Download ETHICAL ISSUES IN CONDUCTING RESEARCH and more Summaries Ethics in PDF only on Docsity!
EETTHHIICCAALL IISSSSUUEESS IINN
CCOONNDDUUCCTTIINNGG RREESSEEAARRCCHH
55
C HAPTER O BJECTIVES
Ensuring the Protection of Human Participants Consent Capacity Information Voluntariness
R ESEARCH IN A CTION Harm Potential Hazards to Participants Determining the Degree of Harm Privacy Issues in Research Affecting Privacy Deception Reasons for Deception
S IMULATION 3.
Ethical Considerations in Quantitative and Qualitative Research Ethics and Quantitative Research Ethics and Qualitative Research
Participant Observation Privacy Issues Degree of Deception
Monitoring Ethics in Research: Institutional Review Boards Federal Requirements for IRBs
Ethics and Professionalism Integrity During Execution of the Study Integrity Related to Publishing Plagiarism and Maintaining Integrity Simultaneous Submission Ethical Issues in Authorship Sanctions for Breaches of Integrity
C HAPTER S UMMARY
K EY T ERMS
S TUDENT S TUDY S ITE
S IMULATION F EEDBACK 3.
C H A P T E R O U T L I N E
Describe the three elements of informed consent for ensuring the protection of human subjects.
Discuss who must consent to participation in a research study and how it should be given.
Describe “harm” as a potential hazard to participants, and identify strategies for determining its level of acceptability within a research study.
Identify the issues in research that directly affect the “privacy” of an individual.
Define “deception” and the rationale for why it may be used in a research study.
Distinguish between ethical considerations in quantitative and qualitative research.
Understand the role of institutional review boards in monitoring ethics in research.
Understand the importance of maintaining integrity in the execution of a research study and the dissemination of results.
56 PART I: O VERVIEW OF R ESEARCH
C H A P T E R O B J E C T I V E S
E
thics has become a cornerstone for conducting effective and meaningful research. As such, the ethical behavior of individual researchers is under unprecedented scrutiny (Best & Kahn, 2006; Field & Behrman, 2004; Trimble & Fisher, 2006). In today’s society, any concerns regarding ethical practices will neg- atively influence attitudes about science, and the abuses committed by a few are often the ones that receive widespread publicity (Mauthner, Birch, Jessop, & Miller, 2003). This chapter focuses primarily on issues, activities, and situations that are far from black and white in terms of ethics. A balanced presentation remains our goal. There are no simple answers or solutions. Although guidelines are suggested, each researcher must be responsible for ethical issues within an investigation. This chapter will not provide a “complete ethical checklist” that covers all questions, settings, or concerns. Such a task would be impossible and not terribly helpful in considering the variety of research situations that may be encountered. Although some guidelines are relatively firm, the best insurance against unethical research practices is the knowledgeable indi- vidual scientist who can intelligently consider the circumstances being faced. Researchers with their inquiring minds are compelled to press on in the develop- ment and dissemination of new knowledge. The question becomes not only a matter of whether it is ethical to conduct research, but also whether it is unethical not to engage in inquiry. Are researchers guilty of unethical conduct if they do not engage in systematic inquiry conducted with the safety of participants in mind, and with fully informed consent? While ethical considerations may initially be viewed as road- blocks to beginning a study, they are clearly integral to the process. Attention to the ethics of an investigation requires extra thought and effort, but the payoff for a study that is both methodologically intact and ethically sound is extremely exhilarating.
ENSURING THE PROTECTION OF
HUMAN PARTICIPANTS
Every researcher has a responsibility to protect the participants in an investigation. The Ethical Principles of Psychologists and Code of Conduct of the American
After reading Chapter 3, you should be able to do the following:
The second factor in determining competence is the individual’s ability to give consent. (Does the individual understand the nature and consequences of giving consent?) This factor is again relevant to the involvement of children in research studies but is also applicable to others as well (e.g., persons with cognitive or emo- tional disabilities, or those who are incarcerated) (Hardman et al., 2006). First, from a legal standpoint, a person is considered competent unless they have been adjudi- cated through the court system as being incompetent. It cannot be assumed that because individuals of adult age may have a cognitive or emotional disability they are not competent to make decisions regarding their individual welfare. Second, even if the individual is deemed legally competent, it does not relieve the researcher of the responsibility to ensure that the participant understands the information provided and can exercise the free power of choice. Clearly the capacity element is not simple. It often becomes even more complex in the context of imple- mentation, which is discussed in later sections of this chapter.
I n f o r m a t i o n Determination of whether information has been communicated to a participant in an effective manner is based on both substance and manner. What information was given, and how was it presented? The information must be planned and pre- sented so it can be completely understood, and it must be fully understood by the participant. It is the researcher’s responsibility to see that this is accomplished. This perspective places a great responsibility on an investigator and makes assurance that effective consent has been obtained even more complicated.
