ayush Units 7-12 Psych

Asch ran a range of laboratory experiments to test majority influence on conformity and to investigate the factors that affect conformity levels. Basic set-up: a naive participant is placed in a group of confederates (people secretly working with the researcher) to see whether the naive participant conforms to the group norm.

50 male college students (American, tested in the 1950s) were put into groups of eight. In each group one person was the naive participant and the other seven were confederates.

Each group member had to match the length of a standard line with one of three comparison lines (one matched the standard line, two clearly differed in length). Each person judged publicly, aloud, in front of the group. Unknown to the naive participant, the seven confederates unanimously gave the same wrong answer, so the participant faced a group that all chose an incorrect line.

Conformity was measured as the frequency of incorrect responses made by the naive participant (errors that matched the majority's wrong answer) over the 12 trials in which the group gave an incorrect answer. Straight after the trials the participant was interviewed and debriefed about the true nature of the study.

50 male US college students were tested in groups of eight (seven confederates plus one naive participant); each publicly judged which of three comparison lines matched a standard line; on 12 trials the confederates unanimously gave the same wrong answer; the naive participant's errors were counted as conformity; each participant was then interviewed and debriefed.

Over the 12 trials where the group gave an incorrect answer, 32% of estimates (responses) made by the naive participants were errors in line with the majority's wrong answer; 68% of estimates were correct. There were clear individual differences: some participants conformed on almost every trial and some never conformed.

32% of RESPONSES (estimates) given were errors. It is not 32% of participants and not 32% of trials. (The book's checkpoint asks exactly this: the correct option is 'responses given'.)

Observations suggested that those who stayed independent were confident, maintained individuality or maintained task focus. Those who conformed were thought to lack confidence in their own perception or to want to fit into the group. (These are Asch's observations/interpretations, not directly measured variables.)

Conformity could only be explained by a combination of situational pressure and the individual's response in that situation. Majority influence produced conformity when the majority was unbrokennanimous and at an optimal majority size of three, but individual differences between participants were also important in predicting conformity.

Two variations tested a broken majority: (1) two naive participants in the group; (2) one naive participant plus one confederate who gave the correct answer alongside six confederates giving the wrong answer. Error rates fell to 10.4% and 5.5% respectively. The reduced majority lowered conformity, perhaps because the partner acted as social support against the majority influence.

Losing a partner midway made the participant lose independence: the loss of support caused them to conform. The reverse (a partner who was at first conforming and then broke away to give correct answers) also reduced conformity, but the effect was smaller because participants found it difficult to switch from conforming to independent behaviour.

Majorities of 1, 2, 3, 4, 8 and 16 were tested, all unanimous and all giving a wrong answer. One confederate had little real influence; with two confederates the naive participant gave the wrong line 12.8% of the time; the greatest error rate was with three confederates; larger majorities did not produce any greater effect than three. Conclusion: optimal majority size of three (the book also says a majority of three to four achieved conformity).

Male college students in groups of seven to nine gathered in a classroom. The confederate majority were chosen as acquaintances. Procedure followed the original: the majority gave the correct line on the first two trials; on the third trial, where the majority gave a wrong answer, one individual (a dissenter) in the group said the majority's answer was wrong. From time to time this individual repeated his dissent and sometimes agreed with the incorrect response, so the naive participant faced majority pressure plus someone who disagreed with the majority.

One third (33.2%) of the naive participants' responses were incorrect and consistent with the majority's error - a similar rate to 1951. A control group who wrote their responses down privately made errors on 7.4% of responses. Because conformity stayed high even though the dissenter questioned the majority's judgement, participants were not simply conforming because they believed the majority was correct; they were yielding to social pressure. There was again significant individual variation.

(1) Majority chose a line that was very obviously different from the standard: error rate fell slightly to 27.2% but the majority influence was not drastically reduced. (2) Majority reduced to one single person: any effect on decision-making was completely abolished and the naive participant ended up challenging the single person. (3) A single group member consistently gave the correct judgement against the majority: errors fell to 13%, with few participants giving multiple wrong judgements; those who erred did so by abandoning the partner for the majority. (4) Majority comprised 16 naive participants and a single person responding incorrectly: the single person (a minority) had no impact, was mocked by the naive majority and suspected of deliberately giving the wrong answer.

That we are still likely to yield to majority influence even when it is questioned and when it is obviously wrong.

Aim: to test whether the majority could command compliance even if their judgement was obviously incorrect. Sample: 123 college students in groups of seven to nine. One naive participant, seated in the second-to-last chair, heard other group members give correct or incorrect line judgements. On 12 trials the majority chose an incorrect line that was either moderately different or extremely different from the standard line. The naive participants' public judgements were compared with a control group of naive participants who wrote their judgements down privately.

The control group of 37 naive participants gave only three incorrect line judgements (as printed in the book), compared with one third incorrect responses in the experimental group. When the majority chose a moderately different wrong line, naive participants also made a moderate selection; when the majority gave an extreme judgement, four-fifths modelled the extreme judgement and one-fifth compromised by choosing the moderately different line. Considerable individual variation again (some fully independent, some yielding completely).

To test whether conformity extended to other stimuli, participants judged whether a comparison disc was brighter or darker than a standard disc. Over 12 trials where the group deliberately gave wrong judgements, 39% of the naive participants' responses were incorrect. In a variation where the group gave public judgements but the naive participant responded privately, conformity fell to 12.5% incorrect estimates.

The experiments show the power of the majority over a minority; participants were complying because of normative social influence (the need to fit in and gain approval). Situational factors such as having social support and the size of the majority affect conformity, but individual differences between people also affect how they respond.

1951: 50 male students, unanimous majority of seven, 32% errors, variations on broken majority and size of majority. 1952: confederates were acquaintances and one dissenter questioned the majority; 33.2% errors; private-response control group at 7.4%; variations including a single partner giving correct answers (13%). 1956: 123 students; moderately vs extremely wrong majority; control group of 37 writing privately; also tested coloured discs (39%) and public group/private participant response (12.5%). All series found considerable individual variation.

Point: Asch's findings may not generalise. Evidence: he tested American, male, university students in the 1950s (50 in 1951), and later studies with different populations and in different decades have not produced the same conformity rates as Asch. Explain: the sample excludes women, other age groups and cultures, so the 32% figure may reflect that group and era rather than a stable human characteristic. Link: conformity levels should be treated as specific to the sample and time tested; a more representative sample including females would increase generalisability.

Point: low ecological validity. Evidence: the task was a trivial line judgement made in a laboratory. Explain: in real life, decisions to conform often matter more, so pressure to conform could be stronger or might not exist at all; we cannot be sure the findings extend beyond the laboratory. Link: Asch shows the process of majority influence but tells us little about conformity in everyday settings; a more realistic task (e.g. photographs of real objects) would improve validity.

Point: highly standardised and controlled. Evidence: every participant over the trials received the same line-judgement tasks in exactly the same way and conformity was measured with an objective measure, the error rate. Explain: this means the effect of the majority can be retested to check reliability, and the controls increase internal validity because the errors were affected only by the majority and not by other variables. Link: supports the conclusion that the majority caused the conformity observed.

Point: participants were deceived and not fully informed. Evidence: they were deceived about the other group members being confederates and were put under considerable stress by having to give a public judgement inconsistent with the majority. Explain: this breaches informed consent and protection from harm. Counterpoint: the deception was a necessary part of the procedure because knowing the true nature would have changed behaviour; participants were interviewed and debriefed straight afterwards (1951). Link: the ethical cost may be justified by the validity of the findings, but it remains a limitation.

