Burgers 2009 Replication of the Milgram Experiment

Stanley Milgram’s groundbreaking psychological experiments on obedience remain famous not just because of what they revealed about human behavior, but also because of how they drew attention to the need for more robust ethical codes in psychological research. Burger (2009) replicates Milgram’s most famous obedience studies in “Replicating Milgram,” tweaking the methodology to ensure ethical treatment of research participants. Therefore, the Burger (2009) experiment uses slightly different experimental approaches. Experimental approaches “have long been a mainstay of the natural science disciplines,” and the “value and applicability of these approaches is relatively new to research examining the psychological, political, and economic dimensions of human life online,” (Oxford Internet Institute, 2017, p. 1). No non-experimental approaches were used in the original Milgram or the Burger (2009) studies.
The Milgram experiments were so groundbreaking as to lead to new theory development in behavioral and social psychology. Milgram worked with the hypothesis that obedience is a factor of authority; people are more likely to obey orders “when that person’s authority is seen as legitimate,” (Burger, 2009, p. 3). In other words, the researchers predicted that the legitimacy of authority determines obedience. Obedience, the dependent variable, was operationalized quantitatively—via measuring the response to the order to deliver an electronic pulse. However, Burger (2009) claims that the primary dependent variable is not obedience itself but “the point in the procedure at which the participant refused to continue,” (p. 1).
Burger (2009) notes that experimenter characteristics may have played a major role in the original Milgram study precisely because an experimenter is viewed as someone with legitimate authority. In fact, the participants might have believed that their obedience would reflect on their performance and behaved accordingly in the study—what is known as demand characteristics. Because many participants would have been familiar with the Milgram study in 2009, the demand characteristics would have been doubly as problematic for Berger. For this reason, Berger (2009) implemented a screening test that eliminated participants with too much familiarity with the original experiment—something that Miller (2009) claims to be inefficient and unnecessary (p. 23). Miller (2009) also criticizes the Burger (2009) study on account of its overly reducing the perceived harm, which impacts the validity and reliability of the results.
Nevertheless, the Burger (2009) study as accurately and ethically as possible recreates the experimental conditions of the original Milgram obedience experiments, using similar independent and dependent variables, changing mainly how they are measured and defined. The primary difference with Burger’s (2009) study is that it attempts to conduct a more ethical experiment while still yielding usable, meaningful results. Independent variables in both the original and the Berger (2009) replications included variations in the experimenter’s behavior and verbalizations. Berger (2009) also notes that situational constraints and situational variables may impact subject responses and performance. Dispositional empathy and desire to control or motivation to control events were the subject variables that Berger (2009) mentions, but not Milgram. Instructional variables were also at play, particularly in the different ways Milgram and Berger would have provided instructions and in Berger’s case too, informed consent. Unlike the original Milgram study, Burger (2009) also informed the participants that they were free to leave at any time and could still retain their incentive pay.
Neither the original Milgram experiments nor the Berger (2009) replications used true random selection of participants. In Berger’s (2009) case, participants self-selected by responding to an advertisement. However, both the original and the Berger (2009) experiment did rely on random assignment to bolster reliability and validity. Both the original and the Berger (2009) replications involved between-subjects designs. One of the noted problems with the Milgram original experiments was obtrusive approaches, which Berger (2009) does not significantly attempt to correct for in his modernized, updated version of the obedience experiments. In fact, the role of the experimenter is integral to the research design, as the experimenter is positioned as an authority figure. As Miller (2009) points out, previous attempts to replicate the Milgram experiments substantially changed the research design by artificially attempting to improve internal validity. For example, some researchers introduced “mediated violence” instead of the more direct violence being assessed in the original experiment (p. 22). Burger (2009) avoids using the mediated violence method.