Vo l u n t a r i n e s s Voluntary consent is concerned with each individual’s ability to exercise the free power of choice without the intervention of force, fraud, deceit, duress, or other forms of constraint or coercion. This right to exercise choice must be present throughout the entire research process. The intent of this interpretation is that no such “constraint or coercion” must be either explicit or implicit on the part of the investigator. The collective consideration of all three elements of consent has great impact on the manner in which a study is planned and executed. Each research situation pre- sents a different set of circumstances, and consent procedures must be adapted accordingly. For investigators in certain areas (e.g., child development, cognitive disabilities, mental health), the type of participants frequently studied requires spe- cial consideration and protection (see, for example, Hardman et al., 2006; Roberts et al., 2005). Such participants will receive repeated attention in our discussion and examination of consent issues regarding who and when.
Who may give consent? The first answer to this question is the parent, guardian, or other agent legally responsible for the person and authorized to act on the person’s behalf. Although this may seem simple, it is not, and the prerogative of consent is not constant. It varies greatly from situation to situation and among participants. The question of who may give consent is not a difficult one for those who have not been adjudicated as being “incompetent.” Adults can give their own consent to par- ticipate in research. It is also clear that they must consent on a voluntary basis and with complete information concerning the nature and consequences of participa- tion. Many individuals deemed incompetent and in need of special protection are thought to be unable to give consent on their own. They may be lacking in capacity because of cognitive or emotional disabilities, or because they are legally too young.
58 PART I: O VERVIEW OF R ESEARCH
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 59
They may appear very vulnerable to either explicit or implicit coercion (thus violat- ing the truly voluntary element of consent). They may have a combination of these and other characteristics that render them unable to exercise free will and make decisions. For such individuals, the question then becomes one of who can give con- sent on their behalf (and what should be considered in the process). The parent, guardian, or other agent (e.g., institutional administrator) may give consent unless it comes to someone’s attention that the interests of the person who is consenting are at odds with what may be the interests of the subject. For situations in which participants cannot provide their own consent, the researcher should not assume that “parental” consent is sufficient (Drew & Hardman, 2007). There is a need to constantly be vigilant in these situations. The American Psychological Association (2002) suggests that obtaining legal consent from the parent or guardian does not relieve the researcher of respecting the wishes of the child. The APA suggests,
For persons who are legally incapable of giving informed consent, psycholo- gists nevertheless (1) provide an appropriate explanation, (2) seek the individual’s assent, (3) consider such persons’ preferences and best interests, and (4) obtain appropriate permission from a legally authorized person, if such substitute consent is permitted or required by law. (p. 7)
Thus, consent from both the guardian and the child is needed. It is important to note that the conventional wisdom associated with consent may not be appropriate for children. Consent, as a legal term, is not considered appropriate to use with minors. As noted in the APA code above, a child’s agreement or refusal to participate is more appropriately described as assent or dissent. Researchers have investigated the ability of children to give consent because of ongoing concerns that these young participants are at risk for some level of implicit coercion (see, for example, Roberts et al., 2005). Findings do confirm that children have some capacity to make decisions about their participation, but there appear to be age-related problems regarding the degree to which they can do so freely. Some evidence suggests that children at about the seventh-grade level struggle to under- stand their prerogatives compared to adults (Diekema, 2005). The notion of not agreeing with their adult consent-giver (parent or guardian) is not only difficult to understand for children, it is inconsistent with much of the other information they receive daily. This is even more challenging for younger children and for those with cognitive disabilities (Field & Behrman, 2004; Gelfand & Drew, 2003). Investigators must use their best professional judgment in determining who should be involved in the consent process (see Box 3.1).
When must consent be given? Some research circumstances clearly require consent, whereas others are more subject to debate. Risk to participants may include a number of different situations, ranging from actual physical harm in the form of pain to potential psychological harm such as emotional stress. Similarly, the forfei- ture of personal rights may vary greatly, from a serious invasion of privacy to a sit- uation in which the participants are not threatened in any significant fashion. The factors involved when consent should be obtained are similar to those regarding who can provide effective consent. Researchers must exercise their best professional judgment. However, the rule of thumb is when in doubt, ask for permis- sion. If it is unclear whether or not consent is needed, it is prudent to err on the side of caution. Smith (2003) noted that the APA allows for no consent to be obtained
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 61
KEY POINTS IN THE CHAPTER REFLECTED IN THIS BOX:
- Anticipating the various ethical issues in conducting action research.
- Determining what approvals will be needed from institutional organizations.
- Checking guidelines regarding participant consent and assent.
- Anticipating questions about potential hazards, confidentiality, and other risk fac- tors presented in action research studies (all of which will be asked by approval groups such as IRBs and school research committees).
