation of consent under certain conditions, researchers sometimes request such waivers, particularly for large scale surveys where the researchers point to the low level of risk involved and the difficulties of securing adequate unbiased samples if active parental permission is required. For some parents, however, it is the sensitivity of survey topics and the potential invasion of privacy that concerns them, not the degree of risk involved.

The School Site

The question of who has authority over what happens to students in schools also raises the issue of the convenience of conducting research with students in schools. Most primary and secondary schooling is compulsory in affluent countries; consequently, researchers can expect to find in schools large and fairly representative samples of young people, neatly segregated by age, under adult supervision, engaged in activities not of their own choosing, and legally required to stay. The opportunities for relatively inexpensive and efficient data collection are obvious. Thus some research efforts seek to involve populations of students as research subjects not because the research objective is focused on understanding education or the lives of students per se, but rather because students in schools offer a convenient way to study a wide variety of phenomena concerning youth. This approach may be viewed by some as a form of exploitation that could be unwelcome and disruptive to the educational process. Frequently students are surveyed about their health and development, extracurricular activities, and perceptions of themselves and society, for reasons unrelated to their educations. Research studies focused on areas outside of education may detract from the school's pursuit of its educational mission, and may pose risks for students by asking for sensitive information about criminal, antisocial, or private behavior.

Education Research and Practice

Education research involves students for the specific purpose of studying the formal and informal processes of learning. Such research projects raise their own set of important ethical concerns, particularly when they involve educational practice and practitioners. Some of these concerns resemble those raised in clinical trials of therapeutic medical interventions.

Education research may involve the educational equivalent of the therapeutic misconception: the mistaken belief that the nature of the subject's involvement in research in designed to improve that subject's welfare, rather than in developing generalizable knowledge (Appelbaum, Roth, Lidz, et al.). Students, parents, and researchers may all fall prey to the educational misconception, especially in the context of educational practice.

Education researchers often involve teachers or other practitioners in data collection or as co-researchers, and the dual roles of researcher and practitioner sometimes conflict (Hammack), as they do in biomedical research (Koski). What should a practitioner/investigator do in a classroom situation where pursuing a research question comes at the expense of delivering an important lesson? What if one uncovers sensitive information about the students that an educator would not otherwise have known? If a practitioner/investigator discovers that a student cheated on a test, should he or she, as an educator, discipline the student and alter the grade, or, as a researcher, protect the research subject from harm? Note that in such circumstances the protective research device of confidentiality is rendered useless, because the person who collects the information is the selfsame person as the one from whom the potentially harmful information is supposed to be kept.

Education research also raises issues of justice or fairness with regard to the selection of research objectives and selected student populations. Should the focus be on research that will benefit the largest possible number of students, with current level of success in the middle range? Or should the focus be on those students who possess the potential to improve the most from better educational interventions, even if they are already doing relatively well? And what of those students who are currently doing relatively poorly in the current educational system, whose level of achievement may be the most difficult to improve? Do students of one ethnic or linguistic minority deserve more attention than students of another, because their numbers in the education system are larger? Such questions of beneficence and justice are reflected in any research project proposing a selective sample of students, and are framed by deeply held cultural beliefs about the importance of education as a vehicle for equality of opportunity and social mobility.

Education research frequently involves the evaluation of interventions delivered at a collective level in classrooms and schools. Reducing class size, changing teacher behavior, altering curricula, attaching high stakes to test performance, and reforming school culture are all educational interventions that can only be accomplished and studied at a group level. In a study where classrooms or schools are randomly assigned to an innovative approach (treatment) or to the standard educational practice (control), by the time the results are available, students may have outgrown the opportunity to benefit from the more effective intervention if they did not receive it during the study. Individual students and their guardians may be able to decline to have the data about them collected or included in research analyses, and sometimes accommodations such as classroom re-assignment can be made to enable students to opt out of a research study. In many cases, though, if a student or guardian wants to avoid the student's participation in a research evaluation of an educational practice at a participating school, the options may be severely constrained or costly. Such collective decisions about the involvement of students as research subjects must face the challenge of striking a reasonable balance between majority will and minority freedoms (Oakes)

