AS Psych: Piliavin et al. (subway Samaritans)

Psychology Being Investigated
1. Bystander apathy is when individuals are less likely to offer help to a victim in the presence of other people
2. There is a diffusion of responsibility when more people are around during an emergency situation
– Individuals feel less of a personal responsibility to act
– Causes people to be less likely to help

Background
Murder of Kitty Genovese in 1964
– Followed and attacked by a man after returning from work in the early morning
– 38 people saw the crime being committed, but didn’t do anything to prevent the murder → bystander apathy
– Triggered Research into bystander apathy

Darley & Latane
– Bystanders who believed there were others witnessing the event were less likely to help in an emergency situation
– But if there’s a model, makes us more likely to help

Aims
1. To investigate bystander apathy and diffusion of responsibility in real life
2. To investigate if “reason” (ill/drunk) affects helping
3. To investigate if race affects helping
4. To investigate if models affect bystander helping

Method
Research method: Field Experiment and Observation
Research design: Independent Measures
Sampling technique: Opportunity

IVs:
1) Type of victim: ill/drunk
2) Race of victim: black/white
3) Size of group of bystanders
4) Behavior of the model: helping 70 secs/150 secs after the collapse

DVs:
1) Time taken for first passenger to help
2) Race, sex
3) Location in carriage
4) Number of passengers who helped

Sample
– Around 4450 men and women
– Passengers traveling on the NYC subway
– From Harlem to the Bronx
– Weekdays between 11am-3pm
– Race composition: 45% Black, 55% White
– Mean no. of passengers per carriage: 43
– Mean no. of people in the critical area (where incident took place): 8.5

Procedure
– 4 teams of 4 student researchers (2 male, 2 female each)
– Same subway route: replicable, standardized, 7.5mins gap (ample time to run trial)

Standard procedure
1. 2 male + 2 female student researchers boarded at different doors
2. Females in area adjacent to critical area*
3. Males = victim & model (in critical area)
– Location of experimental car varied
– Victim stood next to pole in center of critical area
– 70 secs after departing, victim stagger forward and collapse
– Remain supine (lie upwards) until helped
– If NO assistance by passengers: model helped him up when train stopped
– Team disembarked and waiting separately until passengers left station
– Proceed to another platform to board train in opposite direction
– 6-8 trials in a day (same victim condition in each)

Victim (standing)
– 3 white, 1 black
– Dressed identically (Eisenhower jackets, old slacks, no tie)
– 38/103 trials: smelled of liquor, carry liquor bottle wrapped tightly in brown bag
– 65/103 trials: sober, carry black cane
– Behaved identically

Model (standing)
– 4 white males
– 24-29y
– Dressed in informal clothes, but NOT identical
– Raised victim to sitting position, stay with him for rest of trial
– 4 conditions:
1. Critical area/early → 70 secs to help
2. Critical area/late → 150 secs to help
3. Adjacent area/early
4. Adjacent area/late

Female observers (seated)
– BOTH noted comments made by passengers
– 1st observer recorded:
1. Race, sex, location of every passenger in critical area
2. Total number of passengers in car
3. Total number of individuals who helped
4. Race, sex, location of every helper
– 2nd observer recorded:
1. Race, sex, location of passengers in adjacent area
2. Latency of first helper after victim collapse
3. Latency of first helper after model arrived

Results
– A person who is ‘ill’ is more likely to receive help when they collapse than a person who is ‘drunk’
– General public are more likely to help someone of the same race if they are drunk
– Men are more likely to help in an emergency situation than women
– Cane condition received spontaneous help on 62/65 trials
– The drunk victim received spontaneous help on 19/38 trials
– The difference was not due to group size
– 60% of the 81 trials where the victim received help from multiple people
– 90% of spontaneous first helpers were males
– Same race helping was seen for white victims but not for black victims
– Comments by women: “I wish I could help him” or “I am not strong enough”

Conclusion
1. A person who is ‘ill’ is more likely to receive help than a person who is ‘drunk’.
2. Men are more likely to help than women.
3. People are more likely to help people of the similar race, especially if drunk.
4. There is no evidence of diffusion of responsibility in a natural setting; many would offer spontaneous help.

Ethical Issues
✘ Deception: P were led to believe that the drunk/ill victim were actually in need of help
✘ Confidentiality: no individual data was published; we only know the P were people on a New York subway train
✘ Protection from Harm: Participants may have been psychologically distressed when they witnessed the person collapsing or guilty if they had not helped them
✔ Informed consent: P did not know a study was taking place, hence could not give informed consent to participate
✔ Right to withdraw: while P could not exit the train while the study was taking place, they could move to the next car if they wished

Application to Daily Life
1. Findings of the study can be used to educate the public on bystander effect/diffusion of responsibility.
2. Teachers/school curricula can be adapted to teach children about prosocial behavior and how they should help others regardless of the conditions or race of the victims.

*Critical area: area in the vicinity of where the victim stood in the subway carriage