OBJECTIVES TO LEARN FROM THIS BOX :
- Understand how action research conducted in a field setting present challenging contexts pertaining to anticipating ethical issues.
- Reflect on how action researchers anticipate and resolve potential risk issues.
- Determine how consent and assent are obtained for these action research studies.
SCENARIO 1: STUDENT ENGAGEMENT IN READING
Emily is beginning to study the effects of engagement on how young students learn to read. She is using what is labeled as concept-oriented reading instruction (CORI) as a means of engaging the experimental group, and her comparison group is going to receive drill and practice (DAP) on words and sentences. Emily reviews the notes on action research found in the AERA Guiding Standards and is a little confused by the wording. Both reading instruction approaches are found in different teacher’s teaching styles and she is not sure whether she can consider both of her instructional methods to be normal teaching processes. Surely Emily intends to maintain confidentiality, protect her students’ welfare, and the information will be used to benefit students’ instruction in general. But do the guidelines mean that she doesn’t need consent because this can be considered a normal part of the teaching process? Emily will need to have her research approved by the Institutional Review Board (IRB) for her university. The best guideline for her to follow is found in the wording “when in doubt, obtain consent.” In fact since she is studying young students, Emily will need to obtain consent from her students’ parents and also have her students complete an assent form that indicates they agree to participate. Before she can begin the study she will need to outline the details of her protocol and have it approved by the IRB. She will need to also obtain permission to conduct her study in the school district where the students are enrolled and then seek consent from parents and assent from the students themselves. It is unlikely that there is much if any potential hazard to the students. Emily will need to maintain con- fidentiality of her students’ performance (her data). Although confidentiality is an important matter, the data do not raise huge red flags regarding sensitive information. Emily’s study will not likely encounter sig- nificant ethical issues.
SOURCE: This action research scenario is roughly based on Swan, E. A. (2003). Concept-oriented reading instruction: Engaging classrooms, lifelong learners. New York: Guilford Press.
R E S E A R C H I N A C T I O N
For more research in action tips visit www.sagepub .com/drewstudy
How should consent be obtained? The method of obtaining consent will vary greatly depending on the situation involved and the degree of potential risk for par- ticipants. Consent may be obtained less formally (i.e., verbally) if the study creates little or no risk or potential invasion of privacy. In such circumstances, participants may be verbally informed of the nature of the investigation (assuming that the capacity element is present) and give consent verbally. Consent should be obtained in writing in other situations when participants are placed in a more “risky” posi- tion (e.g., potential harm, stress, substantial invasion of privacy). This is typically accomplished by providing both written and verbal explanations of the study, and the subject indicates consent by signing the written form. In certain cases, the pro- vision of both verbal and written explanations is not possible, such as in a survey using mailed questionnaires. Usually the subject’s willingness to complete a ques- tionnaire is adequate indication of consent. A written explanation of the survey should be included with the questionnaire (often as a part of the instructions on the questionnaire) with a clear indication that responses to the questions are voluntary. In all cases, the researcher must remain cognizant of the three elements of consent and also inform the subject that consent can be withdrawn at any time. It is also routine and required that research conducted through colleges or universities involve written consent. Figure 3.1 is a sample assent form for children (parental agreement would also be needed).
62 PART I: O VERVIEW OF R ESEARCH
SCENARIO 2: STUDENT SELF -ESTEEM
Daniel is beginning a study to see if his students with learning disabilities have different self-esteem depending on whether they are of Latino or Caucasian descent. He believes that his students have rather low self-esteem and it seems to him that his Latino students’ self-esteem is somehow different than that of his Caucasian students. Both groups of students are normally in Daniel’s classes although they are not typically asked questions about their self-esteem. Daniel will need to obtain approval from the IRB at the university because he will be conducting this study as part of his degree program. He will also need to obtain permission from the school district as well as consent from the parents and have the students complete an assent form. Consent and assent forms will be part of the protocol to be approved by the IRB and the school district as will any questionnaire that Daniel intends to use with his students. Because of the nature of his study, Daniel will have to pay particularly close attention to two of the stan- dards in the AERA Standards. These include being attentive to cultural, religious or other significant matters in all aspects of his research since he is comparing Latino and Caucasian students. The second will be focus- ing on and minimizing the any potential negative social consequences of asking his participants about self- esteem matters. Daniel’s study does raise some attention level because of his comparison groups but any issues can be minimized by careful protocol review and adequate sensitivity.
TThhiinnkk aabboouutt tthhiiss:: Are you using children and what is the definition of children? What participant consent actions must you take for using children as participants that differ from using adults? What risk factors might be present if any? Who should approve the plan for the study and how do you contact them?
SOURCE: This Action Research scenario is roughly based on Rubin, D. (2000). Race and self-esteem: A study of Latino and European-American students with learning disabilities. Unpublished masters thesis, University of Utah, Salt Lake City.