Point: conformity often occurs in groups whose members are familiar (family, friends), not virtual strangers or loose acquaintances. Evidence: Kazuo Mori and Miho Arai (2010) partially replicated Asch with participants who knew the confederates and found that males (but not females) were less likely to be swayed by the incorrect confederates. Explain: this suggests the original findings on strangers may not show conformity in everyday group life. Link: such research may give more useful insight into everyday conformity.

Improvement 1: use a more representative sample that includes female participants instead of just 50 male students - this makes findings about conformity to a majority more generalisable to a wider target population. Improvement 2: use a more realistic task, such as photographs of real objects - this increases validity of the findings about conformity in real-life situations. (Exam tip: always justify why the change improves the study.)

In 1951 some participants conformed on almost every trial and some never did; Asch observed that independent participants were confident, individualistic or task-focused. In 1956 some stayed completely independent and some yielded completely. This supports Asch's conclusion that conformity depends on a combination of situational pressure (unanimity, majority size, support) and the individual's response.

To see what would happen if ordinary people were placed in a simulated prison environment and some were designated guards and some prisoners - how would they behave in their new social roles? It is used as evidence of conformity to social roles.

24 emotionally stable male university students, selected from volunteers who responded to a newspaper advertisement, paid \$15 per day. Two dropped out, leaving 22 (teacher slides).

A mock prison was set up in the basement of Stanford University. Participants were assigned the role of prisoner or prison guard. Prisoners were unexpectedly arrested at home, given prison uniforms and assigned a prisoner ID number. Guards wore uniforms, carried batons and whistles and wore reflective sunglasses (to avoid eye contact). Zimbardo acted as the prison superintendent (he was not a prisoner). The experiment was planned to last 14 days (teacher slides).

24 emotionally stable male students (22 after two dropped out) volunteered for \$15/day and were assigned as prisoners or guards in a mock prison in the Stanford University basement; prisoners were arrested at home and given uniforms and ID numbers; guards had uniforms, batons, whistles and reflective sunglasses; Zimbardo was superintendent; planned to last 14 days.

Within the first few days the guards grew increasingly abusive and controlling towards the prisoners - waking them in the night, making them clean toilets with bare hands and carrying out other degrading activities - while the prisoners became passive. Guards and prisoners conformed to their roles even when they were unaware they were being watched. Five prisoners were released early because of emotional distressreakdown. The study was terminated after 6 days (planned 14). Zimbardo, as superintendent, lost insight.

Participants adopted the social roles they were given, showing the power of conformity to social roles. This is the type of conformity called identification: we conform to the norms of the role while we are in that role (but not necessarily when out of it).

The mock prison was at Stanford (not Yale) University; 24 were recruited, 22 completed; Zimbardo was the superintendent (not a prisoner); five (not eight) prisoners were released early; the study was stopped after 6 days (not 10) although it had been planned for 2 weeks (14 days).

It was a replication of Zimbardo's study and showed that conformity to social roles is not simply unthinkingly playing the role you are given; it is dynamic and influenced by factors such as identification with the group you represent.

Point: unrepresentative sample. Evidence: only 22 male American students took part. Explain: males could behave very differently from women, and people from other countries might not conform to social roles as quickly as Americans. Link: the findings cannot be generalised to the whole population.

Point: it was a highly controlled study so it can be replicated. Evidence: Zimbardo controlled variables such as participant selection (e.g. only emotionally stable participants) and the research has been conducted again (the BBC study). Explain: control and a repeated study mean the findings can be retested to check consistency. Link: supports reliability, although the BBC study qualified the conclusion about conformity to roles.

Point: the study helps us understand real-life events. Evidence: it can help explain cases such as Abu Ghraib, where guards abused prisoners of war. Explain: it suggests such abuse can arise from the social role and situation rather than just from individual 'bad' characters. Link: gives the research practical relevance beyond the laboratory.

Point: lack of realism and demand characteristics. Evidence: it was suggested participants were play-acting and reflecting stereotypes of how guards and prisoners should behave; one guard based his role on a character, and prisoners rioted because they thought that is what real prisoners did. Explain: demand characteristics are cues that lead participants to behave as they think is expected, so behaviour may not show genuine conformity to roles. Link: reduces validity of the conclusion that people automatically conform to social roles.

Point: breached protection from harm. Evidence: five prisoners left early because of distress; participants felt they could not withdraw because they were being paid and were heard saying they could not leave until their sentences were up. Explain: this undermines the right to withdraw. Counterpoint: Zimbardo followed Stanford University's ethical guidelines and debriefed participants afterwards, but acknowledged the study should have been stopped earlier. Link: serious harm outweighs these safeguards.

Pick two clear criticisms (e.g. lack of realism/demand characteristics and ethics/protection from harm). For each: state the criticism, give specific evidence from the study (play-acting, one guard based on a character; five prisoners released early, felt unable to withdraw), then explain why it weakens the study's conclusions.

Class slide ('Asch's baseline procedure'): Aim - to investigate whether social pressure from a majority could affect a person to conform. Sample - 123 American men. Procedure - groups of 6-8 with one naive participant and the rest confederates; the naive participant always sat last or next to last; the task was to match a standard line to a comparison line; 18 trials in total, of which 12 were critical trials where the confederates gave the same wrong answer. Results - 36.8% of participants' answers were incorrect; 25% never conformed; 75% conformed at least once. Conclusion - pressure from a majority in an unambiguous situation can lead to a person conforming. (The book gives different trial/percentage figures - see the discrepancy card.)

Class slides: 18 trials (12 critical), 36.8% incorrect answers, 75% conformed at least once, 25% never conformed. Book: 12 trials where the majority gave a wrong answer, 32% of responses were errors (68% correct). Linked video (Sprouts, 'Asch's Conformity Experiment on Groupthink'): 123 students, 18 rounds with 12 critical trials, 23% always gave the right answer, 72% conformed at least once, 5% always gave the wrong answer. The SimplyPsychology link also gives 18 trials, 75% at least once and 25% never. These sources do not agree, so always say what each figure refers to (responses or participants) and check with your teacher which figures the Edexcel mark scheme expects.

The class slide states that most participants continued to privately think differently from the group but changed their public answer to avoid disapproval (normative social influence / compliance). A quoted participant said he felt he was standing out like a sore thumb, did not want to make a fool of himself, felt sure he was right but thought the group might think him peculiar.

Group size: conformity increased only to a certain point - above 3 confederates there was no further change. Unanimity: conformity decreased significantly when there was a dissenter within the confederate majority. Task difficulty: when the line lengths became more similar (a harder, more ambiguous task) conformity increased.

One limitation 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); identifying lines was trivial so there was really no reason not to conform; and, according to Susan Fiske (2014), Asch's groups were 'not very groupy' - they did not really resemble the groups we experience in everyday life. So the findings do not generalise to real situations, especially where the consequences of conformity might be important.

Mark scheme on the class slide: up to 2 marks for accurate identification of each weakness (AO1) and up to 2 marks for justification/exemplification of each weakness (AO3). So for each weakness give one clear identification, then justify it with a link to the study. Note: the model answer on the class 'Self-assess' slide actually shows STRENGTH-style points (standardised line length and presentation -> same controlled conditions -> can be replicated to check reliability; controlled laboratory experiment -> validity because only the majority, not extraneous variables, affected responses) - use the same identify-then-justify pattern for weaknesses such as artificial task, sample of American men only, or deception.