The use of qualitative methods and the involvement of practitioners in many education research studies significantly transform the relations and ethical orientations among researchers, practitioners, and students. Qualitative research strategies in education characteristically intertwine the prescriptive and descriptive dimensions, place research activities in specific moral and political frameworks, and recognize the essential contributions of the "insider" subjects' perspectives, abandoning disinterested stances in favor of "advocacy" positions (Howe and Moses). Likewise, practitioner/investigators often find themselves sharing control over the nature, objectives, and credit for research projects with their colleagues and students, making the relationships among the participants more egalitarian, changeable, and complex (Zeni). In such cases the ethical responsibilities shift accordingly.

Tertiary Education Students

Undergraduate and graduate students generally have the background knowledge, literacy skills, and abilities to appreciate potential harms and benefits at a level resembling those of adults. Indeed, nearly all are adults, removing the need for parental permission for the vast majority of tertiary education students. The typical college student also has relatively good health, more flexible schedule commitments, few or no dependents, limited financial resources, considerable disposable time, and openness to new experiences. More than fifteen million students attend degree-granting tertiary education institutions in the United States, where they are easily accessible to academic researchers. Tertiary education students are prime candidates for research involvement.

DEPARTMENTAL SUBJECT POOLS. Academic researchers, most notably psychologists, have capitalized on the ready availability of students, often their own students. The vast majority of findings in human studies by psychologists come from research involving students as research subjects (Chastain and Landrum). In order to avoid the coercive situation of faculty asking their students to participate in their own research, many institutions set up departmental subject pools (DSPs) through which they arrange for students to participate in faculty research projects.

The practical arrangements of DSPs vary. Some are entirely voluntary, while others are attached to course selection—usually introductory psychology or other lower-division social science courses—and are either required of students or award extra credit for the course. Most DSPs allow students other options, such as writing a paper, as an alternative to participation in research. The ethical rationale generally put forward for this practice is that the DSP arrangements for research participation provide an educational benefit to the research subjects while efficiently supplying sufficient numbers of research subjects to enable faculty to carry out a more robust research agenda of valuable studies.

Some DSPs are ethically better than others, depending on specific features of the DSP, including the following:

• Clear, timely information is provided to prospective students.
• Investigators demonstrate respect for research subjects through their conduct.
• Student participation in the system is efficient and non-punitive.
• Students choose from a variety of research studies.
• Research studies are appropriate for a sample population of college students.
• Research studies are generally low in risk.
• Subject participation is structured to provide educational consent and debriefing experiences.
• A variety of alternatives to participation in research are offered, involving comparable educational value, time commitment, and enjoyment (Sieber).

The 2002 revision of the American Psychological Association's ethical code reaffirms its previous position on students as research subjects, allowing DSPs if students have alternative options: "8.04 Client/Patient, Student, and Subordinate Research Participants (b) when research participation is a course requirement or an opportunity for extra credit, the prospective participant is given the choice of equitable alternative activities" (American Psychological Association).

There are reasons for skepticism about whether DSPs are ethical at all, unless they are entirely voluntary. Given the imbalances in power, authority, and autonomy inherent in the relationship between teacher and student, complete voluntariness may not be possible. Whatever degree of latitude DSPs permit in student choice of research projects or alternatives, required participation still reflects researchers' use of their control over students' educational choices to induce them to participate.

