Milgram's Study into Obedience

• what was the aim?

  • to test the 'germans are different' hypothesis, claimed that germans are highly obedient and that Hitler could not have exterminated the jewish people/other minority groups in the 1930-40s w/out the unquestionning co-operation of the german pop.
  • to see if individuals would obey the orders of an authority figure that incurred negative consequences and went against one's moral code.
• what was the sample?

  • 40 male volunteers, ages (20-50), recurited through an ad, believeing they were in a memory/learning study at yale. p's became the teacher, while confederate, 'mr wallace' played the learner.
• what was the design?

  • no order effects.
  • lab exp
  • standardised scripts/order of volts.
• what was the method?

  • learner was strapped to electrodes while teacher read word pairs and gave "electric shocks" for wrong answers.
  • shocks ranged from 15-450 volts, increasing w/each mistake. learner (confederate) protested at 150 volts, complained of heart problems at 300 & fell silent after 315 volts.
  • experimenter (in a lab coat), urged p's to continue w/standardised prods like "the exp requires you to continue".
• what were the results?

  • gathered quantitative data.
  • many p's showed stress, but continued when prodded.
  • the shocks appeared real, but no one was actually harmed (acting).
  • exp demonstrated how far people would go in obeying authority, even when it conflicted w/personal morals.
  • main version of exp, obedience rate = 62.5% (25 out of 40).
  • 100% of p's continued up to at least 300 volts (earlier version).
  • caused short-term distress, tweaking, shaking.
• what were the conclusions?

  • revealed the powerful influence of authority on obedience, showing ordinary people could commit harmful acts under instruction.
• what were the strengths?

  • have been reflected in other research on obedience, e.g. Hofling et al (1966), good external validity/replicable.
  • use of 15-volt intervals on the shock generator was effective in showing how destructive obedience doesn't happen immediately, it's a 'drip-drip' effect of the slow erosion of personal values & morals when faced w/prevailing social values.
• what were the weaknesses?

  • lacking internal validity- p's realised that the shocks were fake & were simply playing along.
  • hugely comprised in terms of ethics:

  - p's were deceived as to the true nature of the study.

  - physical & psychological harm inflicted on p's means study can't be replicated today.

  - right to withdraw not tacitly given, experimenter's prods made leaving the study more difficult.
• like w/asch...

  • milgram conducted a no. of diff variations to assess how diff situational factors influence obedience.
• what does data from the study suggest?

  • that people are more likely to obey & cause harm (in the name of obedience) than predicted.
• milgram changed location from yale uni to rundown office block. experimenter also no longer wore a lab coat, instead wearing regular clothing like jeans & a t-shirt. level of obedience?

  • decreased to 50%.
• milgram moved teacher and learner so that they were in the same room and teacher could see learner. level of obedience?

  • decreased to 40%
• milgram asked teachers to force hand of learners on to the electric shock plate in order to receive the punishment. level of obedience?

  • decreased to 30%
• experimenter telephoned the instructions to the teacher and was no longer in the room with them. level of obedience?

  • decreased to 20%
• another teacher (stooge) was bought in to shock the p'. the naive teacher (real p') read out word pairs and stooge teacher shocked the learner. level of obedience?

  • increased to over 90%
• one or two other 'teachers' (stooges) were brought in. at specific points in the study, these 'teachers' refused to continue. level of obedience?

  • decreased to 10%
• variations in milgram's research?

  • setting, uniform, proximity to learner how instructions were given, increase in teachers.
• internal validity?

  extent to which study measures what it's designed to measure.

  • need to consider demand characteristics, i.e. did p's know what was expected of them?
• external validity?

  how well can results be generalised to the pop. ?

  • does the setting represent the real-world?
  • is the sample representative of the whole pop. ?
  • can the results be generalised to today? (temporal)
• internal validity of milgram's study- evaluation:

  • milgram conducted many diff versions of his study, helped to control for some of the confounding variables.
  • e.g. could argue having experimenter in the room would cause the obedience, however, when the experimenter left the room and telephoned the instructions, 20% still went up to 450 volts.
  • people obey due to situational & dispositional factors.
• what does orne (1966) say?- criticism, (internal validity of milgram):

  • says p's were not really fooled. they were just playing along w/the demand characteristics of the situation. (theoretical, he can't prove this)
  • milgram counteracts- "if the p's weren't really fooled, why did they get so stressed? this would suggest they thought the shocks were real." (natural reactions)
  • orne replies- "the stress came from having to play along w/the situation. they still didn't believe they were hurting mr wallace really."
  • milgram responds- "if they didn't believe that mr wallace was really hurt, why did they cheat when the experimenter was absent?"
• criticism of ecological validity of milgram's study (external validity):

  • criticism- milgram's study involved a bizzare task in an artificial situation. people don't really behave that way in real life.
  • response- "but hofling et al. found that nurses would obey an order to hurt a paitent. this shows that authority can make people do bad things (there's other research that supports milgram's findings)."
  • criticism responds- "but the nurses were only doing their job. they thought it was for the paitent's benefit (didn't know they were hurting anyone) and most didn't notice the incorrect dosage."
  • response- "okay, but bickman (1974) showed that just wearing a uniform increases people's obedience. that's what milgram showed"
• criticism of external validity of milgram (androcentric):

  • criticism- milgram only used men in his research. that means that we can't generalise his results to women. we can't wholly trust his results.
  • response- "but milgram did a later study w/female p's and found that rate of obedience was 65% exactly the same as in male samples."
  • criticism- milgram only used 40 females (small sample). sheridan & king (1972) did a similar study where p's shocked puppies and found women to be more obedient (100%) than men (54%).

  - can criticise this- puppies aren't human, women conform more than men (social expectation). no voice line to hear the puppies.

  • response that's a freak finding. the exp has been done many times, and usually m & f don't differ, regardless of culture of p's.