The slide asks you to match 8 terms with definitions A-I. The definitions given: A - conforming out of a desire to be right; B - changing your behaviour; C - a situation with a completely obvious correct answer; D - deeply accepting group beliefs both publicly and privately; E - a situation with an unclear or subjective answer; F - agreeing with a group publicly while secretly disagreeing; G - conforming out of a desire to be liked and so wanting to fit in with the group; H - a change in a person's behaviour as a result of real or imagined pressure from a group; I - publicly changing your behaviour and privately accepting it while you are with the group; not permanent, it lasts only as long as you are part of the group or continue to value them. Likely matches (inferred - the slide has no key): conformity = H; compliance = F; identification = I; internalisation = D; normative social influence = G; informational social influence = A; ambiguous task = E; unambiguous task = C.

2 marks (Milla attends a recycling meeting then buys recycling bins - describe, using one type of conformity, why she changed): mark scheme example - the information from the group meeting has been internalised by Milla (1); therefore she has a change in both her public and private beliefs and behaviour of recycling her waste (1). 8 marks (school: students conform to punctuality, tidy uniform, attendance and homework; teacher says internalisation, her friend says compliance - discuss how compliance and internalisation can explain it, referring to the context): teacher feedback on student answers - write continuous prose rather than bullet points, and when giving a conformity example use Asch's study (conformity) not Milgram's (obedience). 4 marks (outline one study into conformity): cover aim, sample, procedure, results and conclusion.

A form of social influence where people reject the established norms of the majority and move towards the behaviour and opinions of a minority group. Minorities lack power in numbers, so influence comes from behavioural style. Minority influence is linked to informational social influence and results in individuals internalising new beliefs or behaviours (private acceptance); it explains how social change occurs.

Consistent conviction in their views (successful minorities are consistent), combined with flexibility rather than rigidity. An uncompromising minority is seen as extreme and is less likely to be believed; a flexible minority that accepts alternative opinions is seen as more reasonable and casts doubt or uncertainty over the majority's beliefs, so change is negotiated.

Conversion is the gradual shift in which individuals in the majority come to accept the minority viewpoint. Because the minority has no power in numbers, change in majority opinion builds up gradually through the minority's consistent, dedicated viewpoint (e.g. social movements such as feminism, civil rights, the right to education and environmental change all began as minority views).

A mock jury study: 72 male students decided how much compensation to award the victim of a ski accident caused by a ski company's negligence. When a compensation offer suggested by a minority was consistent and uncompromising it was less effective than when the offer was negotiated. This supports the idea that minority flexibility is an important factor in social influence.

Accepting a minority viewpoint is influenced by whether we identify with the minority group. According to social identity theory we are more likely to identify with a minority if they form part of our ingroup.

(1) To investigate whether social pressure from a minority could influence a majority to give an incorrect response in a colour perception task; (2) to investigate whether a consistent minority would exert greater influence than an inconsistent minority; (3) to investigate whether minority influence created only a change in verbal response or also an actual change in perception.

Female liberal arts, law or social science students, placed in groups of six: two confederates and four naive participants. Women were chosen because the researchers thought they would have greater involvement in evaluating colour.

Groups sat in a row in front of a screen and were told it was a study of colour perception and judging colour and light intensity of slides (with a brief explanation of light intensity). All members were given a test to check for visual abnormalities, which also showed participants that everyone had normal vision (an important control, so naive participants would not attribute different judgements to poor vision). Blue slides of varying light intensity were projected using filters; participants said aloud the colour and rated intensity 0-5 (0 dimmest, 5 brightest). Six slides were each shown six times for 15 seconds, so each participant judged 36 slides. After a practice trial the two confederates consistently said every slide was green. A post-experimental questionnaire was given at the end.

(1) Position of confederates: in 12 groups the confederates sat next to each other and gave the first and second responses; in the other 20 groups they gave the first and fourth responses, to make the second confederate appear independent of the first. (2) Interruption: 13 groups had one-minute pauses after slides 12 and 24, with confederates giving the first and fourth responses, to test the impact of commitment to their first response and allow a possible change of response.

Ten groups completed Experiment 1 and were then asked to take part in another researcher's study on the effect of exercise on visual phenomena. Participants were seated around a table, isolated by cardboard screens, and privately wrote down the colour of 16 discs, each shown for five seconds on a neutral background: three blue, three green and ten ambiguous. It tested whether Experiment 1 had a lasting effect on perception in those who yielded, and whether those who did not publicly yield had still changed their perceptual code.

11 groups were tested to see whether consistency of the confederates affected minority influence: the confederates said the slide was green 24 times and blue 12 times (inconsistent). A control group of 22 participants made colour judgements in groups of six with no confederates present (there was a control group for each experiment).

Groups of six women (four naive participants and two confederates) judged the colour and light intensity (0-5) of 36 blue slides aloud; after a practice trial the confederates consistently called every slide 'green'; variations in the confederates' positions and pauses; Experiment 2 involved private judgements of 16 bluereen/ambiguous discs; Experiment 3 used inconsistent confederates (green 24 times, blue 12 times); control groups had no confederates; a post-experimental questionnaire was given.

Minority influence was measured by the number of 'green' responses by naive participants: 8.42% of answers in the first two experiments (n = 128) compared with 0.25% green responses in the control group (who saw the slides as blue). 32% of naive participants said a blue slide was green at least once. In groups that yielded, 57% of naive participants (about two per group) gave the same response as the confederates. There was no difference between confederates in first and second position or first and fourth. Influence was greatest when the light intensity of the slide was weak, but green responses were higher in experimental than control groups irrespective of this.

Experiment 2: participants who did NOT publicly agree with the minority in Experiment 1 were more likely to shift their judgement in the second phase than those who did publicly agree - suggesting a modification in perceptual code more than a change of verbal response. Experiment 3 (inconsistent minority): only 1.25% of responses agreed with the minority, suggesting consistency is an important behavioural style.

Participants who tended to agree with the minority perceived the slides as having a range of green-blue shades, whereas those who did not agree said the slides were simply blue. This suggests members of the majority genuinely made an effort to consider the minority's viewpoint, possibly because the minority exerted informational social influence.

Minorities can exert some, albeit small, influence in a laboratory situation with female participants, particularly if they are consistent. A consistent minority can modify the perceptual code or norm of a majority. A consistent minority is more capable of modifying a majority's perceptual code than their social response - in contrast to majority influence, which is more likely to influence a social response than a perceptual code.

A way of perceiving or interpreting information. In Moscovici's experiments, when a minority consistently presents a viewpoint it can challenge the majority's perceptual code.

Point: unrepresentative sample. Evidence: only female liberal arts, law and social science students took part (and the book says only American students). Explain: findings may not represent males or non-students, limiting conclusions about minority influence to a female student population; cross-cultural research shows social influence varies between countries. Link: the findings cannot safely be generalised. Improvement: use male and female participants.

Point: highly standardised procedure with tightly controlled variables. Evidence: all participants had a visual-abnormality test (so naive participants would not dismiss the minority as having poor vision); participants were randomly allocated to experimental or control groups (avoiding personality bias in groups); light intensity was varied and analysed. Explain: this means the effect can be attributed to the minority and the study can be replicated to check whether minority influence is consistent. Link: strengthens validity and reliability of the findings.

Point: low ecological validity. Evidence: a laboratory colour-judgement task with confederates. Explain: the laboratory is artificial and does not reflect how minorities exert influence in real life over real issues (such as social movements). Link: we cannot be sure minorities would have the same influence in real situations.

