Moscovici et al (1969)

• Aim 

  • They aimed to investigate the influence of a minority upon a majority within a group
  • They investigated whether the consistency of the behaviour of a minority is a powerful source of influence
• Sample 

  •  liberal arts, law and social science students (172 female americans)
  •  Female participants preferred ⇒ involvement in evaluating the colour of an object

  Decieved: told that this would be an experiment on colour perception (an explanation of the meaning of ‘light intensity’ was given
• Each experimental group: four participants and two confederates
• Took a polack test: elemination of participants with visual abnormalities / to emphasise to everyone that the group had normal vision.
• Procedure 

  • participants were shown 36 slides that were different shades of blue and asked to state what colour the slide was. 
  • 2 individuals were confederates (minority) while 4 individuals were actual participants (majority) in each condition. 
  • In the consistent condition confederates answered green for all 36 slides (1).
  • In the inconsistent condition confederates answered green 24 times and blue 12 times (1).
  • Moscovici et al. (1969) used a control condition which involved no confederates, just six participants.
• Results 

  • ⇒ 8.42% of responses in the consistent minority condition were green (1)  
  • ⇒ Only 1.25% of responses in the inconsistent minority condition were green (1)
  • Participants were more likely to give similar responses to the confederates when light intensities were weak than when they were strong.
• Validity 

  • Screened participants using polack ⇒ participant variables controlled ⇒ high internal validity 
  • Control group indicates that manipulation of IV was what resulted in minority influence 
  • Lacks ecological validity ⇒ lab experiment ⇒ real life pressure group scenario 
  • Lacks population validity ⇒ only used females ⇒ not generalisable to males
• Reliability 

  • Controlled variables such as light intensity ⇒ test-retest reliability can be measured 
  • Test-retest reliability ⇒ supporting evidence from Wood et al. (1994) whose meta-analysis with 97 studies ⇒ found that when minority is consistent ⇒ more likey to influence
  • Random allocation of participants ⇒ no experimenter bias ⇒ results reliable to understand minority influence 
  • Participants were split into groups of 6 ⇒ inter-rater reliability of similar results in minority influence