This paper examines Stanley Milgram's landmark 1963 obedience experiment, tracing its origins to the Nuremberg trials and analyzing its experimental design, methodology, and surprising findings — 65% of participants administered maximum-level shocks. The paper evaluates the study's methodological limitations by contemporary standards, discusses the ethical concerns it raised regarding informed consent, participant welfare, and deception, and explains how it contributed to the formation of Institutional Review Boards and stricter American Psychological Association guidelines. Solomon Asch's conformity experiments are also discussed as a parallel case illustrating the necessity of deception in social psychology research.
The paper uses a compare-and-contrast strategy alongside a then-versus-now evaluative framework. By measuring Milgram's methods against modern standards — randomized controls, minimal risk, informed consent — the writer demonstrates how to apply current disciplinary criteria to historical research without dismissing its contributions. This technique is particularly effective in psychology and social science essays.
The paper opens with historical motivation (Nuremberg trials), then moves through experimental design, findings, and methodological critique before pivoting to ethics. The Asch discussion serves as a supporting case study for the deception argument. The conclusion ties together the methodological and ethical threads, affirming the study's enduring relevance. This funnel structure — broad context to specific critique to broader legacy — is well-suited for review essays in the social sciences.
The classic Milgram studies on obedience were inspired by the Nuremberg trials of Nazi war criminals, who consistently argued in their defense that they were simply carrying out orders. In his original study, Stanley Milgram (1963) wanted to determine whether people would inflict pain to the point of serious injury or death as the result of being ordered to do so by an authority figure. Milgram used a sham learning experiment and a confederate learner to test his hypothesis that few people would actually progress to the point of inflicting damage on strangers at the behest of an authority figure. As we will see, his original hypothesis proved incorrect.
The learning experiment required a "learner" (the confederate) to memorize a list of word pairs. The "teachers" — recruited participants — were required to administer what they believed to be painful electrical shocks to the learner whenever the learner failed to recall a word pair. The shocks were also a sham. The independent variable in this study is the authority figure and the prodding used to get the teacher to administer a shock to the learner. The dependent variable is the level of shock the teacher would actually administer.
The authority figure — Milgram, dressed in a white lab coat — met with both the teacher and the learner before the start of the experiment. Unbeknownst to the teacher (a participant recruited through a local newspaper advertisement offering $4.00 to participate in a "learning experiment"), the learner was always the same person: a confederate who was part of the experiment. Milgram, posing as the experimenter in a white laboratory jacket, had the two individuals draw lots to determine who would be the teacher and who would be the learner. This process was, of course, fixed as described above.
Once the experiment started, the learner was taken to a separate room and the teacher could not see him. They communicated via an intercom. The procedure was relatively straightforward: the teacher would read a list of word pairs to the learner. The teacher would then go back and repeat the first word of each pair, and the learner was supposed to reply with the second word. The teacher stood or sat before a panel of knobs labeled according to shock intensity, ranging from 15 to 450 volts in 15-volt increments. If the learner gave an incorrect response, the teacher was instructed to administer a shock, starting at 15 volts and increasing by one level with each missed word.
The shocks were a sham, and the learner followed the same scripted responses for each participant: initially feigning being shocked with a grunt, later claiming to have a bad heart and wanting to stop, and eventually going silent as if he had passed out.
When teachers became resistant to shocking the learners, Milgram — the authority figure ordering the shocks, operating under the auspices of a research university — used up to four verbal prods to encourage them to continue. The experiment was terminated either after the fourth prod or after the participant had administered the 450-volt shock three times. Milgram had pre-tested the experiment by explaining the proposed procedure to university students, who predicted that fewer than 10% of participants would follow through to the ceiling level of 450 volts. Milgram himself predicted an even lower rate. However, during the actual experiment, 65% of subjects administered the maximum shock, despite the learner's protests, grunts of pain, complaints of a bad heart, and eventual silence at high shock levels. Milgram subsequently conducted several variations of his original experiment (see Brown, 1986, for a thorough review).
The experiment was demonstrated to be quite reliable by follow-up studies. Milgram varied the physical distance between teacher and learner — an additional independent variable that effectively made the study a factorial design — and found that the tendency to obey authority decreased the closer and more visible the learner was to the teacher. The initial study was considered so valid in its findings that it helped stimulate the formation of Institutional Review Boards in research universities and hospitals.
The study was also conducted before the widespread adoption of randomized controlled designs. By today's standards, the control condition would be considered inadequate: polling students about how they would react in such an experiment does not constitute a true control. The lack of randomized assignment also raises questions about the causal inferences the study made. For example, including a control condition without an authority figure present — and randomly assigning participants to either the control or experimental condition — would allow for a more appropriate inference that it was the presence of authority that caused teachers to continue administering shocks. Additionally, the convenience sampling and recruitment method used in this study raise concerns about the generalizability of the findings.
You’re 48% through this paper. Sign up to read the remaining 3 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.