Ethical issues with milgram's obedience study

ETHICS IN BEHAVIORAL RESEARCH CHP. 3

LEARNING OBJECTIVES

· Summarize Milgram's obedience experiment.

· Discuss the three ethical principles outlined in the Belmont Report: beneficence, autonomy, and justice.

· Define deception and discuss the ethical issues surrounding its use in research.

· List the information contained in an informed consent form.

· Discuss potential problems in obtaining informed consent.

· Describe the purpose of debriefing research participants.

· Describe the function of an Institutional Review Board.

· Contrast the categories of risk involved in research activities: exempt, minimal risk, and greater than minimal risk.

· Summarize the ethical principles in the APA Ethics Code concerning research with human participants.

· Summarize the ethical issues concerning research with nonhuman animals.

· Discuss how potential risks and benefits of research are evaluated.

· Discuss the ethical issue surrounding misrepresentation of research findings.

· Define plagiarism and describe how to avoid plagiarism.

Page 44ETHICAL PRACTICE IS FUNDAMENTAL TO THE CONCEPTUALIZATION, PLANNING, EXECUTION, AND EVALUATION OF RESEARCH. Researchers who do not consider the ethical implications of their projects risk harming individuals, communities, and behavioral science. This chapter provides an historical overview of ethics in behavioral research, reviews core ethical principles for researchers, describes relevant institutional structures that protect research participants, and concludes with a discussion of what it means to be an ethical researcher.

MILGRAM'S OBEDIENCE EXPERIMENT

Stanley Milgram conducted a series of studies (1963, 1964, 1965) to study obedience to authority. He placed an ad in the local newspaper in New Haven, Connecticut, offering a small stipend to men to participate in a “scientific study of memory and learning” being conducted at Yale University. The volunteers reported to Milgram's laboratory at Yale, where they met a scientist dressed in a white lab coat and another volunteer in the study, a middle-aged man named “Mr. Wallace.” Mr. Wallace was actually a confederate (i.e., accomplice) of the experimenter, but the participants did not know this. The scientist explained that the study would examine the effects of punishment on learning. One person would be a “teacher” who would administer the punishment, and the other would be the “learner.” Mr. Wallace and the volunteer participant then drew slips of paper to determine who would be the teacher and who would be the learner. The drawing was rigged, however—Mr. Wallace was always the learner and the volunteer was always the teacher.

The scientist attached electrodes to Mr. Wallace and placed the teacher in front of an impressive-looking shock machine. The shock machine had a series of levers that, the individual was told, when pressed would deliver shocks to Mr. Wallace. The first lever was labeled 15 volts, the second 30 volts, the third 45 volts, and so on up to 450 volts. The levers were also labeled “Slight Shock,” “Moderate Shock,” and so on up to “Danger: Severe Shock,” followed by red X's above 400 volts.

Mr. Wallace was instructed to learn a series of word pairs. Then he was given a test to see if he could identify which words went together. Every time Mr. Wallace made a mistake, the teacher was to deliver a shock as punishment. The first mistake was supposed to be answered by a 15-volt shock, the second by a 30-volt shock, and so on. Each time a mistake was made, the learner received a greater shock. The learner, Mr. Wallace, never actually received any shocks, but the participants in the study did not know that. In the experiment, Mr. Wallace made mistake after mistake. When the teacher “shocked” him with about 120 volts, Mr. Wallace began screaming in pain and eventually yelled that he wanted out. What if the teacher wanted to quit? This happened—the volunteer participants became visibly upset by the pain that Mr. Wallace seemed to be experiencing. The experimenter told the teacher that he could Page 45quit but urged him to continue, using a series of verbal prods that stressed the importance of continuing the experiment.

The study purportedly was to be an experiment on memory and learning, but Milgram really was interested in learning whether participants would continue to obey the experimenter by administering ever higher levels of shock to the learner. What happened? Approximately 65% of the participants continued to deliver shocks all the way to 450 volts.

Milgram went on to conduct several variations on this basic procedure with 856 subjects. The study received a great deal of publicity, and the results challenged many of our beliefs about our ability to resist authority. The Milgram study is important, and the results have implications for understanding obedience in real-life situations, such as the Holocaust in Nazi Germany and the Jonestown mass suicide (see Miller, 1986).