Harm Psychologists must take reasonable steps to avoid harming their clients/ patients, students, supervisees, research participants, organizational clients, and others with whom they work, and to minimize harm where it is foresee- able and unavoidable. (American Psychological Association, 2002, p. 6)
When psychologists become aware that research procedures have harmed a participant, they take reasonable steps to minimize the harm. (American Psychological Association, 2002, p. 12)
Participants, or their guardians, in an [educational] research study have the right to be informed about the likely risks involved in the research and of potential consequences for participants, and to give their informed consent before participating in research.... Participants in research should be made aware of the limits on the protections that can be provided, and of the efforts toward protection that will be made even in situations where absolute confi- dentiality cannot be assured. It should be made clear to informants and par- ticipants that despite every effort made to preserve it, confidentiality may be compromised. (American Educational Research Association, 2005)
The most basic concern in all research is that no individual is harmed by serving as a participant, as suggested above by the APA and AERA codes of ethics. In the context of research ethics, harm may be broadly defined to include extreme physical pain or death, but also involves such factors as psychological stress, personal embar- rassment or humiliation, or myriad influences that may adversely affect the partici- pants in a significant way.
Po t e n t i a l H a z a r d s t o Pa r t i c i p a n t s Certain types of investigations present potential harm to participants. Research that involves physically dangerous treatment may present real possibilities for harm if the treatment is “inflicted” on the participants. Unfortunately, there are examples of investigations in which ethical principles were violated in an extreme fashion (see Young, 2005). Other areas of research are specifically designed to investigate the effects of psychological or emotional stress. Such research represents extremely diffi- cult circumstances, especially when the procedures involve actual “infliction” of stress. There is always the possibility that a subject may become seriously ill (e.g., have a stroke or heart attack) as a result of the stress. In addition, the possibility exists that the stress itself may be harmful to participants from a psychological standpoint.
Vulnerable populations. Some studies may be characterized as high risk for harm because of the participants involved. By virtue of characteristics associated with age or disability, an individual can be rendered relatively powerless in exercising free will when choosing whether or not to serve as a research participant. These partici- pants may be less capable of understanding potential harm, or they may feel openly or implicitly coerced in some fashion. People who are institutionalized or incarcer- ated, such as prisoners, persons with severe disabilities, or people with serious mental illness, may agree to participate in a study either because they “should to show evidence of good behavior” or to gain approval from supervisors. Unfortunately, some troubling examples of ethical violations have occurred with studies involving these individuals (Field & Behrman, 2004; Moser et al., 2004). Highly vulnerable populations should not be taken advantage of in the name of science. Researchers investigating topics involving these individuals must exercise extreme care. Very young children, the elderly, or people with disabilities may be easily convinced that most activities are important, are of little harm, and should be engaged in for the benefit of society (Drew & Hardman, 2007; Quadagno, 2005).
64 PART I: O VERVIEW OF R ESEARCH
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 65
D e t e r m i n i n g t h e D e g r e e o f H a r m
Vulnerable participants pose a dilemma because of the important need to study these individuals in order to improve their lives and to benefit society. Consequently, a researcher will likely encounter such ethical dilemmas sooner or later. Thoughtful and knowledgeable consideration must guide the process. The researcher must determine what constitutes significant risk, and carefully weigh potential benefits versus potential harm to the subjects.
How much is harmful? Researchers continually face questions regarding what is meant by a high degree of vulnerability and a potential, substantial, or significant risk. Although the above terms suggest quantification, they refer to issues that are diffi- cult to quantify. Basically, such terminology focuses on perceptions, values, or judg- ment. As such, there is little probability of consensus concerning their meaning. Even so, these issues cannot be ignored and every researcher must eventually face the question, “How much [harm] is harmful?” Researchers must decide how much threat, stress, or pain should be considered harmful. Their determination might fall near one end or the other of the contin- uum (i.e., nothing should ever be done in research that would pose the slightest imposition or threat of stress on a subject versus the interests of science and society are always paramount—individual distress or harm should be discounted in favor of the potential benefit for the larger number). Neither of these extremes can be accepted as a blanket rule. The continuum must be explored more completely and judicious caution exercised. Some researchers may attempt to fall back on consent as a means of dealing with issues related to harm. They may argue that potentially harmful studies can only be undertaken if the participants have effectively consented (have the capacity to understand the potential risk, full information concerning any harm to them that may occur, and voluntarily agree to subject themselves to such potential harm). However, since consent is an essential prerequisite to these investigations, it adds little to our understanding of how much harm is acceptable. Consent, even under the most careful precautions, does not render an investigator free to ignore further responsibility regarding potential harm.