The claim that DSP participation represents a genuinely beneficial and authentic educational experience is open to objection. DSPs present students with consent situations that do not reflect the ideal of voluntary participation: Outside of DSPs, research subjects are either volunteers or are offered compensation in forms they value. Extraordinary briefing or debriefing experiences may provide students with better understanding of the substance of a particular research project, but to the extent that they succeed in this regard they also provide a distorted picture of the educational benefits of the typical non-DSP research subject's experience. It is also difficult to accept the idea that well-designed debriefing exercises will overcome the inherent differences in the educational potential of the research subject's actual experience of participation in a given research study. If the goal throughout is really supposed to be educational, it seems a more effective approach would be for faculty to involve their students in mock research activities specifically designed to demonstrate key features of research subject participation connected to the rest of the course's curricular content. In sum, the rationale for DSPs may be said to represent the institutionalization of the educational misconception. What is especially intriguing in this regard is the fact that the educational value of research subject participation in DSPs is seldom even assessed, and few if any rigorous research studies have been conducted comparing the educational effectiveness of research participation to the frequently utilized alternative educational options. Some educational institutions have taken steps to eliminate DSPs or impose additional oversight procedures to ensure that when and if they are permitted, they are closely scrutinized. For some, recruitment of students through broad-based appeals to the general public is considered preferable both ethically and scientifically, as a less selected population of subjects may enhance the generalizability of study results.

STUDENTS IN GRADUATE OR PROFESSIONAL TRAINING. Outside of DSPs, students sometimes participate in research studies related to their field of study, some of which focus on their regular education and training. These studies present challenges resembling those discussed above with regard to primary and secondary school practitioner researchers, with respect to the general issues of possible coercion, the educational misconception, and the inherent limits of confidentiality protections. What is different, however, is that the researcher/subject relationship has grown in some ways: The student research subject is now an adult, and the researcher is now someone whose authority over the student has taken on a different shade. For example, graduate or medical students may view the investigator as an important mentor and career influence, in addition to whatever control the investigator might possess over students' grades, recommendations, teaching or research assignments, and opportunities for postdoctoral work or residency programs. Even if the researcher wishes the students to view a decision to participate as a research subject as entirely voluntary, the students may see themselves as having no choice.

At the same time, academic faculty may view participation in research as an obligation students should accept as a function of their having chosen to pursue a profession in which research plays an integral part. Where faculty are doing research to evaluate the effectiveness of their educational practices, they may feel that students have an obligation to contribute to improving those practices, because the students are benefiting from lessons drawn from studying previous students' experiences (Dubois). At the same time, faculty may be conscious of the importance of providing role models of researchers who treat their subjects with the utmost respect, and must therefore solicit their voluntary consent (Henry and Wright). Hans Jonas argued that those persons who are most knowledgeable, committed, and autonomous should be the first to participate as research subjects, which presumably implies that graduate and medical students should be the first to volunteer for such studies, after the faculty themselves. This argument construes the idea of autonomy more broadly, in the sense that while the student may feel pressured to volunteer for a study at the given moment, the student has chosen to pursue a highlyrewarded profession in which research—with its incumbent risks and sacrifices for human subjects such as themselves—plays an important role. To the extent that a student's autonomy is limited by the personal and professional circumstances of their participation, Jonas's presumption may not be true.

Conclusion

Students' decisions to participate in research may be affected by various influences, incentives, rewards, or compensation, and yet the pressure of these factors does not always rise to the level of undue influence or coercion. Students occupy a wide range of locations along the spectrum of opportunities for participation in research, ranging from invitation to attraction to enticement to pressure to force. Some influences may be altered, while others are endemic to the student's natural condition. As long as investigators and institutions are cognizant of and responsive in the design and execution of their studies to the special situations that arise in research involving students, they can reduce the likelihood that additional social and regulatory limits to their work will be imposed. Unless society is willing to forego all research in which students are the research subjects, the challenges of enlisting students as research subjects under circumstances of mixed voluntariness will continue.

ivor anton pritchard

greg koski

SEE ALSO: Informed Consent, Consent Issues In Human Research; Research Policy, Subject Selection; Military Personnel as Research Subjects; Minorities as Research Subjects; Prisoners as Research Subjects; Research, Unethical

BIBLIOGRAPHY

American Association for the Advancement of Science. 1993. Benchmarks for Science Literacy. New York: Oxford University Press.