But the Milgram study is also an important example of ethics in behavioral research. How should we make decisions about whether the Milgram study or any other study is ethical? The Milgram study was one of many that played an important role in the development of ethical standards that guide our ethical decision making.

What do you think? Should the obedience study have been allowed? Were the potential risks to Milgram's participants worth the knowledge gained by the outcomes? If you were a participant in the study, would you feel okay with having been deceived into thinking that you had harmed someone? What if it was a younger sibling? Or an elderly grandparent? Would that make a difference? Why or why not?

In this chapter, we work through some of these issues, and more. First, let us turn to an overview of the history of our current standards to help frame your understanding of ethics in research.

HISTORICAL CONTEXT OF CURRENT ETHICAL STANDARDS

Before we can delve into current ethical standards, it is useful to briefly talk about the origin of ethics codes related to behavioral research. Generally speaking, modern codes of ethics in behavioral and medical research have their origins in three important documents.

The Nuremberg Code and Declaration of Helsinki

Following World War II, the Nuremberg Trials were held to hear evidence against the Nazi doctors and scientists who had committed atrocities while forcing concentration camp inmates to be research subjects. The legal document that resulted from the trials contained what became known as the Nuremberg Code: a set of 10 rules of research conduct that would help prevent future research atrocities (see http://www.hhs.gov/ohrp/archive/nurcode.html).

Page 46The Nuremberg Code was a set of principles without any enforcement structure or endorsement by professional organizations. Moreover, it was rooted in the context of the Nazi experience and not generally seen as applicable to general research settings. Consequently, the World Medical Association developed a code that is known as the Declaration of Helsinki. This 1964 document is a broader application of the Nuremberg that was produced by the medical community and included a requirement that journal editors ensure that published research conform to the principles of the Declaration.

The Nuremberg Code and the Helsinki Declaration did not explicitly address behavioral research and were generally seen as applicable to medicine. In addition, by the early 1970s, news about numerous ethically questionable studies forced the scientific community to search for a better approach to protect human research subjects. Behavioral scientists were debating the ethics of the Milgram studies and the world was learning about the Tuskegee Syphilis Study, in which 399 African American men in Alabama were not treated for syphilis in order to track the long-term effects of this disease (Reverby, 2000). This study, supported by the U.S. Public Health Service, took place from 1932 to 1972, when the details of the study were made public by journalists investigating the study. The outrage over the fact that this study was done at all and that the subjects were African Americans spurred scientists to overhaul ethical regulations in both medical and behavioral research. The fact that the Tuskegee study was not an isolated incident was brought to light in 2010 when documentation of another syphilis study done from 1946 to 1948 in Guatemala was discovered (Reverby, 2011). Men and women in this study were infected with syphilis and then treated with penicillin. Reverby describes the study in detail and focuses on one doctor who was involved in both the Guatemala and Tuskegee studies.

As a result of new public demand for action, a committee was formed that eventually produced the Belmont Report . Current ethical guidelines for both behavioral and medical researchers have their origins in The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979). This report defined the principles and applications that have guided more detailed regulations developed by the American Psychological Association and other professional societies and U.S. federal regulations that apply to both medical and behavioral research investigations.

The three basic ethical principles of the Belmont Report are:

· Beneficence—research should confer benefits and risks must be minimal. The associated application is the necessity to conduct a risk-benefit analysis.

· Respect for persons (autonomy)—participants are treated as autonomous; they are capable of making deliberate decisions about whether to participate in research. The associated application is informed Page 47consent—potential participants in a research project should be provided with all information that might influence their decision on whether to participate.

· Justice—there must be fairness in receiving the benefits of research as well as bearing the burdens of accepting risks. This principle is applied in the selection of subjects for research.

APA ETHICS CODE

The American Psychological Association (APA) has provided leadership in formulating ethical principles and standards. The Ethical Principles of Psychologists and Code of Conduct—known as the APA Ethics Code —is amended periodically with the current version always available online at http://apa.org/ethics/code. The Ethics Code applies to psychologists in their many roles including teachers, researchers, and practitioners. We have included the sections relevant to research in Appendix B.