Participants were deceived: they were told it was a colour perception study and the 'other researcher's' study on exercise was a cover, while two group members were confederates, so they could not give fully informed consent. Possible ways to avoid: debrief fully afterwards and gain consent/withdrawal rights (the book notes a post-experimental questionnaire was given). Note: this card is an inference from the described procedure, not a point stated in the book.

Use male and female participants, as Moscovici only used female students. This gives a more representative sample so we can be sure minority influence over a majority extends to the wider population, increasing generalisability.

Both used laboratory colour/line judgement tasks with confederates and measured conformity as incorrect (influenced) responses. Asch: a unanimous majority of confederates (e.g. seven) pressured one naive participant; 32% of responses were errors (1951); explained by normative social influence (compliance). Moscovici: a consistent minority of two confederates in groups of six influenced a naive majority; 8.42% green responses; explained as informational social influence that can change perceptual code (internalisation). The book notes majority influence tends to change social response more than perceptual code.

Supplementary source (an AQA-based revision site, not the Edexcel book). Consistency: intra-individual (the same stance over time) and inter-individual (all minority members agree) - this disrupts the majority's illusion of unanimity and creates doubt (Hogg and Vaughan, 2002). Commitment: dedication and willingness to bear personal cost make the majority take the minority seriously; augmentation principle - if a group is willing to suffer for its cause, the cause is seen as more valid or important (example given: the Suffragettes' hunger strikes and chaining themselves to railings). Flexibility: a successful minority balances consistency with adaptability - a rigid, dogmatic minority is off-putting; Mugny (1982) said minorities must appear cooperative and reasonable, but too much flexibility weakens them.

Supplementary AQA-based source. Snowball effect (Van Avermaet, 1996): minority influence starts slowly, but once a few majority members adopt the view it gains momentum until a tipping point where the minority view becomes the new majority. Group membership (Hogg and Vaughan, 1998): people are more influenced by minorities they see as part of their in-group; an out-group minority can be rejected on identity alone, so in-group members who support the cause act as a bridge. Minority influence leads to conversion (genuine private change) whereas majority influence leads to public compliance (Moscovici, 1980).

Supplementary AQA-based source. Only female participants (gender bias; may not generalise to mixed or male groups). A single laboratory session is a snapshot - real minority influence can take years or centuries. Judging blue-green slides is not controversial and has no personal or societal stakes, unlike real minority movements that face hostility. Later meta-analyses (Wood et al., 1994) found only weak effects of minority influence, so Moscovici's findings may have been overstated. Its summary of the study: consistent minority more influential than inconsistent; the change was private/internalised more than public.

Class assignment brief: 8 marks = 4 marks for AO1 and 4 marks for AO3, and you must include a conclusion. So describe the study accurately (aims, sample, procedure, results, conclusion) for AO1, then give PEEL/GRAVE-style evaluation points for AO3 (e.g. sample, controls, ecological validity), and finish with a judgement.

Aspects such as personality and gender. They are NOT the same as situation and culture, which are separate factors. Exam questions may ask you to identify which factor a scenario shows (personality, gender, situation or culture).

A person with an authoritarian personality is typically submissive to authority, possibly because they were raised in a strict household and punished for non-compliance; they tend to maintain social rules and order, which is linked to higher obedience. Theodor Adorno et al. (1950) devised the F-Scale (Fascism Scale), a questionnaire (Likert-style) used to detect it. Teacher slide: the scale is scored by rating agreement 1 (strongly disagree) to 5 (strongly agree) and totalled; 40+ = strongly authoritarian, 30-39 = authoritarian (for the 10-item class version).

Fascism: a very nationalist and authoritarian position, intolerant of others' views, based on an extreme right-wing political position. Class examples of statements: 'Children should learn to obey authority', 'People that do not follow norms should be treated the same as criminals and severely punished.'

They compared F-Scale scores for 20 obedient participants (who gave the full 450 volts) and 20 disobedient participants (who refused to continue) from Milgram's experiments; participants also completed several personality questionnaires and open-ended questions on their relationship with their parents and with the experimenter and learner. Obedient participants scored higher on the F-Scale than dissenters, were less close to their fathers in childhood (all participants in Milgram's original experiment were male) and admired the experimenter - the opposite of disobedient participants. Conclusion: the obedient participants displayed higher levels of authoritarian personality.

They ran a simulation of Milgram's experiment in a virtual (computer) environment and found authoritarianism was linked to obedience: those with high authoritarian scores were less likely to withdraw, perhaps because they were submissive to the experimenter's authority or wanted to punish the failing learner.

Point: the link to obedience is only correlational and measured by self-report. Evidence: the F-Scale is a self-reported measure, and Elms and Milgram compared scores of obedient and dissenting participants. Explain: because it is a correlation, we cannot be sure authoritarianism causes obedience. Counterpoint (teacher slide): some of the most obedient participants in Milgram's study did not have an authoritarian personality, so it cannot explain all obedience. Support: Dambrun and Vatine's virtual simulation also linked authoritarianism to obedience.

Teacher-slide structure: identify the personality type (authoritarian personality is linked to higher levels of obedience and conformity); explain how it is measured (an F-scale where individuals rate their agreement with statements; a high score indicates an authoritarian personality); give two characteristics (obeys people with high status and power, so more likely to obey orders from an authority figure; more likely to conform to society's laws and norms because those with higher status create the social norms and rules).

A personality theory about the belief in how much control a person has over events that happen to them. Internal locus of control: believe they are responsible for their own actions and are less influenced by others. External locus of control: believe their behaviour is largely beyond their control and is due to external factors (the environment), so they are more influenced by others.

118 participants from Experiments 1-4 (obedient and disobedient) judged the relative responsibility for the shocks by moving three hands on a round disc for the experimenter, the teacher (themselves) and the learner. Dissenters gave proportionately more blame to themselves (48%) and then the experimenter (39%); obedient participants were more likely to blame the learner (25%) than dissenters (12%). Interpretation: dissenters take more responsibility whereas obedient people displace blame - consistent with obedient people having an external locus of control.

Participants with an external locus of control were more likely to obey the experimenter's instructions to shock the learner and also more likely to blame the experimenter. But Blass stressed the situation's role: obedience should be considered an interaction between situation and personality, as some externals may be susceptible to obedience in particular situations.

Point: personality alone does not explain obedience. Evidence: Blass found external locus of control predicted obedience, but concluded that situation and personality interact; Milgram's blame measure was self-reported after the study and only shows blame attribution, not causation. Explain: the evidence is consistent with Rotter's theory but cannot show locus of control causes obedience. Link: the evidence supports locus of control as one contributing factor alongside situational factors.

Internal locus of control: feel less pressure from others, less likely to comply with group norms or feel a need to fit in, so less susceptible to normative social influence. External locus of control: look to others to guide decisions and rely on others' opinions, so more susceptible to normative social influence.

Biondo et al. (1971): students with an external locus of control were more influenced by overt attempts to persuade attitude change than internals; internals reacted against overt influence and ignored subtle influence. Ritchie and Phares (1969): externals were more persuaded by communication from a high-prestige source than internals. Avtgis (1998) meta-analysis: most research found higher external locus-of-control scores were more conforming.

Meta-analysis (a researcher gathers and analyses other people's data to look at effect sizes) found most research showed externals conform more, but evidence is mixed: some studies suggest a link between external locus of control and higher conformity, but it depends on the type and source of persuasion. This reflects that many factors play a role in social influence.

