One limitation of Asch's research is that the task and situation were artificial
Participants knew they were in a research study and may simply have gone along with what was expected (demand characteristics)
The task of identifying lines was relatively trivial and therefore there was really no reason not to conform
According to Susan Fiske (2014), Asch's groups were not very groupy, ie they did not really resemble groups that we experience in everyday life
This means the findings do not generalise to real-world situations, especially those where the consequences of conformity might be important
Another limitation is that Asch's participants were American men
Other research suggests that women may be more conformist, possibly because they are concerned about social relationships and being accepted (Neto 1995)
Furthermore, the USA is an individualist culture (ie. where people are more concerned about themselves rather than their social group)
Similar conformity studies conducted in collectivist cultures (such as China where the social group is more important than the individual) have found that conformity rates are higher (Bond and Smith 1996)
This means that Asch's findings tell us little about conformity in women and people from some cultures
One strength of Asch's research is support from other studies for the effects of task difficulty
For example, Todd Lucas et al. (2006) asked their participants to solve 'easy' and 'hard' maths problems. Participants were given answers from three other students (not actually real). The participants conformed more often (ie agreed with the wrong answers) when the problems were harder
This shows Asch was correct in claiming that task difficulty is one variable that affects conformity
However, Lucas et al's study found that conformity is more complex than Asch suggested. Participants with high confidence in their maths abilities conformed less on hard tasks than those with low confidence
This shows that an individual-level factor can influence conformity by interacting with situational variables (eg task difficulty). But Asch did not research the roles of individual factors
Conformity
Assuming that the majority opinion or behaviour is correct and changing one's own opinion or behaviour to match the group
Types of conformity
Internalisation (genuine acceptance of group norms), Identification (conforming to be accepted by the group), Compliance (going along with the group in public but not privately changing opinions/behaviour)
Informational social influence (ISI)
Conforming because the group is perceived to have better information
Normative social influence (NSI)
Conforming to gain social approval and avoid rejection
Evidence supports NSI as an explanation of conformity. For example, when Asch's participants wrote their answers down privately, conformity fell to 12.5% because there was no normative group pressure
Evidence also supports ISI as an explanation. The Lucas et al. study found more conformity on difficult maths problems because participants relied on the group's answers when they were uncertain
However, it is often unclear whether NSI or ISI is at work in research studies, as both processes probably operate together in most real-world conformity situations
One limitation is that NSI does not predict conformity in every case. Some people are more concerned with being liked by others (nAffiliators) and are more likely to conform
This shows there are individual differences in conformity that cannot be fully explained by one general theory of situational pressures
In the Stanford Prison Experiment, the guards took up their roles with enthusiasm, treating the prisoners harshly. The prisoners became subdued, depressed and anxious
Social roles appear to have a strong influence on individuals' behaviour. The guards became brutal and the prisoners became submissive
Such roles were very easily taken on by all participants, even those who came in to perform specific functions
One strength of the Stanford Prison Experiment is that the researchers had control over key variables, such as randomly assigning participants to roles
This increased the internal validity of the study, allowing stronger conclusions about the influence of roles on conformity
One limitation is that the experiment lacked realism, as participants may have been merely play-acting based on stereotypes rather than genuinely conforming to roles
However, evidence suggests the participants did behave as if the prison was real to them, giving the study a high degree of internal validity