What is the cost/benefit ratio? Another approach to addressing the issue of harm is the cost/benefit ratio. This approach involves a comparison of the potential bene- fits of a given study with the potential risks to the participants. Presumably, if the benefit of the study outweighs the potential harm, the study is considered ethical, and the opposite would also be true. For example, suppose an investigation were proposed that had potential for solving the inflation problem in the United States. However, to conduct the study, a sample of 1,000 participants would be required to divulge personal details concerning amount and sources of income, as well as the manner in which every dollar was spent. Certainly this represents a substantial inva- sion of privacy—but there is also a dire need to solve the inflation problem. Do the benefits outweigh the harm? This becomes a matter of individual judgment. The idea of assessing costs and benefits has considerable appeal. Yet how is this task accomplished? It may be a rather simple matter in business, but in many cases we are confronted with factors that defy measurement. How does one evaluate and predict the harm that might be inflicted on a participant? An equally difficult ques- tion involves how one measures and predicts benefit. Or, how does the process of balancing subject harm with potential benefit proceed? These problems lack precise solutions and require careful evaluation. The evaluation process must involve a review by professionals other than the investigator proposing the study (often an institutional or agency “human subjects review board”). If assessments were the sole prerogative of the researcher, a conflict of interest could influence the decision. (After all, the researcher has a vested interest in the study.)
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 67
public. This makes it possible for the public to identify individuals. Such a breach of confidence should never occur, but this example is not without some base in reality. The researcher must remain very alert concerning the degree to which private infor- mation remains confidential, particularly when such information is of a sensitive nature. A researcher cannot ignore privacy in any study. If a potential problem exists, consent should always be obtained. In addition, the participants should be assured that the data will be held in strict confidence to protect anonymity. Few studies require that data be maintained in a form in which participants can be personally identified. As a rule of thumb, researchers should invade the privacy of participants as minimally as possible. If a potential of privacy risk exists, the investigator should take every precaution. Any study that substantially invades privacy may obviously place the participants at risk. Such procedures also place the conduct of science at risk by reducing the public trust vital to the successful pursuit of knowledge.
Deception
Research deception involves an intentional misrepresentation of facts related to the purpose, nature, or consequences of an investigation. In this context, deception refers to either an omission or a commission on the part of the researcher in terms of interactions with participants. An omission deception could mean that the investi- gator does not fully inform participants about important aspects of the study. Part or all of the information is withheld. A commission involves a situation in which the researcher gives false information about the investigation, either partially or totally. In the first case, participants may not even be aware that a study is in progress, or they may only be informed about a portion of the investigation. In the second situ- ation, participants generally are aware that they are involved in some type of study or activity that is out of the ordinary, but may be given misleading information regarding the actual purpose of the study or activity. In either case, the researcher is misrepresenting the study. Regardless of the precise nature of deception, it has become a very prominent issue for investigators concerned with the ethics of conducting research. As we move through the first decade of the 21st century, deception is receiving widespread attention in educational and social science research with increasing concerns regarding its use on the Internet (Keller & Lee, 2003; Lichtenberg, Heresco-Levy, & Nitzan, 2004; Mishara & Weisstub, 2005; Nagy, 2005c; Pittenger, 2003).
R e a s o n s f o r D e c e p t i o n
There are a variety of reasons that deception may be used in research. These include deceiving participants who have become “test-wise” and are suspicious that the pur- pose of the study is hidden, reducing the amount of influence a group can have on an individual respondent, and pragmatics (limited finances, time, and data sources).
Hidden agenda. Some research participants have become “test-wise” and often believe the “real” purpose of a given study is hidden. This is truer for certain popula- tions than others (notably college sophomores, one of the most “studied” groups in the United States). If participants are that suspicious, they may be expected to answer, perform, or generally respond in an atypical manner. Such responses may seriously threaten the soundness of an investigation. Participants may attempt to respond in a manner that they think the researcher desires, or they may try to out- guess the researcher and “sabotage” the study. Some populations are studied fre- quently, and others are even professional research participants. (Some college students earn part of their monthly income by serving as participants in studies about response to advertisements.) There is even evidence that participants are “wis- ing up” to the research process.
The APA Code of Ethics notes that “Psychologists do not conduct a study involv- ing deception unless they have determined that the use of deceptive techniques is justified by the study’s significant prospective scientific, educational, or applied value and that effective non-deceptive alternative procedures are not feasible” (American Psychological Association, 2002, p. 11). APA continues by noting that deception is not used if it can be expected to cause physical or emotional suffering, and, if used, the deception must be explained to participants as early as feasible.