Burger and Cooper (1979) proposed individual differences in motivation to control events. High desire for control: assertive, decisive, takes leadership roles, more likely to resist others' opinions. Low: passive and influenced by others. Burger (1987): 55 students in groups of four rated the funniness of cartoons on a 1-100 scale publicly after three confederates had given their ratings; participants with a high desire for control were less likely to conform to the group norm.

Low self-esteem, high need for social approval, anxiety and insecurity.

(1) Majority size - Asch found three optimal and larger majorities exerted no greater influence; minority size research is mixed (a large minority has more visual presence, a small one may be seen as extreme but can work if its argument is strong and consistent). (2) Majority unity - a broken majority reduced conformity (the book cites a fall from 33% to 5.5%). (3) Ambiguityncertainty - more conformity through informational social influence, giving a more persistent change. (4) Source status, reputation, credibility or expertise - increases influence whether majority or minority (e.g. environmental campaigners).

Because the evidence is correlational and mixed (e.g. Avtgis's meta-analysis depends on the type and source of persuasion; Blass emphasised situation-personality interaction), and personality is only one of several factors (situation, culture) that affect social influence.

(1) Someone who feels an active dislike of people from lower social classes. (2) Someone with fixed and conventional ideas of right and wrong, good and evil. (3) Someone who cannot deal with any uncertainty about the right way to behave and acceptable attitudes or beliefs. (4) Someone who is always willing to be 'bossed' by those of higher status than themselves. The class graphic also lists conformity, respect for authority, conventional thinking, obedience to rules and intolerance of ambiguity.

Many other situational factors affect obedience (proximity, uniform, location), so authoritarian personality contributes but cannot explain all obedience. Middendorp and Meleon (1990) found less-educated people are more likely to display authoritarian characteristics than well-educated people - if correct, it may be level of education, not authoritarian personality, that leads to obedience. Elms and Milgram used the F-scale, which may suffer from response bias or social desirability (participants give socially acceptable answers, e.g. appear more authoritarian because they think those answers are 'correct'), so people may be misclassified. The F-scale is a self-report questionnaire.

Holland (1967) repeated Milgram's study and measured whether participants had an internal or external locus of control. 37% of internals did not continue to the highest shock level (showing independence) compared with only 23% of externals (the homework brief words this as '37% of the internals were disobedient compared to 23% of externals'). This supports the explanation that people with an internal locus of control are more resistant to obedience. (Source: class slides and homework; not found in the textbook.)

Internal locus of control (Rotter, 1966): a high level of personal control over lives and behaviour; hold themselves responsible for actions ('I made it happen'); base decisions on their own beliefs; more self-confident. External locus of control: belief that life is determined by external/environmental factors such as luck ('wrong place, wrong time'); do not believe they exercise personal control; more likely to be influenced by others. Link (from the slides' task and the book): externals are more likely to obey and conform; internals are more likely to resist. Slide scenarios: not applying for Head Girl because she is 'not popular enough' and checking a horoscope daily = external; confident of getting a job because of good exam results and interview = internal. Homework also asks you to find one study supporting LOC and conformity (textbook p.33; see the Avtgis card).

Supplementary video (not Edexcel-specific): authoritarian personality is a dispositional explanation linked to strict, punitive parenting (repressed hostility displaced onto weaker or outgroup members), developed by Adorno et al. after the Second World War and measured by the California F-scale (F = fascism). Dimensions named: conventionalism, authoritarian submission, aggression, anti-intraception (rejecting emotion, imagination, introspection), superstition and stereotyping, power and toughness, destructiveness and cynicism, projectivity (projecting one's own unacceptable impulses onto others) and excessive concern with sex.

Experiment 8 used 40 women (most of Milgram's participants were male). Women were virtually identical to men in obedience (65%), with 27.5% breaking off at the 300-volt level. Their rated anxiety was much higher than men's among those who were obedient. Rationale: earlier research suggested women were more compliant, yet women are traditionally seen as more empathetic and less aggressive, which might make them less likely to shock an innocent victim.

In the baseline condition, 66.7% of men and 72.7% of women pressed the 150-volt switch (went on to continue after 150 volts) - little difference. The point at which men and women first needed a prod was also similar. Burger concluded no difference between genders in obedience (and that higher-empathy participants showed early reluctance but this did not affect ultimate obedience).

They adapted Milgram's research so that the victim was a live puppy receiving genuine shocks from college student participants. All 13 women delivered the maximum level of shock to the puppy, compared with only half of the men - women were much more compliant.

Blass reviewed ten obedience experiments and found obedience levels between men and women were consistent in nine of them. The exception was Kilham and Mann (1974), a partial replication of Milgram in Australia, which found women far less obedient (16%) than men (40%). Because the studies used different procedures and gave inconsistent outcomes, this may indicate there are few gender differences in obedience despite traditional beliefs that women are more compliant.

Point: the evidence is inconsistent so gender is not a reliable predictor of obedience. Evidence: Milgram (Experiment 8: 65% for women, equal to men), Burger (66.7% vs 72.7%) and nine of ten studies in Blass's review found no gender difference, but Sheridan and King found all 13 women obeyed vs half the men, and Kilham and Mann found women 16% vs men 40%. Explain: the studies used different procedures (e.g. a puppy victim, a different country), so differences may reflect procedure rather than gender. Link: concluding gender strongly affects obedience is not supported.

Limited evidence is given. Mori and Arai (2010) replicated Asch with participants who knew the confederates and found that males (but not females) were less likely to be swayed by the incorrect responses. Neto (1995) found a higher rate of conformity among female Portuguese university students than Asch's participants. (The LMS lists an Eagly and Carli (1981) meta-analysis task on gender and conformity but the book does not include it - see the Eagly source-conflict card.)

Individualistic cultures emphasise personal goals, independence and self-expression (teacher slide examples: USA, UK); they tend to behave more independently and tolerate dissent, so resist conformity or obedience. Collectivistic cultures emphasise group harmony, family/community goals and interdependence (examples: Japan, India); compliance is important for group stability (Smith and Bond), so we might assume they are more obedient/conforming and less tolerant of dissent.

Percentage of full obedience: Milgram (1962) US 65%; Edwards et al. (1969) South Africa 87.5%; Bock (1972) US 40%; Kilham and Mann (1974) Australia 28%; Shanab and Yahya (1977) Jordan 73%; Miranda et al. (1981) Spain 50%; Schurz (1985) Austria 80%; Ancona and Pareyson (1968) Italy 85%; Burley and McGuiness (1977) UK 50%.

On closer inspection, differences may be a product of procedure rather than culture. Example: Ancona and Pareyson (Italy) used a maximum of 330 volts (vs Milgram's 450), and Milgram found 73% in his proximity studies - closer to Italy's 85% - suggesting 330 volts was seen as less dangerous; they also used only student participants, whom Milgram avoided because of their compliant and competitive nature. Burley and McGuiness (UK) used only 20 students and a maximum of 225 volts.

Frey analysed contact-tracing data from COVID-19 lockdowns: collectivistic countries recorded less in-country mobility than individualistic countries despite similarly stringent policies, suggesting individualistic countries were more defiant of the rules or at least recorded as less compliant.

Point: the effect of culture is complex and unclear. Evidence: apparent differences in Blass's table are confounded by procedural differences (voltage limits, student samples, small samples), and there is a lack of research in some cultures or of research that can be compared between countries. Explain: we cannot separate cultural variation from methodological variation. Link: culture may influence obedience, but the exact effect cannot be established.

