Milgram

• Milgram (1963)

  Sample:

  • 40 volunteers (men age 20-50)
  • For human memory study
  • Offered $4.50 (now £25)
  • Diverse occupations, educational level, and age
  • Volunteer sample

  Procedure:

  • Participants met Mr Wallace in rigged role allocation
  • Participant = “teacher”
  • Mr Wallace = “learner”
  • Separate rooms, Wallace heard, not seen
  • Naïve participant given initial shock (45v) at beginning
  • Shocks went up 15V up to 450V (‘XXX’)
  • Mr Wallace script
  • ~3 wrong answers to each correct
  • 180V; Complained of weak heart
  • 300V; Banged on wall
  • 315V; Silent
  • If ‘teacher’ objected, researcher gave 4 prods
• Milgram's Aim

  • Investigate obedience in situation where following orders would break personal moral codes
  • Test hypothesis “Germans were different” (from Holocaust)
• Results

  • Experts predicted
  • <1% go to max
  • 2.6% go to 240V
  • 65% went to max (450V)
  • 100% went to 300V
  • Participants showed signs of extreme stress (shook, sweated and stuttered)
  • 14/40 had nervous laughing fits and argued with researcher (obeyed anyway)
  • 35% resisted pressure (stopped 300-450V)
  • Most people default to truth, assume they can trust
  • Doubters:
  • 56% fully believed
  • 24% had some doubts, still believed
  • 6% unsure
  • 11% had some doubts, didn’t believe
  • 2% sure untrue
• Conclusion

  • Milgram concluded social situation is a determinant of behaviour
  • As we are socialised from a young age to recognise authority and obey those with perceived power
• Evaluation

  Strengths:

  • Generalisability:
  • All US males
  • But had wide range of backgrounds
  • Found similar results with females and cross-culturally
  • Blass
  • Experimental validity:
  • 45V shock given to naïve participants to increase realism
  • Equipment and screams were realistic
  • Participants were in clear distress - implying belief

  Weaknesses:

  • Ecological validity:
  • Giving electrical shocks to strangers is unlikely in real life
  • High internal but low external value
  • Population validity:
  • Volunteer sample used
  • Participants may have been more compliant/authoritarian
• Ethics

  • Withdrawal
  • Difficult, replicated real-life
  • Disobedience possible; 35%
  • Protection from harm
  • Argued distress was more 'momentary excitement' than harm (cost-benefit analysis)
  • Maintained people would not have cared for ethical issues had there been lower obedience levels
  • Obedient participants told behaviour normal
  • Disobedient participants told behaviour was desirable
  • Questionnaire:
  • 92% responded
  • 84% glad to have done it
  • 2% regretted
  • 74% learned personal importance
  • Independent psychiatrist assessed psychological damage a year post-experiment (found nothing)
• Gina Perry(2012)

  • Claims Milgram manipulated research, onle one variation had 65% obedience (in over half his variations 60% disobeyed)
  • Methodological problems:
  • Improvisation
  • 'Researcher’ deviated from his script, with Milgram's approval
  • Would leave lab to check on learner and assure participant he was fine
  • Command 25 times or more to continue
  • Teachers blocked to swap or check themselves