Individual influences. Suppose a study were being conducted on group control of individual actions. In this type of investigation, the study might focus on the amount of influence a group had on individual opinions about the use of drugs. This might be the type of study that would involve intentional deception to a con- siderable degree. If the participant being observed was suspicious about the purpose of the investigation, he or she might be more or less resistant to group pressure than normally would occur. The researcher would need to account for such an influence by deceiving the participant. The individual may be directly deceived about the pur- pose of the study, or information may be withheld. If the participant were openly informed about the nature of the study, the outcome would be jeopardized. In this case, the use of intentional deception could backfire. The researcher cannot reveal the exact nature of the investigation, but the participant may be generally suspicious and alter behavior because of those suspicions in a manner unknown to the investi- gator. Intentional deception is used to control factors important to the study, but suspicion of deception creates problems for control. The control exerted may relate to the technical soundness of the study or it may relate to the generalizability of the findings (matters that will be examined in detail in later chapters). If one group of participants receives different treatment (decep- tion) than another, the comparison may not really involve the treatment. Likewise, if deception is either real or perceived by participants, results might not be transfer- able to the outside world. Methodological reasons for using intentional deception represent a double-edged sword: they have both advantages and disadvantages. The judgment of the investigator regarding this issue will likely be tested.
Pragmatics. Some reasons for using deception are purely pragmatic. Limited finances, time, and data sources have often led researchers to use deception. For example, if it is too expensive, time consuming, or not feasible to observe a natural occurrence of a particular phenomenon, it might be best studied by creating a sim- ulated incident. Such a situation may arise when the researcher is studying some- thing that only occurs rarely or creates dangerous circumstances when it does occur (e.g., emergencies). This type of deception could involve a staged purse-snatching or mugging with observations focused on the reactions of witnesses. Such an example would necessarily involve deception (and also no pre-study consent) if the witnesses are to behave naturally.
Is deception ethical? The previous discussions illustrate the reasons why researchers use deception and also some of the problems involved in its use. But is deception ethical? One of the most fundamental arguments against deception is that it is unethical and degrading to the participants involved. Lying is generally viewed unfavorably in our society and even deemed immoral. This value places the conduct of science in a dilemma. Science is presumably undertaken for the benefit of society, so should it be associated with acts that are generally thought to be malicious? This is a question that has no direct answer since it is couched in such polarized terms. Most people would agree that research cannot be clearly declared either ethical or unethical. There are many shades of gray and myriad matters that must be con- sidered. One of the most serious issues facing researchers and the use of deception is the issue of consent. If participants are involved in activities that present potential
68 PART I: O VERVIEW OF R ESEARCH
In addition to the problems related to participants who receive an experimental treatment, there are also difficult ethical issues involving those who are in a “placebo” or control group. Such would be the case where one group of students in a high school receives a newly developed science program (experimental treatment) that appears to be very effective, and a second group receives the science program that was used for many years with limited effectiveness (control group). One ethical perspective is that the researcher has the responsibility to provide the new treatment to all participants. However, some researchers may have a very different view. This opposing perspective is often called the natural state argument. This argument con- tends that the untreated participants are not being denied a benefit they already have; they are merely being left in their natural state. In the example above, the high school students in the control group continued to receive the science program that had been used in the school for many years. Clearly, neither of the above positions is acceptable for all research (Field & Behrman, 2004; Gross, 2005; Roberts et al., 2005). There remains a critical need to conduct a scientifically sound test. One approach to this dilemma represents a com- promise that attempts to meet the goals of research and also the ethical need for pro- viding benefit to all participants. In our example above, it may be possible to first conduct the study as initially planned—with a treatment group and a placebo com- parison group. Then, after the investigation is complete and the data collected, researchers could provide the new science program to the participants in the compar- ison group if in fact it was a more effective program. In this way, treatment is not with- held from anyone in the study, and the data are not contaminated. Although this is a reasonable compromise, ethical issues remain. Some would contend that the delay in treatment is still harmful to the placebo group. This may be true if the duration of the study is extended over time, but if not, such an argument loses its potency. Another issue may be expense: if the treatment is quite expensive, the researcher may have dif- ficulty funding treatment for the second group. Although this may be a problem, it is not a very compelling argument. To remain ethical, the researcher should plan a bud- get to allow for treatment of the comparison participants. None of this discussion has great significance for interventions or treatments of unknown effectiveness, but becomes more relevant as evidence of promise accumulates for any given treatment. Ethical issues also exist in conducting nonexperimental research where an inves- tigator does not impose or manipulate conditions. Although ethics in nonexperi- mental designs (e.g., survey research) are often less complex or harmful than experimental studies, it is important for investigators to be aware of basic principles for protecting the participants, including “full disclosure and consent.” For example, in survey research, each respondent should be fully informed as to the purpose of the study, participant demographics (e.g., teachers, college students, the general public), confidentiality of responses, how the results are intended to be used, and who will have access to the data. Bacon and Olsen (2005) also indicate that survey researchers have the ethical responsibility of “not wasting” a respondent’s time and to only collect data that has utility (real use). Schenk and Williamson (2005), in dis- cussing the ethical responsibilities involved in conducting nonexperimental research on children, suggest “if the information gathering activity will not directly benefit the children involved or their community, do not proceed” (p. 17).