A meta-analysis of 133 conformity studies that used an Asch-type procedure across 17 countries found that collectivistic countries tended to show higher levels of conformity than individualistic countries. Teacher slide questions ask: what is a meta-analysis, how many studies and countries, which cultures had the highest conformity, and what conclusions can be drawn.

Berry hypothesised the Temne (Sierra Leone) would show more conformity than the Canadian Inuit. The Temne live on a single crop, so an annual effort is needed to ensure the harvest and this requires cooperation; the Inuit are hunter-gatherers who rely less on collective effort. Using an adaptation of Asch's line study, the Temne showed higher conformity than the Inuit, explained by the cooperative values instilled by food production.

Neto (1995): higher conformity in female Portuguese students than Asch. Whittaker and Meade (1967): Brazilian students similar to Asch, Zimbabweans higher. Chandra (1973): native Fijians (36%, comparable to Asch's Americans) more conforming than Indians. Timaeus (1968): German students less conforming than Americans. Milgram (1961): higher conformity in Norwegians. Perrin and Spencer (1980): a virtual absence of conformity in England. Triandis et al. (1988): Japanese students reported conforming less than US students but differentiated in-groups from out-groups more sharply (more likely to conform to other students than to strangers).

Point: general conclusions can be drawn, but the exact influence of culture cannot be determined. Evidence: Smith and Bond's meta-analysis of 133 studies across 17 countries found higher conformity in collectivistic countries, but results elsewhere are mixed (e.g. Triandis found Japanese students reported conforming less than US students; Perrin and Spencer found almost no conformity in England). Explain: many variables (participants, period, task, ingroup/outgroup membership) influence conformity. Link: culture is a contributory factor, not a single cause.

Identify one specific factor and explain how it affects conformity/minority influence (e.g. individual difference: external locus of control - more reliant on others' opinions, more susceptible to normative social influence; situational: majority size or unity; cultural: collectivistic cultures place greater emphasis on group cooperation and adhering to social norms, so higher conformity).

Class slide: overall, Milgram found obedience levels in men and women tend to be equal. Kilham and Mann (1974) tested Australian students in a replication and found only 16% of women obeyed compared with 40% of men (Milgram found 65% for both men and women). Gilligan (1982) distinguished the ethic of justice (the value of equality and fairness) from the ethic of care (values of nurturing and supporting needs). The slide asks you to explain the result in relation to these principles: the women were being ordered to shock another woman by a male authority figure, which might fit an ethic of care. (Source: class slide; Gilligan is not in the textbook.)

Class slide: Eagly (1987) meta-analysis - reviewed many studies on gender and conformity; women often conform slightly more than men but only in some situations; sometimes linked to social expectations, desire to maintain harmony or cooperation. Class self-assess table: men prioritise personal goals over group harmony and are less likely to change their opinion in public; women tend to conform more in social groups and may conform to avoid conflict; both can conform or resist depending on the situation. SOURCE CONFLICT: the LMS Quizlet link and a student-made group presentation instead cite Eagly and Carli (1981) (148 studies); the student presentation claims the overall difference was tiny, that women conformed slightly more only when publicly watched, and that male researchers more often reported women as easier to influence - these are student summaries and the wording/date is not confirmed by the book, so check with your teacher.

Student presentation (uploaded to the LMS; not an authoritative source): Milgram Experiment 8 (1974) used 40 women aged 20-50; 65% gave 450 V, identical to men, but women showed significantly higher anxiety and stress. Burger (2009): 70 adults aged 20-81 screened for well-being; study stopped at 150 V; 70% were prepared to go beyond 150 V versus 82.5% in Milgram at that point; no significant gender difference. A second student PDF (collectivist vs individualistic) repeats Smith and Bond (133 studies, 17 countries, higher conformity in collectivist cultures) but its conclusion gives muddled percentages ('up to 32% and 25%'), so do not use those numbers.

Each is 2 marks: 1 mark for identification and 1 mark for justification. Personality example from the class self-assess slide: if a person has an authoritarian personality they are more likely to be obedient (1) as they are more likely to obey to maintain order and social rules and values (1). Culture example: Milgram's study replicated in collectivist cultures shows higher obedience (administering 450 V) than in individualist cultures (1) because collectivist cultures are interdependent and place more emphasis on group harmony and obedience to authority (1). Gender: say whether evidence shows a difference (e.g. Milgram Experiment 8: women 65%, equal to men) and justify with a study or explanation.

They investigated conformity between human and non-human peers to see whether social conformity is specific to humans (primates were used because they are the most similar animals to humans). Aim: to investigate human and non-human animals' tendencies to adjust their behaviour to a majority of their peers in a similar situation - specifically whether two-year-old children, chimpanzees and orangutans would adjust an individually acquired behaviour strategy after exposure to a majority of peers displaying an equally effective but different strategy.

18 two-year-old children (nine male, nine female; mean age 28 months), 12 chimpanzees (the book and the LMS Quizlet set say seven female and five male; the class slide says 7 males and 5 females - source discrepancy; mean age 121 months) and 12 orangutans (six male, six female; mean age 102 months).

A box with three sections of different colours arranged horizontally (the two outer sections equal in height and the middle section shorter), each with a hole in the top. The box was on the ground for children; for primates it was attached to the outside of a steel mesh of an observation room. A ball dropped into a hole triggered a reward at the bottom of that section; researchers controlled which holes were active. Rewards: peanuts for primates and chocolate drops for children.

(1) Initial learning phase: participants could drop a ball into any hole and learned that only one colour section dispensed a reward; learning criterion = ball in the correct section eight out of ten times. (2) Participants were taken two metres from the box and watched three familiar conspecific peers (demonstrators) use it one at a time; each demonstrator placed the ball in a section that was a different colour from the one the participant had learned and received a reward, twice (six demonstrations in all, creating a majority). (3) The ball was handed to a demonstrator only when the participant was looking; trials were repeated if the participant was not paying attention. (4) Demonstrators were led away but could still see; after 30 seconds the participant was given three balls, one at a time, and could choose any hole - all three choices were rewarded. (5) Choices were video recorded and coded by two independent observers as a switch to the majority, staying with their own preference, or a third option.

Participants (children, chimpanzees, orangutans) first learned that only one colour section of a three-section box rewarded a ball dropped in it (8 out of 10 correct); they then watched three familiar peers each put the ball in a different, also rewarded, section twice (six demonstrations); after 30 seconds they were given three balls to place in any section; choices were video-coded by two independent observers as switch, stay or other.

'Stay' responses were subtracted from 'switch' responses. A score of -3 meant they chose their own preferred response (no conformity); +3 meant they changed to match the peer majority (conformity). Negative scores = no conformity; positive scores = conformity.

Number of participants who chose switch / stay / other: children (n = 18) 12 / 6 / 0; chimpanzees (n = 12) 2 / 10 / 0; orangutans (n = 12) 2 / 8 / 2. Children, but not primates, were more likely to change from their initial preferred strategy to that of their peers. The book states children conformed in over half of trials whereas the primates almost exclusively stayed with their preferred choice.

In a situation where children, chimpanzees and orangutans observe a conspecific peer majority perform a strategy different from the one they learned, children were more likely to adjust their strategy than the primates.

A member of one's own species (peers/demonstrators were the same species as the participant).

Point: artificial task. Evidence: watching a peer place a ball in a hole to get a reward is not a typical everyday task. Explain: behaviour change was shown only under contrived conditions and may not reflect conformity in everyday settings such as the playground. Counterpoint: conformity was objectively measured as switch, stay or other response. Link: lowers ecological validity but not the objectivity of coding.