APA Ethics Code: Five Principles

The APA Ethics Code includes five general ethical principles: beneficence and nonmaleficence, fidelity and responsibility, integrity, justice, and respect for rights and responsibilities. Next, we will discuss the ways that these principles relate to research practice.

Principle A: Beneficence and Nonmaleficence As in the Belmont Report, the principle of Beneficence refers to the need for research to maximize benefits and minimize any possible harmful effects of participation. The Ethics Code specifically states: “Psychologists strive to benefit those with whom they work and take care to do no harm. In their professional actions, psychologists seek to safeguard the welfare and rights of those with whom they interact professionally and other affected persons and the welfare of animal subjects of research.”

Principle B: Fidelity and Responsibility The principle of Fidelity and Responsibility states: “Psychologists establish relationships of trust with those with whom they work. They are aware of their professional and scientific responsibilities to society and to the specific communities in which they work.” For researchers, such trust is primarily applicable to relationships with research participants.

Researchers make several implicit contracts with participants during the course of a study. For example, if participants agree to be present for a study at a specific time, the researcher should also be there. If researchers promise to send a summary of the results to participants, they should do so. If participants are to receive course credit for participation, the researcher must immediately let the instructor know that the person took part in the study. These may seem Page 48to be little details, but they are very important in maintaining trust between participants and researchers.

Principle C: Integrity The principle of Integrity states: “Psychologists seek to promote accuracy, honesty and truthfulness in the science, teaching and practice of psychology. In these activities psychologists do not steal, cheat or engage in fraud, subterfuge or intentional misrepresentation of fact.” Later in this chapter, we will cover the topic of integrity in the context of being an ethical researcher.

Principle D: Justice As in the Belmont Report, the principle of Justice refers to fairness and equity. Principle D states: “Psychologists recognize that fairness and justice entitle all persons to access to and benefit from the contributions of psychology and to equal quality in the processes, procedures and services being conducted by psychologists.”

Consider the Tuskegee Syphilis study, or the similar study conducted in Guatemala. In both cases there was a cure for syphilis (i.e., penicillin) that was withheld from participants. This is a violation of principle D of the APA Ethics Code and a violation of the Belmont Report's principle of Justice.

Principle E: Respect for People's Rights and Dignity The last of the five APA ethical principles builds upon the Belmont Report principle of Respect for Persons. It states: “Psychologists respect the dignity and worth of all people, and the rights of individuals to privacy, confidentiality, and self-determination. Psychologists are aware that special safeguards may be necessary to protect the rights and welfare of persons or communities whose vulnerabilities impair autonomous decision making. Psychologists are aware of and respect cultural, individual, and role differences, including those based on age, gender, gender identity, race, ethnicity, culture, national origin, religion, sexual orientation, disability, language, and socioeconomic status, and consider these factors when working with members of such groups. Psychologists try to eliminate the effect on their work of biases based on those factors, and they do not knowingly participate in or condone activities of others based upon such prejudices.”

One of the ethical dilemmas in the Milgram obedience study was the fact that participants did not know that they were participating in a study of obedience. This limited participants’ rights to self-determination. Later, we will explore this issue in depth.

Protecting Research Subjects

The preamble to the APA Ethics Code states: “Psychologists are committed to increasing scientific and professional knowledge of behavior and people's understanding of themselves and others and to the use of such knowledge to improve the condition of individuals, organizations and society.” By internalizing and adhering to ethical principles we support and nurture a healthy science. Page 49With this in mind, we will consider the ways in which research subjects—humans and animals—are protected in behavioral research.

ASSESSMENT OF RISKS AND BENEFITS

The principle of beneficence leads us to examine potential risks and benefits that are likely to result from the research; this is called a risk-benefit analysis. Ethical principles require asking whether the research procedures have minimized risk to participants.