Ethics and Qualitative Research Qualitative research involves data that are recorded in narrative descriptions, not numbers. Researchers use qualitative methods to observe and describe conditions rather than control them. A basic ethical principle for qualitative researchers is this: Do not tamper with the natural setting under study.
70 PART I: O VERVIEW OF R ESEARCH
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 71
Participant and nonparticipant observations are integral components of qualitative research and are used widely in the fields of education, sociology, and anthropology. Each presents unique ethical issues in regard to consent, privacy, and deception (Brinkmann & Kvale, 2005; Haverkamp, 2005). Informed consent is necessary but can be problematic when relying on observa- tions in a qualitative research study. Although potential harm from treatment is not generally a threat, there are other ethical concerns. Clearly, there is a substantial threat to privacy. A revelation of observed conversations and behaviors could cause harm to participants in their families, communities, or place of employment. In addition, the actual research participants, who have given consent, may not be the only people observed. In natural settings, people move in and out of interactions and settings for many reasons (Creswell, 2005; Denzin & Lincoln, 2005). For example, a researcher has permission to study formal and informal counselor–student interac- tions in a middle school setting. Informed consent has been obtained from the coun- selors, students, and parents of the students who are in formal counseling. However, the researcher plans to follow the counselors while they work during the day and record informal interactions among students throughout the school and between the counselors and the students. How is consent obtained from all the students and their parents? Is consent necessary? The accepted rule of thumb for such nonparticipant observation research is that consent is not necessary when (a) access to the setting is approved by the agency or institution, (b) participants who are actively involved have given informed consent, and (c) other observed behavior is considered public and observable by anyone present in the setting. In other words, if other students or teachers in the school overhear the conversations that the researcher is observing in a public place, they do not require additional consent. This rule of thumb does not always guarantee, however, that difficulties with consent will not occur. The definition of “public” is open to debate. For example, a doctoral student at a leading university was recently conducting nonparticipant observational research for his dissertation. The student obtained consent to observe a YWCA board meeting from the president and members of the board. However, an influential community member speaking at the meeting refused to participate when told he was under observation as part of a research study. The entire board meeting came to a halt.
Pa r t i c i p a n t O b s e r v a t i o n
Participant observation also presents unique ethical challenges. The participant observer often lives, eats, and sleeps on a daily basis with those under observation. In such a study, a broad range of details is observed and recorded regarding participants’ cultural mores, interaction patterns, social structures, and daily behaviors. The partici- pant observer may directly probe into many facets of the lifestyle of the participants— public, private, and often sensitive. With such a pervasive scope of observation, an invasion of privacy is inevitable. This makes informed consent absolutely necessary. The nature of participant observation research raises some serious difficulties with regard to consent. Specifically, this research methodology often requires that only the core group of participants know the researcher is not just another member of the group. Is it adequate to obtain consent only from the group leader (e.g., the tribal leader, the school principal, the teacher, the fraternity president)? Another problem in observation studies involves the effects of obtaining consent on the behavior of those being studied. An important strength of observation research is the study of participant routine behaviors within the natural environment. By obtain- ing consent, a researcher may alter this natural behavior. Many researchers seek to mitigate these effects by asking an inside informer to watch for changes in the
C HAPTER 3: E THIC AL I SSUES IN C ONDUCTING R ESEARCH 73
Psychologists [must] inform participants about (1) the purpose of the research, expected duration, and procedures; (2) their right to decline to participate and to withdraw from the research once participation has begun; (3) the foresee- able consequences of declining or withdrawing; (4) reasonably foreseeable factors that may be expected to influence their willingness to participate such as potential risks, discomfort, or adverse effects; (5) any prospective research benefits; (6) limits of confidentiality; (7) incentives for participation; and (8) whom to contact for questions about the research and research participants’ rights. They provide opportunity for the prospective participants to ask ques- tions and receive answers. (p. 11)
BOX 3.2 Ethics of Undercover Research
Ethnographies, or detailed descriptions of a culture, require anthropologists to immerse themselves in the day-to-day lives of their subjects. But what if immersion involves a researcher not revealing his or her identity? This question was important for Rebekah Nathan, who determined that she could not have made a meaningful study of students unless she blended in with them—and did not fully explain who she was. That, the professor writes in My Freshman Year, required “a delicate balancing act between truth and fiction.” (Rebekah Nathan is a pseudonym, and her university is not identified in her book.) Ms. Nathan says she followed the ethical protocol of “informed consent” when conducting formal interviews with students. She identified herself as a researcher (though not as a professor at the university), explained her study, and obtained written permission to publish the subjects’ words. However, in day-to-day interactions, Ms. Nathan let students assume she was one of them. Lois Weis, a professor of sociology of education at the State University of New York at Buffalo who read the book, does not think Ms. Nathan compromised her ethical obligations as a researcher. “It seems to me that she made wise decisions,” says Ms. Weis, who has conducted numerous ethnographical studies and is the coauthor of Speed Bumps: A Student-Friendly Guide to Qualitative Research (Teachers College Press, 2000). “Certainly in this kind of research there can be a lot of misunderstandings, miscommunications, and certainly research subjects could end up feeling hurt or used,” Ms. Weis says. “But that could happen whether they knew that she was a researcher or not.” Margaret A. Eisenhart, a professor of educational anthropology and research methodology at the University of Colorado at Boulder, who also read My Freshman Year, disagreed with Ms. Nathan’s undercover approach.