Point: findings may not generalise beyond those tested. Evidence: the primates were captive, so behaviour may not reflect wild chimpanzees and orangutans; sample sizes were 18, 12 and 12; and there is considerable individual variation in conformity. Explain: one group of two-year-olds and captive primates cannot represent all children or apes. Link: conclusions about species differences should be treated cautiously.

Point: highly standardised procedures and controls. Evidence: all children and primates took part in exactly the same way; the peers were familiar to the participant; coding was by two independent observers. Explain: the behaviour changes can be retested (reliability) and we can be sure changes were a direct result of observing a peer demonstrating a different strategy; findings reflect conformity to someone familiar, not a stranger. Link: supports internal validity.

The study did not test whether the peers' influence had any lasting effect on behaviour change, so we cannot say conformity lasted beyond the immediate test.

Point: it can be used to promote positive behaviour change. Evidence: peers are a source of social influence that mediates behaviour change, so peers could be used to encourage compliant behaviours such as picking up litter or turning off electrical appliances. Explain: e.g. if we know we consume more electricity than our neighbours, it may influence householders to reduce energy consumption. Link: shows the findings have real-world use (the book frames these as examples).

The class 'Key Findings' slide states: children conform more readily than chimpanzees and orangutans (more likely to change behaviour to align with peers); conformity is driven by social motivation - the desire to fit in; and the number of peers (one or three) did not significantly affect whether children copied the behaviour. The linked video (Tute Star, 'Haun et al. 2014') says the same: 2-year-olds, chimpanzees and orangutans were compared; children were much more likely to conform to peers than apes; children's conformity is driven at least partly by social motivation (belonging), not just copying better information; apes' behaviour is more about efficiency or survival; and children conformed as much to a single peer as to a group. (Social-motivation and single-peer points come from the class slide and video, not the textbook.)

Both test conformity to a peer majority in a controlled task. Haun: two-year-old children and primates; a majority of three familiar peers with a rewarded alternative method; measured by choice of hole (switch/stay/other); 12 of 18 children switched. Asch: 50 male college students; a majority of seven strangers/confederates; public line judgements; 32% of responses were errors. Haun found children switched far more than primates (12 of 18 children vs 2 of 12 chimpanzees and 2 of 12 orangutans); Asch focuses on majority size and unanimity in adult humans.

The teacher asked for more examples of how the study was standardised. Examples from the procedure: the same three-section coloured box with a ball-in-hole reward mechanism for all participants (on the ground for children, outside a steel mesh for primates); the same learning criterion (correct section 8 out of 10 times); participants taken about two metres away to watch the same number of demonstrations (six: two from each of three familiar conspecific demonstrators); the ball handed to the demonstrator only when the participant was watching; after 30 seconds, three balls given one at a time with every choice rewarded; choices video recorded and coded by two independent observers as switch, stay or other. This standardisation makes the procedure easy to repeat, so it is reliable.

Class prompts (GRAVE): generalisability - are the samples representative (captive primates, 18/12/12 participants, individual variation)? reliability - standardised methods and why that is good (retestable, same procedure for all)? application - how can the study help use peers positively (e.g. encouraging litter picking or reducing household energy use via peer comparison, per the book)? validity - an artificial task was used (watching a peer put a ball in a hole is not an everyday behaviour, so low ecological validity)? ethics - any ethical issues to consider? (The book does not discuss ethics; possible points are inferences only, e.g. captive animals and children's consent obtained via parents.)

Self-report: methods of gathering data where participants provide information about themselves without interference from the researcher (e.g. how they feel, attitudes, opinions, personality). Questionnaire: a written self-report technique where participants answer a preset number of questions, designed to gather a lot of data from a large sample (post, email, face to face or online). Interview: respondents answer questions directly rather than writing them - a conversation or preset questions asked directly by the researcher.

Closed questions have preset fixed answers (yes/no, a list of options, age bands) and give quantitative data. Strengths: easy and quick to answer and easy to analyse. Limitations: respondents can be frustrated if the options do not match what they want to say, and the researcher cannot know why they answered that way, so the level of detail is limited.

Attitude scales offer more than yes/no so the strength of opinion can be gauged. Likert-style scale: respondents rate agreement with a statement from a fixed set of options (e.g. 5 = strongly agree to 1 = strongly disagree). Ranked scale: respondents rank choices relative to other options; to score it each preference is given a weighting (e.g. most feared animal weighted highest) and the highest total is the top-ranked item.

Open questions have no preset answers so respondents answer freely, elaborating and justifying their opinions (qualitative data). Cons: more time and effort for the respondent, and the answers need qualitative analysis, which can lead to subjective interpretation by the researcher.

Quantitative data is numerical (or can be converted to numbers) and is objective (not open to interpretation). Qualitative data is non-numerical prose and can be subjective because meanings in prose are open to interpretation.

A respondent gives not a genuine answer but one that shows them in a more favourable light, according to prevailing social norms (e.g. few would say they agree with cheating in an exam). It is a problem for socially sensitive issues. Social psychology example (book): people may say they are more obedient than they really are, or less disobedient, to avoid being seen negatively - reducing validity. Many questionnaires are designed to detect socially desirable responses.

Questions must not be too technical, ambiguous or complex (the researcher is usually absent), and must not lead or mislead respondents; personal questions must respect the right to privacy (not intrusive; do not obtain data participants would not voluntarily disclose). Likert scales: an odd number of options lets respondents pick the middle value more often; an even number forces a choice.

If all statements are worded favourably (or all unfavourably), respondents can slip into just agreeing (or disagreeing) with all of them. Solution: reverse some statements and mix them up (e.g. 'Pets make people happy' vs 'Pets do not make people happy').

Reliability = consistency. External reliability: consistency of a measure over time, tested with the test-retest method (the same people complete the same questionnaire on a different occasion; similar responses = reliable). Internal reliability: consistency within the measure, tested with the split-half method (split the questions into two halves and compare; both halves should achieve the same score if all measure the same concept).

Validity = measuring what it claims to measure. Face validity: looking at each question and deciding whether it makes sense for the construct (can be confirmed by an expert). Predictive validity: the test accurately predicts the same construct in the future (e.g. a high intelligence score correlating with later exam success). Concurrent validity: comparing the test with another test measuring the same construct (or evidence from several studies testing the same thing) to see whether they agree.

Structured: standardised, all respondents asked the same preset questions in the same way; easy to administer, no need to build rapport, easily repeated, but data can be superficial and respondents may feel stifled. Semi-structured: preset questions but the interviewer can deviate from the script, giving a more conversational flow; gathers quantitative and qualitative data. Unstructured: begins with a loose aim and is a conversation; the interviewer must probe, build rapport, listen and use non-judgemental language; time-consuming to conduct and analyse and not replicable because different questions are asked each time.

Point: in-depth answers increase validity. Evidence (book example): detailed answers can give reasons why individuals consider others to have authority or power, which helps research into obedience. Explain: richer, qualitative data shows meaning that closed questionnaires cannot. Counterpoint: unstructured interviews are time-consuming, not replicable and open to researcher effects; structured interviews give superficial data.

Qualitative data can quote respondents directly, so all respondent details must be anonymised and personal details disguised. The interviewer must behave sensitively when asking for personal information so as not to breach the respondent's right to privacy.

Interviewer characteristics (sex, age, manner, personality) can influence how truthfully a person responds and whether they disclose (e.g. a male interviewer may get less detail from a female participant about friendships - control by using a female interviewer). Researchers can also unintentionally influence responses through expectations, e.g. smiling approval, or asking questions that confirm their expected outcome (confirmation bias: conducting research in a way that confirms the intended outcome). Reduce by using an independent, trained interviewer unaware of the study aims.

