To investigate the extent to which social pressure from a majority group could affect a person to conform
Procedure - Asch’s study
Tested conformity
123 American male undergraduates
5 confederates and 1 naïve ‘real’ participant
Participants looked at 3 lines of different lengths
They took turns to call out which of the 3 lines was the same as the ‘standard’ line
Each participant took part in 18 trials
12/18 trials were critical (meaning confederates were instructed to give the same incorrect answer)
Findings
On the 12 critical trials, the average conformity rate was 33% (i.e. participants agreed with the incorrect response given by other group members on average in 1/3 of the trials)
25% did not conform on any trials therefore 75% conformed at least once
Half of participants conformed on 6 or more of the critical trials.
1 in 20 conformed on all of the critical trials.
Findings
A control condition was used where confederates did not give the wrong answer. In this condition, participants made mistakes 1% of the time.
Those that conformed did so even when the situation was unambiguous (The Asch effect)
In post interviews with the participants most said they conformed to avoid disapproval from other group members, even though they knew the confederates were wrong.
The type of conformity shown here is COMPLIANCE. The explanation of conformity shown here is NORMATIVE SOCIAL INFLUENCE.
Group size - Very little conformity with 1 or 2 confederates. 3 confederates needed for a conformity rates of approx. 30%. More than 3 confederates showed no further rise
The unanimity of the majority
In Asch’s original study, confederates unanimously gave the same wrong answer. However, when the real participant was given the support of a confederate who gave the right answer, conformity levels dropped significantly. In this case the percentage of wrong answers given by the real participant dropped from 33% to just 5.5%.
The difficulty of the task
When Asch made the difference between the line lengths much smaller, conformity increased. Informational Social Influence may be at play here, as we look to others for guidance when the situation is unclear (ambiguous). Lucas et al. (2006) found that, when exposed to maths problems in an Asch style experiment, high self-efficacy (self-belief) individuals remained more independent and were less likely to conform than low self-efficacy individuals.
A limitation of Asch’s research into conformity is that not all participants conformed when the majority unanimously gave the wrong answer
In two-thirds of the trials, participants kept their original answer, despite a large majority expressing a completely different view
Asch argued this indicated a tendency for participants to show independent behaviour rather than simply conform
This suggests that majority influence is not as strong as it might seem as a higher proportion of trials produced an independent response rather than conforming to the majority
Another limitation of Asch’s experiment is that it may not be true today
The research took place in the 1950s, at a time when conformity was high. The USA was affected by McCarthyism at the time, so people were scared to go against the majority
Perrin and Spencer (1980) replicated Asch’s study in the 1980s and only had one conforming response in 386 trials
This suggests that conformity levels change over time and that Asch’s research could be regarded as a ‘childofitstime’ rather than a universal phenomenon
One limitation of Asch’s research is that there are ethical issues associated with it.
Naive participants were deceived. They thought the others in the procedure (confederates) were genuine
But this ethical cost should be weighed against the benefits of the study
The main benefit was highlighting people’s susceptibility to group conformity and the variables affecting it.
A limitation of Asch’s research is that the task and situation were artificial
Participants knew they were in a study so may have just responded to demand characteristics
The line task was trivial so there was no reason not to conform. Also the naive participants were in a ‘group’, but not like groups found in everyday life
Finding therefore do not generalise to everyday situations where consequences of conformity are important, and where we interact with groups more directly