Part 2 - Variables affecting conformity: group size...Asch

• Asch's Baseline Procedure (1951): Aim - To investigate if participants would yield (conform) to majority influence. Participants - 123 male American undergraduates recruited using a volunteer sampling technique groups of 6 consisting of 1 true participant, 5 confederates (actors/people involved)...
• ...Procedure - Participants and confederates were presented with 4 lines; 3 comparison, 1 'standard' line. Asked to state which of 3 lines was the same length as a stimulus line. The genuine participant always answered last or second to last. Confederates would give same incorrect answer for 12/18 'critical' trials. Asch observed how often the participant would conform to confederates vs providing the correct answer...
• ...Findings - Naive participant gave a wrong answer on 36.8% of trials, 25% never conformed, 75% conformed at least once. In a control trial, only 1% of responses given by participants were incorrect (which eliminates eyesight/perception as an extraneous variable, thus increasing the validity of the conclusions drawn). Conclusion - Participants exhibited a 'distortion of action' - they knew what was the right answer but conformed to the majority to avoid ridicule. (compliance and NSI are present).
• Evaluation of Asch's study: High control - Uses the experimental method, thus there was strict control over extraneous variables such as timing of assessment and the type of task used. Participants did the experiment before without confederates to see if they actually answer correctly thus removing the confounding variable of a lack of knowledge...
• ...Also when using a lab experiment, researcher can be more confident that it is the IV (confederates answers) that is causing the DV (participants' answers). Confident, results about conformity aren't being affected and thus have internal validity suggesting valid and reliable 'cause and effect' relationships can be established meaning replication is easy, increasing reliability of findings as it reduces likelihood of anomalies.
• Ethical issues - The researchers breached the BPS ethical guideline of deception and consequently, the ability to give informed consent. However, the participants were debriefed. Furthermore, it can be argued that there was deception as participants were unaware of confederates involved thus could not provide full informed consent but psychologist would contradict this by saying the deception was necessary...
• ...because if participants knew of true aim they may have shown demand characteristics thus deception helps to increase validity of findings about conformity. Ethical issues do not threaten the validity or reliability of findings, but rather suggest a cost-benefit analysis is required.
• Supports NSI - Participants reported that the conformed to fit in with group and to avoid rejection, so it supports the idea of NSI, which states that people conform to fit in when privately disagreeing with the majority.
• Low Population validity - For example, Asch's participant's were American male students. The study could be criticised for being gender bias and not representative of female behaviour in situations of conformity. Other research suggests that women may be more conformist because they are concerned about social relationships and being accepted (Neto, 1955). Limitation as results about conformity may not be able to be generalised to the behaviours of others females (beta bias) or people that aren't student (low external validity).
• Low ecological validity - Participants knew they were in a research study and may have simply gone along with what was expected (demand characteristics). Artificial situation and task as relatively trivial and there was no reason not to conform. Findings cannot be generalised to real life as it does not reflect the complexity...
• ...Additionally, according to Fiske (2014), 'Asch's groups were not very groupy', did not resemble groups in everyday life. Limitation as difficult to generalise such as making decisions amongst people they know (low external validity) or in situations where consequences of conformity might be important.
• Historical bias - The social context of the 1950s may have affected results. For example, Perrin and Spencer (1980) criticised the study by stating that the period that the experiment was conducted in, influenced the results because it was an anti-Communist period in America when people were more scared to be different e.g. McCarthyism. Thus the study can be said to lack temporal validity because the findings cannot be generalised across all time periods.
• Cultural bias - Smith et al. (2006) analysed results of Asch-type studies across different cultures. The average conformity rate across them was 31.2% . Average for individualistic cultures (e.g. Europe and US) was 25, whereas for collectivist cultures (e.g. Africa, Asia, S. America) it was much higher at 37%. Markus (1991) suggests the higher conformity rate in collectivist cultures is because it is viewed more favourably and is a form of 'social glue' that binds communities together. Limitation as unable to generalise to other cultures other than the US (low external validity).