Appelbaum, Paul S.; Roth, Loren H.; Lidz, Charles W.; et al. 1987. "False Hopes and Best Data: Consent to Research and the Therapeutic Misconception." The Hastings Center Report 17(2): 20–24.

Bebeau, Muriel J.; Rest, James R.; and Narvaez, Darcia. 1999. "Beyond the Promise: A Perspective on Research in Moral Education." Educational Researcher 28(4): 18–26.

Bruzzese, Jean-Marie, and Fisher, Celia B. 2003. "Assessing and Enhancing the Research Consent Capacity of Children and Youth." Applied Developmental Science 7(1): 13–26.

Chastain, Garvin, and Landrum, R. Eric, eds. 1999. Protecting Human Subjects: Departmental Subject Pools and Institutional Review Boards. Washington, D.C.: American Psychological Association.

DuBois, James. 2002. "When is Informed Consent Appropriate in Educational Research? Regulatory and Ethical Issues." IRB: Ethics & Human Research 24(1): 1–8.

Hammack, Floyd M. 1997. "Ethical Issues in Teacher Research." Teachers College Record 99(2): 247–265.

Henry, Rebecca C., and Wright, David E. 2001. "When Do Medical Students Become Human Subjects of Research? The Case of Program Evaluation." Academic Medicine 76(9): 871–875.

Howe, Kenneth R., and Moses, Michele S. 1999. "Ethics in Educational Research." In Review of Research in Education, ed. Ashghar Iran-Nejad and P. David Pearson. Washington, D.C.: American Educational Research Association.

Jonas, Hans. 1969. "Philosophical Reflections on Experimenting with Human Subjects." In Experimentation with Human Subjects, ed. Paul Freund. New York: George Braziller.

Koski, G. 1999. "Resolving Beecher's Paradox: Getting Beyond IRB Reform." Accountability in Research 7: 213–225.

Oakes, J. Michael. 2002. "Risks and Wrongs in Social Science Research: An Evaluator's Guide to the IRB." Evaluation Review 26(5): 443–479.

Sieber, Joan E. 1999. "What Makes a Subject Pool (Un)Ethical?" In Protecting Human Subjects: Departmental Subject Pools and Institutional Review Boards, ed. Garvin Chastain and R. Eric Landrum. Washington, D.C.: American Psychological Association.

Singer, Eleanor. 1993. "Informed Consent and Survey Response: A Summary of the Empirical Literature." Journal of Official Statistics 9(2): 361–375.

Steinberg, Laurence; Brown, B. Bradford; and Dornbusch, Sanford M. 1996. Beyond the Classroom: Why School Reform Has Failed and What Parents Need to Do. New York: Touchstone.

Strike, Kenneth A.; Anderson, Melissa S.; Curren, Randall; et al. 2002. Ethical Standards of the American Educational Research Association: Cases and Commentary. Washington, D.C.: American Educational Research Association.

Zeni, Jane, ed. 2001. Ethical Issues in Practitioner Research. New York: Teachers College Press.

INTERNET RESOURCES

American Psychological Association. 2002. "Ethical Principles of Psychologists and Code of Conduct." Available from <http://www.apa.org/ethics/>.

American Sociological Association. 1997. "American Sociological Association Code of Ethics." Available from <http://www.asanet.org/members/ecoderev.html>.

Family Policy Compliance Office, Department of Education. Family Educational Rights and Privacy Act. Available from <http://www.ed.gov/offices/OII/fpco/ferpa/leg_history.html>.

Family Policy Compliance Office, Department of Education. Protection of Pupil Rights Amendment. Available from <http://www.ed.gov/offices/OII/fpco/hot_topics/ht_10–28-02.html>.

National Committee on Science Education Standards and Assessment, National Research Council. 1996. National ScienceEducation Standards. Available from <http://www.nap.edu/books/0309053269/html/index.html>.

Office for Human Research Protections, Department of Health and Human Services. Federal Policy for the Protection of Human Subjects. Available from <http://ohrp.osophs.dhhs.gov/humansubjects/guidance/45cfr46.htm>.