Evaluation 2

• Valid consent

  Participants had consented to a psychology experiment but one they believed was about memory and learning, so no valid consent
• Deception

  There were many ways that Milgram deceived participants, but arguably to reduce demand characteristics
• Risk of harm


  The responses recorded by Milgram show that participants clearly suffered as a consequence of the study, and there is a debate about whether they genuinely had a right to withdraw
• Confidentiality


  Although no names were published, photographs and film footage of the study were published
• Privacy


  Many participants showed they had the potential to kill another, and this is something they may not have wanted revealed about themselves – even to themselves
• Long term risks to beliefs, values, etc.


  Most participants reported they were glad to have taken part and some experienced growth as a result, but does this justify the potential long term negative effects?
• Internal validity


  The observation was controlled but there was no actual independent variable so no cause and effect was established
• Population validity


  The sample included a wide range of occupations and ages but was gender and culturally biased making some generalisations difficult
• Ecological validity


  The lab setting, the nature of the authority figure and the task itself were all unrealistic and unlikely to occur in real-life
• Construct validity


  Even if we argue following orders to harm someone is realistic, this is still a narrow measure of what obedience truly is
• Temporal validity

  Levels of obedience are subject to change as society changes so Milgram's findings are likely to be out-of-date
• Demand characteristics


  There is a debate about whether participants realised the experiment was a set up. Some critics argue it was obviously fake that participants were simply playing along having guessed the aim. Later interviews with participants suggested this was not the case. Indeed, their anguished responses to having to shock someone seemed very genuine at the time
• Social desirability


  It seems unlikely that participants shocked another person to appear to be following accepted norms unless we argue that they wanted to do the right thing in front the experimenter and this was who they were concerned about it – however, this would be more akin to conformity than obedience then
• Researcher bias


  Milgram's quantitative measures of obedience were completely objective but it qualitative data less so