Primary data: gathered first-hand directly from participants or sources for the aims of a study (experiment, observation, survey, content analysis); time-consuming to collect but current and specific. Secondary data: data from existing studies by other researchers (e.g. a literature review or meta-analysis); quicker as it is already collected and gives an overview of many studies, but may be out of date and not specific to the researcher's aims. Literature review: reading, summarising and drawing conclusions from other people's research.

Asch (1951): participants interviewed and debriefed afterwards. Moscovici et al.: post-experimental questionnaire and interviews. Adorno et al.: the F-Scale is a Likert-style questionnaire (self-report). Elms and Milgram: personality questionnaires plus open-ended questions. Burger (2009): screening used questionnaires (28-item Interpersonal Reactivity Index for empathy, 21-item Beck Anxiety Inventory, 20-item Desirability of Control Scale, 21-item Beck Depression Inventory) and a roughly 30-minute structured interview based on the Mini International Neuropsychiatric Interview (MINI) with a clinical psychologist.

Target population: the population being investigated (e.g. people at a hospital). A sample is recruited from it using a sampling technique, aiming for a representative sample (one that represents the population's characteristics) so conclusions can be generalised. Sampling bias occurs if the sample over- or under-represents a type of participant.

Firstly identify the subgroups in the target population; secondly calculate the percentage of each subgroup; thirdly randomly select a proportionate number from each subgroup (and ask if they agree to take part). Strength: the most representative method. Limitations: time consuming and selected people may refuse, leaving an unrepresentative sample.

Random: everyone has an equal chance (e.g. computer-generated numbers or names drawn from a hat); should be representative but may not be, is time consuming and people can refuse. Opportunity: use readily available participants (e.g. students in a common room); limited control over who is recruited, not everyone has an equal chance, and asking directly can pressure people. Volunteer: self-selected via an advert; more ethical because people are not approached, but a certain type of participant tends to volunteer, creating sample bias that limits generalisability.

Features: no predetermined questions; the interviewer follows the natural progression of the conversation and can ask follow-up questions; informal, general topic. Strengths: more in-depth (qualitative) information can be collected; builds rapport so participants may be more willing to share personal information; interesting points can be developed. Weaknesses: interviewer bias (questions made up on the spot); needs a highly trained interviewer; hard to compare interviews and not replicable because every interview differs; time-consuming. (Sources: LMS Quizlet 'Edexcel A-Level Psychology - Interviews', two LMS videos and the book.)

Features: a set format of standardised questions asked in order; the same for everyone; can be face to face or over the phone. Strengths: requires less interviewing skill (everything is set out in advance); easy to compare responses because all interviewees have the same experience; easily replicated. Weaknesses: cannot follow up interesting answers because questions are fixed; questions may be interpreted differently by different participants, reducing reliability; data may be superficial. (Sources: LMS Quizlet, two LMS videos and the book.)

Features: a set of prepared questions (open and closed) that can be varied and expanded on - a mix of structured and unstructured. Strengths: more detailed information can be gathered because questions can elaborate on answers; fixed questions can still be compared across participants while allowing unexpected insights. Weaknesses: findings are harder to analyse because each participant may have been asked different questions; still needs a highly trained interviewer to think of questions in the moment. (Sources: LMS Quizlet, two LMS videos and the book.)

Closed-ended questions give a limited set of fixed answers; open-ended questions let respondents answer in their own words. Closed types: Likert scale (e.g. 'I feel well-supported by my parents in managing my workload' - strongly agree to strongly disagree), rating scale (e.g. 'On a scale of 1 to 10, how manageable is the amount of homework?'), fixed-choice option/multiple choice (e.g. 'Which type of study resource do you find most helpful?'). Errors to avoid: overuse of jargon (specialist/technical terms), leading questions (suggest a specific, desired answer) and double-barrelled questions (two questions crammed into one, but only allowing a single answer). Class facts about questionnaires: a primary research method; fixed written questions; open and closed questions; quantitative and qualitative data; distributed online, by post or in person; can gather large samples.

Acquiescence bias: ticking 'yes' to every question regardless of content; check it by asking a similar question later with opposite wording - agreeing with both shows the respondent is not answering honestly (Psych Boost video). Filler questions: mock questions unrelated to the research that stop participants working out the aim, reducing demand characteristics and socially desirable answers (both videos). Questionnaire weaknesses (second video): low return rate because people ignore or throw away questionnaires; the people who respond may be similar (e.g. free or unemployed), so the sample is unrepresentative; open-ended answers can lead to inter-rater reliability problems because researchers interpret them subjectively. Interview weaknesses: interviewer effects (gender, ethnicity, personality of the interviewer can change responses) and cost/time.

Class task: design a questionnaire on gender differences in obedience (how men and women differ in perceived obedience, comparing their perceived own obedience). It must include 8 closed-ended questions (Likert scale, rating scale, multiple choice) and 1 open-ended question, with thought given to format and instructions to participants. Example Likert item on the slide: 'I consider myself to be an obedient person' scored strongly agree (5) to strongly disagree (1). Also on the class starter slide: reflect on whether you were 100% honest and whether the style of questions affected your answers.

Know each part separately and aim to recall at least one point per mark: aim(s) - 2 marks per aim; sample - 2 marks; procedure - up to 5 marks; results - up to 4 marks; conclusion(s) - 2 marks per conclusion. If asked for the procedure you get no marks for the aim, sample, results or conclusions.

Identify a weakness, give an improvement that actually fixes it, then justify why it improves the study. Example: Asch could use a more representative sample including females instead of only 50 male students, making his findings more generalisable; or a more realistic task such as photographs of real objects, which would increase validity for real-life conformity.

G - generalisability (is the sample representative of the target population?); R - reliability (can the study be replicated?); A - application (how does it help us understand real-world events?); V - validity (real or artificial environment? demand characteristics?); E - ethics (deception, psychological harm, informed consent). For each chosen point: Point, Evidence from the study, Explain the effect on the conclusion, Link back. Do not force a category that has no supportable point.

'Describe': give an account only - no need to justify or evaluate. 'To what extent...' (e.g. to what extent can individual differences explain variation in obedience?): review all relevant knowledgeesearch, make a reasoned conclusion that directly answers the question, and flag it with 'In conclusion...'.

Relate the point to social psychology, not just generic research. Example (book): a weakness of questionnaires is social desirability bias - people may say they are more obedient than they are to avoid appearing disobedient; a strength of interviews is in-depth answers on why people consider others to have authority, increasing validity.

Treat it as a process: 'Firstly... secondly... thirdly...'. If asked to define a sampling method, add an example for extra marks (e.g. opportunity sampling: asking students in the common room at the time who agree to take part).

A result is what was measured (e.g. Asch 1951: 32% of responses were errors; Moscovici: 8.42% green responses vs 0.25% in the control group). A conclusion is the interpretation (e.g. majority influence produces conformity when unanimous and at size three; a consistent minority can modify a majority's perceptual code). Do not claim a cause that the evidence cannot support (correlation, e.g. F-scale and obedience).

The LMS assessment note says it is 1 hour long and contains a full Section A and Section C from Paper 1: Section A is 1-8 mark questions on the social psychology content and Section C is one 12-mark question on social psychology. The topics to revise are all social psychology topics listed on the LMS (The visible note lists Units 1-11; the rest of the note is cut off after a 'Read more' link.).