Burger (2009) aims for maximum experimental realism and mundane realism, but there are naturally limitations to the experimental conditions. Mundane realism refers to how well the experiment replicates real life scenario. Mundane realism improves the external validity of the study. Although neither the original nor the Berger (2009) experiment have full mundane realism, the act of shocking a person can be reasonably extrapolated to other acts of violence. In fact, one of the main principles of the Milgram obedience studies is how participants respond to incremental increases in voltage. To ask someone to immediately administer 450 volts would be different from asking someone to start with only 10 volts and then gradually increase the voltage. The main difference in Burger’s (2009) approach from the original experiment was to reduce the shock level from 450 volts used by Milgram to a much safer 150 volts. Burger (2009) still defends the relevance of the research in spite of its relatively low mundane realism, as most people will not be asked in real life to shock someone. “Although one must be cautious when making the leap from laboratory studies to complex social behaviors such as genocide, understanding the social psychological factors that contribute to people acting in unexpected and unsettling ways is important,” (Burger, 2009, p. 10).
Using a cost-benefit analysis, the study has a tremendous bearing on understanding human social behavior. Conformity, obedience, and the willingness to commit violent acts towards others have implications for public safety, legislation, and social controls. Therefore, the minimal costs of conducting the experiment are outweighed by the potential benefits to society. “In general the experimental method is widely accepted as offering exciting potential to make causal inference and even to feed in to policy-making and practice,” (Oxford Internet Institute, 2017, p. 1). The cost-benefit analysis of the Burger (2009) research is a vast improvement over the original Milgram designs because of the way the latter suffered from what are now considered ethical violations. Although Burger (2009) uses a reasonable degree of deception in order to receive accurate results, informed consent was given, confidentiality assured, and institutional approval from the IRB was duly received. Similarly, the Burger (2009) study conforms to the five main ethical tenets of the American Psychological Association: Beneficence/Nonmaleficence, Fidelity/Responsibility, Integrity, Justice, and Respect for the Rights of Others. For example, the Burger (2009) study ensured that no shock over 150 volts would be permissible given the principles of nonmaleficence and respect for the rights of others.
Experimental realism improves internal validity, and refers to how well the experimenter designed the setting and the study overall to accurately measure the variables. The Burger (2009) research has a strong degree of experimental realism overall. Miller (2009) claims that the experimental realism is “both imaginative and convincing” in spite of needing to lower the maximum voltage for ethical purposes (p. 23). Burger (2009) also does a good job of explaining additional issues at stake in obedience, such as whether participants can see their peers. When peers stop administering the shocks, the other participants are more likely to stop administering the shocks too. Obedience is therefore related not just to authority but also to conformity to social norms. Interestingly, no gender differences were found and Burger (2009) implies that lower rates of assertiveness towards authority figures among female populations would override potentially greater measures of empathy or compassion.
In spite of having to tweak the Milgram experiments significantly, the Burger (2009) research suffers from few methodological flaws. The researcher retained the original hypotheses and overall research design and experimental approaches used by Milgram. Miller (2009) claims that the Burger (2009) study is problematic because he overcompensates for the ethical violations known in the original experiment, shielding participants from stress. By overprotecting the participants, the results of the Burger (2009) study may be less reliable than those obtained by Milgram. Yet the results of Burger’s (2009) study remarkably replicate those findings. The Berger (2009) study therefore continues to have relevance for future social psychological research. Applicable to the real world, the obedience experiment can and should be integrated into public policy whenever possible. As Miller (2009) points out, even the original Milgram “research had stimulated so much ethical and methodological controversy rather than substantive research on obedience itself,” (p. 26). New, more ethically designed, research is “sorely needed” to help create a normative culture whereby empathy and ahimsa always override the desire to obey authority figures or to conform to peer expectations.






References

Burger, J.M. (2009). Replicating Milgram. American Psychologist 64(1): 1-11.
“Ethical Issues in Psychology,” (n.d.). Retrieved online: http://psychyogi.org/ethical-issues-in-psychology/
“Experimental Realism,” (n.d.). Psychology. Retrieved online: https://psychology.iresearchnet.com/social-psychology/social-psychology-experiments/experimental-realism/
Miller, A. G. (2009). Reflections on “Replicating Milgram.” American Psychologist 64(1): 20-27.
Oxford Internet Institute (2017). Experimental approaches. Retrieved online: https://www.oii.ox.ac.uk/study/courses/experimental-approaches/