Sunday, 1 March 2015

Social Influence - Obedience

Obedience is compliance to commands given by an authority figure. 

Milgram (1963)

Procedure

- 40 male volunteers paid $4.50, deceived into thinking they were giving electric shocks.
- Participants told study was about punishment and learning.
- Genuine participant always had the teacher role while a confederate played the learner.
- Learner had to remember pairs of words and say the correct word when teacher read out stimulus word.
- Shock system had 30 levers with levels of shock.
- Participant watched confederate get strapped into chair with electrodes attached to his arms.
- Confederate starts by answering correctly and then made mistakes. Shocks started at 15 volts and rose in 15-volt-intervals to 450 volts.
- If teacher hesitated, researcher encouraged him to continue.
- No shocks were actually administered.
- Experiment continued until teacher stopped or reacher the 450 volts.
- Participant debriefed.

Findings 

- All participants went to at least 300 volts.
- 65% of participants went to the end of the shocks.
- Most participants wanted to stop and were stressed but carried on when a researcher told them to.

Conclusion

- People will obey orders that go against their conscience.

Evaluation

+ Lab experiment - high control.
+ Lab experiment - replicable therefore reliable.
+ Lab experiment -  high internal validity. (Participants believed shocks were real so it set out what it was supposed to measure).
+ Same study conducted in different countries making it more generalisable e.g. Germany and Jordan.
+ Participant debriefed.
+ Volunteer sampling made it easy to find participants.
+ Influential research.
+ Participants had different occupations - more generalisable.
+ Participants were immediately and thoroughly debriefed.
+ In social influence research it is often necessary to deceive participants because they may behave differently if they know the true nature of the experiment.
+ Introduced to learner and told the shocks weren't real and got to see the learner was okay. Also told their behaviour was normal and assured them others behaved the same way. (Disobedient participants were told their behaviour was desirable).
+ Gave participants a questionnaire and 84% said they were glad to be in the experiment.
+ External psychiatrist assessed participants a year later and there was no evidence of psychological harm.
+ Cost-benefit analysis.
+ Got presumptive consent, (1% to 450v).
+ It was still possible to withdraw - 35% of participants refused to go to obey.
+ Hofling's field experiment was in a real life setting and still found a high level of obedience. 
+ Milgram disputed claims from Orne and Holland by using video evidence of participants being distressed and gave reports. Some of them also refused who j wouldn't have made sense if they thought the shocks were fake.

- Androcentric - low population validity.
- Small sample size - low population validity- not representative.
- Ethnocentric as it was only conducted on Americans. (Culture bias)
- Lab experiment - lacks mundane realism. Low in ecological validity because it's an artificial setting. (People wouldn't give shocks in real life).
- Unethical - deception (told it was about learning and punishment and there were confederates as well as thinking the shocks were real).
- Unethical - protection from harm (participants would experience extreme stress and thought they had injured or killed someone else and could have caused psychological harm).
- Unethical - confidentiality - videos.
- Unethical - Informed consent.
- Unethical - Right to withdraw made very difficult as there was pressure to continue.
- Outdated.
- Volunteer sample. 
- Orne and Holland criticised validity and claimed participants were going along with the study and didn't think the shocks were real.

--------------------------

Hofling et al (1966)

Procedure

- Experimenter phoned 22 nurses who were alone on wards in different hospitals introducing himself as a doctor who wasn't real. 
- He ordered them to administer a dosage twice the maximum allowance of a drug not on the ward list. 

Findings 

- 21 out of 22 nurses obeyed. 
- They were stopped before they could administer the drug and debriefed. 

Evaluation

+ Field experiment - more likely to show natural behaviour. 

- Couldn't get fully informed consent. 
- Participants deceived (fake doctor).
- Causes distress - protection from harm. 
- Low population validity - small sample size. 
- Low ecological validity - only in hospitals. 

-------------------------

Bickman (1974)

Procedure

- Field experiment in New York. Asked passers by to:
    • Pick up rubbish 
    • Lend money 
    • Stand on other side of sign
- Half of the time the experimenter dressed in a security uniform and street clothes the other half of the time. 
- Measured the number who obeyed. 

Findings

- 92% obeyed to lend money when in uniform. 
- 49% obeyed to lend money when in street clothes. 

Evaluation

+ Field experiment - higher ecological validity. 
+ Less likely to have demand characteristics because they don't know they're taking part. 
+ More likely to be able to establish cause and effect than natural experiment. 

- Informed consent can't always be obtained. 
- Less control than lab - low internal validity. 
- Only one uniform used. 
- Participant variables. 


No comments:

Post a Comment