• Independent behaviour rather than conformity - For example, in Asch's study, only 1/3 of trials where the majority unanimously gave the same wrong answer produced a conforming response. Meaning that in 2/3 of these trials, participants resolutely stuck to their original judgement despite being faced with a majority expressing a different view. Limitation as Asch's research actually demonstrates a commendable tendency for participants to stick to what they believed to be correct behaviour.
• Factors affecting level of conformity: Asch extended his baseline study to investigate the variables that might lead to an increase or decrease in conformity.
• Group size: Asch increased size by adding more confederates so larger majority. Thus a growing pressure to conform -NSI, not only to a pint levelling off when the majority was greater than three. With 3 confederates, conformity to the wrong answer was 31.8% compared to 36.8% in Asch's original study where he used 6 confederates, but with the presence of more confederates this made little difference. This suggests that most people are very sensitive to the views of others as just 1 or 2 confederates was enough to sway opinion.
• Unanimity: The extent to which all the members of the group agree e.g. all give the same answer. In Asch's variations, he introduced a confederate who disagreed with other confederates. the presence of a dissenter. The rate decreased to less than a quarter of the level it was when the majority was unanimous. With the presence of a dissenter or disaffected confederate who gave the correct answer, conformity fell from 32% to 5.5%. If different answers are given, it falls from 32% to 9%...
• ...The more unanimous the group is, the more confidence the participant will have that they are all correct, and therefore the participant’s answer is more likely to be incorrect. Unanimity is vital in establishing a consistent majority view, which is particularly important by providing normative social influence through preventing any conflicting views arising causing unanimity to be disturbed.
• Task Difficulty: An individual is more likely to conform when the task is difficult. For example, Asch altered the (comparison) lines (e.g. A, B, C) making them more similar in length. Since it was harder to judge the correct answer, conformity increased. When the task is difficult, we are more uncertain of our answer so we look to others for confirmation...
• ...The more difficult the task the greater the conformity. This suggests that informational social influence is a major mechanism for conformity when the situation is ambiguous and the individual does not have enough of their own knowledge or information to make an informed decision independently, and so has to look towards others.
• (+) High Control - A strength of Asch's research into variables affecting conformity is that he uses the experimental method, so therefore has high control over extraneous variables and can establish cause and effect...
• ...When using a lab experiment the researcher can be more confident that it is the IV (task difficulty/group size/degree of unanimity) that is causing the DV (participants' answers). We can be confident that results are not being affected by confounding variables and thus have internal validity which strengthens the validity of the claim that task difficulty/group size/unanimity does affect conformity.
• (+) Research Support - One strength of Asch's research is support from other studies for the effects of task difficulty. For example, Lucas et al. (2006) observed that participants with high confidence in their Maths abilities conformed less than those with low. Thus shows Asch was correct in claiming that task difficulty is one variable that affects conformity. Also Ethical issues and supports NSI. However, research has found conformity to be more complex than Asch suggested as an individual-level factor can influence conformity by interacting with situational variables. Limited in EI, HB.
• (-) Low population validity - Limitation of Asch's research is that it is androcentric (a form of gender bias). The focus is on males as sample consists of only male participants and so findings only reflect male behaviour. Asch's research also has beta bias (when studies/theories minimise differences between genders) as he attempts to generalise findings to both males and females (universality) thus does not provide a valid explanation of female behaviour in the context of conformity...
• ...Asch's study is an example of nomothetic research as it attempts to establish universal laws about conformity which can be generalised to any given population. However it does tend to be more scientific than ideographic e.g. testing behaviour using lab experiments thus giving psychology greater credibility as meaningful conduction.
• (-) Artificial situation and task - low ecological validity
• Ethical issues: There was deception as participants were tricked into thinking the study was about perception rather than compliance so they could not give informed consent. There could have been psychological harm as the participants could have been embarrassed after realising the true aims of the study. Such issues simply mean that a cost-benefit analysis is required to evaluate whether the ethical costs are smaller than the benefits of increased knowledge of the field. They do not affect the validity or reliability of findings. Also, this was culturally biased.