Protecting the Subject
“My job as an ethnographer is to protect the people I’m working with in the same way I’d want to be protected if I were in the subject’s shoes,” says Ms. Eisenhart. “It is just simply unethical to do something which is primarily for your own benefit, to advance your own career.” She rejects the idea that Ms. Nathan would have jeopardized her research if she had been up front with her subjects. Ms. Eisenhart says that in one of her own studies she was able to gain the trust of students and accompany them on social excursions, even dates, without concealing her identity. That research was the source for the book Educated in Romance: Women, Achievement, and College Culture (University of Chicago Press, 1990), which she wrote with Dorothy C. Holland. “I don’t think it was hard to get people to open up once they relaxed about who we were,” Ms. Eisenhart says of their study. The American Anthropological Association’s code of ethics does not explicitly advise against undercover research, though it once did. Some anthropologists believe the practice is unethical in all cases, while others say it depends on the context, says Stacy Lathrop, staff liaison to the association’s ethics committee. “We all play multiple roles in our life, and you have to navigate between those,” Ms. Lathrop says. “Participant obser- vation is like life, dynamic and changing. As in human relationships, sometimes you do keep secrets.”
SOURCE: From Hoover, E. (2005, July 29). “The ethics of undercover research,” The Chronicle of Higher Education , A37. Reprinted with permission from The Chronicle of Higher Education.
Many researchers are supportive of IRBs and view them as a protection for not only human participants but researchers as well. The IRB can assist the researcher by identifying potential pitfalls that could be overlooked by an individual investiga- tor, thus protecting reputations and avoiding litigation. Other researchers see the IRB as a governmental intrusion into the research process. They view them as overly cumbersome, unnecessary, and dictatorial. From the latter viewpoint, IRBs were established to deal with the very few who violate ethical standards, while punishing the many ethical researchers with unwieldy governmental requirements (see, for example, Kleiman, 2003; Rae & Sullivan, 2003).
Federal Requirements for IRBs Regardless of the position taken on the need for IRBs, they are a federal require- ment. Institutions receiving federal funding must establish committees to approve all human and animal research. This includes student research for theses and disser- tations, as well as individual faculty endeavors. Investigators must receive IRB approval on any research involving human or animal participants whether or not the study is receiving federal dollars. An institution that fails to meet IRB require- ments may put its federal funding in jeopardy. Collegial review bodies such as IRBs do not negate the individual researcher’s responsibility for ethical decision making. The final ethical responsibility rests with the investigator.
ETHICS AND PROFESSIONALISM
Ethics and professionalism could easily consume an entire volume, but will only receive a cursory examination in this chapter. Our focus is on three issues: (1) integrity during the execution of the study, (2) integrity related to publishing, and (3) sanctions for breaches of integrity. Hopefully, these topics will provide a background and sensitivity that will prompt further thought and inquiry by the beginning researcher as questions arise.
Integrity During Execution of the Study Any breach of integrity during the execution of a research study, whether it be unin- tentional errors or outright falsification of the data, seriously weakens or even inval- idates the investigation. One of the basic purposes of science is the acquisition of objective and accurate data about real phenomena. Reality is fluid, situational, and certainly variable depending on one’s perspective. However, somewhere underlying the conduct of science, there is a philosophical assumption that some “truths” can be determined. This places a very important responsibility on scientists to under- take their efforts in a totally honest fashion. This is not a statement that reflects a value judgment or a moral stand. It simply emphasizes a fundamental principle on which scientific investigation is based. This chapter has addressed the importance of integrity in research and focused thus far on the execution of the study. The question raised in this context relates to why breaches of integrity occur. Although a variety of influences result in these problems, none makes such acts justifiable. However, their discussion assists our understanding and promotes preventive action. Factors influencing breaches of integrity involve some rather powerful social pressures, particularly among beginning researchers. One of the most powerful fac- tors is the pressure to publish, often characterized as a “publish or perish” mentality. Researchers are constantly pressured to “get into print,” which is always particularly acute early in a career (Parasuraman, 2003; Weerasekera, 2004). This pressure may
74 PART I: O VERVIEW OF R ESEARCH