The potential risks to the participants include such factors as psychological or physical harm and loss of confidentiality; we will discuss these in detail. In addition, the cost of notconducting the study if in fact the proposed procedure is the only way to collect potentially valuable data can be considered (cf. Christensen, 1988). The benefits include direct benefits to the participants, such as an educational benefit, acquisition of a new skill, or treatment for a psychological or medical problem. There may also be material benefits such as a monetary payment, some sort of gift, or even the possibility of winning a prize in a raffle. Other less tangible benefits include the satisfaction gained through being part of a scientific investigation and the potential beneficial applications of the research findings (e.g., the knowledge gained through the research might improve future educational practices, psychotherapy, or social policy). As we will see, current regulations concerning the conduct of research with human participants require a risk-benefit analysis before research can be approved.

Risks in Behavioral Research

Let's return to a consideration of Milgram's research. The risk of experiencing stress and psychological harm is obvious. It is not difficult to imagine the effect of delivering intense shocks to an obviously unwilling learner. A film that Milgram made shows participants protesting, sweating, and even laughing nervously while delivering the shocks. You might ask whether subjecting people to such a stressful experiment is justified, and you might wonder whether the experience had any long-range consequences for the volunteers. For example, did participants who obeyed the experimenter feel continuing remorse or begin to see themselves as cruel, inhumane people? Let's consider some common risks in behavioral research.

Physical harm Procedures that could conceivably cause some physical harm to participants are rare but possible. Many medical procedures fall into this category, for example, administering a drug such as alcohol or caffeine. Other studies might expose subjects to physical stressors such as loud noise, extreme hot or cold temperatures, or deprivation of sleep for an extended period of. The risks in such procedures require that great care be taken to make them ethically acceptable. Moreover, there would need to be clear benefits of the research that would outweigh the potential risks.

Page 50Stress More common than physical stress is psychological stress. The participants in the Milgram study were exposed to a high level of stress; they believed that they were delivering fatal doses of electricity to another person. Milgram described one of his participants:

While continuing to read the word pairs with a show of outward strength, she mutters in a tone of helplessness to the experimenter, “Must I go on? Oh, I'm worried about him. Are we going all the way up there (pointing to the higher end of the generator)? Can't we stop? I'm shaking. I'm shaking. Do I have to go up there?”

She regains her composure temporarily but then cannot prevent periodic outbursts of distress (Milgram, 1974, p. 80).

There are other examples. For instance, participants might be told that they will receive some extremely intense electric shocks. They never actually receive the shocks; it is the fear or anxiety during the waiting period that is the variable of interest. Research by Schachter (1959) employing a procedure like this showed that the anxiety produced a desire to affiliate with others during the waiting period.

In another procedure that produces psychological stress, participants are given unfavorable feedback about their personalities or abilities. Researchers may administer a test that is described as a measure of social intelligence and then told that they scored very high or very low. The impact of this feedback can then be studied. Asking people about traumatic or unpleasant events in their lives might also cause stress for some participants. Thus, research that asks people to think about the deaths of a parent, spouse, or friend, or their memories of living through a disaster could trigger a stressful reaction.

When using procedures that may create psychological distress, the researcher must ask whether all safeguards have been taken to help participants deal with the stress. Usually a debriefing session following the study is designed in part to address any potential problems that may arise during the research.

Confidentiality and privacy Another risk is the loss of expected privacy and confidentiality. Confidentiality is an issue when the researcher has assured subjects that the collected data are only accessible to people with permission, generally only the researcher. This becomes particularly important when studying topics such as sexual behavior, divorce, family violence, or drug abuse; in these cases, researchers may need to ask people very sensitive questions about their private lives. Or consider a study that obtained information about employees’ managers. It is extremely important that responses to such questions be confidential; revealing the responses of an individual could result in real harm. In most cases, researchers will attempt to avoid confidentiality problems by making sure that the responses are completely anonymous—there is no way to connect any person's identity with the data. This happens, for example, when questionnaires are administered to groups of people and no Page 51information is asked that could be used to identify an individual (such as name, taxpayer identification number, email address, or phone number). However, in other cases, such as a personal interview in which the identity of the person might be known, the researcher must carefully plan ways of coding data, storing data, and explaining the procedures to participants so that there is no question concerning the confidentiality of responses.

Invasion of privacy becomes an issue when the researcher collects information under circumstances that the subject believes are private—free from unwanted observation by others. In some studies, researchers make observations of behavior in public places without informing the people being observed. Observing people as they are walking in a public space, stopped at a traffic light, or drinking in a bar does not seem to present any major ethical problems. However, what if a researcher wishes to observe behavior in more private settings or in ways that may violate individuals’ privacy (see Wilson & Donnerstein, 1976)? For example, would it be ethical to rummage through people's trash or watch people in public restrooms? The Internet has posed other issues of privacy. Every day, thousands of people post messages on websites. The messages can potentially be used as data to understand attitudes, disclosure of personal information, and expressions of emotion. Many messages are public postings, much like a letter sent to a newspaper or magazine. But consider websites devoted to psychological and physical problems that people seek out for information and support. Many of these sites require registration to post messages. Consider a researcher interested in using one of these sites for data. What ethical issues arise in this case? Buchanan and Williams (2010) address these and other ethical issues that arise when doing research using the Internet.

INFORMED CONSENT

Recall Principle E of the APA Ethics Code (Respect for People's Rights and Dignity)—research participants are to be treated as autonomous. They are capable of making deliberate decisions about whether to participate in research. The key idea here is informed consent—potential participants in a research project should be provided with all information that might influence their active decision of whether or not to participate in a study. Thus, research participants should be informed about the purposes of the study, the risks and benefits of participation, and their rights to refuse or terminate participation in the study. They can then freely consent or refuse to participate in the research.

Informed Consent Form

Participants are usually provided with some type of informed consent form that contains the information that participants need to make their decision. Most commonly, the form is presented for the participant to read and agree Page 52to. There are numerous examples of informed consent forms available on the Internet. Your college may have developed examples through the research office. A checklist for an informed consent form is provided in Figure 3.1. Note that the checklist addresses both content and format. The content will typically cover (1) the purpose of the research, (2) procedures that will be used including time involved (remember that you do not need to tell participants exactly what is being studied), (3) risks and benefits, (4) any compensation, (5) confidentiality, (6) assurance of voluntary participation and permission to withdraw, and (7) contact information for questions.

FIGURE 3.1

Creating an informed consent form

Page 53The form must be written so that participants understand the information in the form. In some cases, the form was so technical or loaded with legal terminology that it is very unlikely that the participants fully realized what they were signing. In general, consent forms should be written in simple and straightforward language that avoids jargon and technical terminology (generally at a sixth- to eighth-grade reading level; most word processors provide grade-level information with the Grammar Check feature). To make the form easier to understand, it should not be written in the first person. Instead, information should be provided as if the researcher were simply having a conversation with the participant. Thus, the form might say Participation in this study is voluntary. You may decline to participate without penalty, instead of I understand that participation in this study is voluntary. I may decline to participate without penalty. The first statement is providing information to the participant in a straightforward way using the second person (“you”), whereas the second statement has a legalistic tone that may be more difficult to understand. Finally, if participants are non-English speakers, they should receive a translated version of the form.

Autonomy Issues

Informed consent seems simple enough; however, there are important issues to consider. The first concerns lack of autonomy. What happens when the participants lack the ability to make a free and informed decision to voluntarily participate? Special populations such as minors, patients in psychiatric hospitals, or adults with cognitive impairments require special precautions. When minors are asked to participate, for example, a written consent form signed by a parent or guardian is generally required in addition to agreement by the minor; this agreement by a minor is formally called assent. The Society for Research on Child Development has established guidelines for ethical research with children (see http://www.srcd.org/about-us/ethical-standards-research).

Coercion is another threat to autonomy. Any procedure that limits an individual's freedom to consent is potentially coercive. For example, a supervisor who asks employees to fill out a survey during a staff meeting or a professor requiring students to participate in a study in order to pass the course is applying considerable pressure on potential participants. The employees may believe that the supervisor will somehow punish them if they do not participate; they also risk embarrassment if they refuse in front of co-workers. Sometimes benefits are so great that they become coercive. For example, a prisoner may believe that increased privileges or even a favorable parole decision may result from participation. Sometimes even an incentive can be seen as coercive—imagine being offered $1,000 to participate in a study. Researchers must consider these issues and make sure that autonomy is preserved.

Page 54

Withholding Information and Deception

It may have occurred to you that providing all information about the study to participants might be unwise. Providing too much information could potentially invalidate the results of the study; for example, researchers usually will withhold information about the hypothesis of the study or the particular condition an individual is participating in (see Sieber, 1992). It is generally acceptable to withhold information when the information would not affect the decision to participate and when the information will later be provided, usually in a debriefing session when the study is completed. Most people who volunteer for psychology research do not expect full disclosure about the study prior to participation. However, they do expect a thorough debriefing after they have completed the study. Debriefing will be described after we consider the more problematic issue of deception.

It may also have occurred to you that there are research procedures in which informed consent is not necessary or even possible. If you choose to observe the number of same-sex and mixed-sex study groups in your library, you probably do not need to announce your presence and obtain anyone's permission. If you study the content of the self-descriptions that people write for an online dating service, do you need to contact each person to include their information in your study? When planning research, it is important to make sure that you do have good reasons not to obtain informed consent.

In research, deception occurs when there is active misrepresentation of information about the nature of a study. The Milgram experiment illustrates two types of deception. First, as noted earlier, participants were deceived about the purpose of the study. Participants in the Milgram experiment agreed to take part in a study of memory and learning, but they actually took part in a study on obedience. Who could imagine that a memory and learning experiment (that title does sound tame, after all) would involve delivering high-intensity, painful electric shocks to another person? Participants in the Milgram experiment did not know what they were letting themselves in for.

The Milgram study was conducted before informed consent was routine; however, you can imagine that Milgram's consent form would inaccurately have participants agree to be in a memory study. They would also be told that they are free to withdraw from the study at any time. Is it possible that the informed consent procedure would affect the outcome of the study? Knowledge that the research is designed to study obedience would likely alter the behavior of the participants. Few of us like to think of ourselves as obedient, and we would probably go out of our way to prove that we are not. Research indicates that providing informed consent may in fact bias participants’ responses, at least in some research areas. For example, research on stressors such as noise or crowding has shown that a feeling of “control” over a stressor reduces its negative impact. If you know that you can terminate a loud, obnoxious noise, the noise produces less stress than when the noise is uncontrollable. Studies by Gardner (1978) and Dill, Gilden, Hill, and Hanslka (1982) have demonstrated Page 55that informed consent procedures do increase perceptions of control in stress experiments and therefore can affect the conclusions drawn from the research.

It is also possible that the informed consent procedure may bias the sample. In the Milgram experiment, if participants had prior knowledge that they would be asked to give severe shocks to the other person, some might have declined to be in the experiment. Therefore, we might limit our ability to generalize the results only to those “types” who agreed to participate. If this were true, anyone could say that the obedient behavior seen in the Milgram experiment occurred simply because the people who agreed to participate were sadists in the first place!

Second, the Milgram study also illustrates a type of deception in which participants become part of a series of events staged for the purposes of the study. A confederate of the experimenter played the part of another participant in the study; Milgram created a reality for the participant in which obedience to authority could be observed. Such deception has been most common in social psychology research; it is much less frequent in areas of experimental psychology such as human perception, learning, memory, and motor performance. Even in these areas, researchers may use a cover story to make the experiment seem plausible and involving (e.g., telling participants that they are reading actual newspaper stories for a study on readability when the true purpose is to examine memory errors or organizational schemes).

The problem of deception is not limited to laboratory research. Procedures in which observers conceal their purposes, presence, or identity are also deceptive. For example, Humphreys (1970) studied the sexual behavior of men who frequented public restrooms (called tearooms). Humphreys did not directly participate in sexual activities, but he served as a lookout who would warn the others of possible intruders. In addition to observing the activities in the tearoom, Humphreys wrote down license plate numbers of tearoom visitors. Later, he obtained the addresses of the men, disguised himself, and visited their homes to interview them. Humphreys’ procedure is certainly one way of finding out about anonymous sex in public places, but it employs considerable deception.

Is Deception a Major Ethical Problem in Psychological Research?

Many psychologists believe that the problem of deception has been exaggerated (Bröder, 1998; Kimmel, 1998; Korn, 1998; Smith & Richardson, 1985). Bröder argues that the extreme examples of elaborate deception cited by these critics are rare.