Social Psychology - Piliavin, Rodin and Piliavin (1969)

Social Psychology - Piliavin, Rodin and Piliavin (1969)

1. Aim

   • To investigate how several variables can effect helping behaviour using the New York Eigth Avenue - Independent Subway as a 'Laboratory on wheels'.
   • The Variables were:
     • 1)
       • Responsibility of the victim It was predicted that a person who was drunk would receive less help than someone who is ill. This suggests that this would be down to the personal cost of helping someone whose behaviour was unpredictable.
     • 2)
       • Race of Victim It was predicted to affect the rate of helping and that people were more likely to help someone of their own race than a different one.
     • 3)
       • Effect of Modelling If the model put in place had any affect on the surrounding participants.
     • 4)
       • Effect size of group

2. Participants/Sample

   • Opportunistic
   • 4450 men and women unknowingly took part
     • 55% were white, 45% were black
     • 43 people on average in a cart, 8 within the critical area (on average)
     • 16 general studies students from the University of Columbia carries out the study.
       • They were spilt into 4 teams of 4.
         • 2 boys and 2 girls in each group.

3. Method

   • IV's
     • 1) Type of victim 2) Race of Victim 3) Presence of Model (Early/late)
   • DV's
     • 1) Time taken for the first passenger to offer help 2) Total number of passengers who helped 3) Gender, race and location of every helper 4)Spontaneous comments made by passengers
   • Field Experiment
     • Location
       • Carried out on the A and D trains of 8th Avenue in New Yor City.
         • There are no stops for 7 1/2 minutes
       • Weekdays between 11am and 3pm between 15 April and 26 June
         • Longitudinal study

4. Prodcedure

   • Confederates
     • Two males in each group were the model & victim, the two females observed in an attempt to create inter-rater reliability.
       • There were four groups in total but only one black male who did not want to act being drunk out of fear of being harmed.
     • Victim Model Conditions
       • Drunk Victim
         • For 36 trials the victims smelt of alcohol and carried a brown bag
       • Early Model (Critical)
         • hadModel stood in critical area and waited for 70 seconds before helping the victim if no-one else did
       • Late Model (Adjacent)
         • Model stood in area adjacent to the critical area and waited 150 seconds before helping the victim if no-one else had
       • Late Model (Critical)
         • Model stood in critical area and waited for 150 seconds before helping the victim if no-one else had
       • Cane Victim
         • For 65 trials the victim appeared sober and carried a black cane
       • Early Model (Adjacent)
         • Model stood in the area adjacent to the critical area and waited for 70 seconds before helping the victim if no-one else had.
   • The emergency consisted of the victim staggering and collapsing, laying there staring blankly at the ceiling until receiving help.
     • If no help was received, the model would help the victim back to is feet and the team would exit at the next stop.
   • To observers would write down their observstions.
     • See the Dependant variables for what they were observing.

5. Results

   • Individuals who appear to be ill are more likely to receive aid than someone who appears drunk, even when the immediate help was of the same kind.
     • The cane victim received 95% spontaneous help from their trials.
     • The drunk victim received only 50% spontaneous help of the trials.
     • The patterns were the same for both the black and white victims.
   • 90% of the spontaneous first helpers were male
   • There was a tendency for same-race helping within the trials, especially when the victim was 'drunk'.
   • No strong relationship between the number of bystanders and the speed of helping. The expected increase of 'diffusion of responsibility' with a greater number of bystanders was not obtained from groups of this size.
   • People mainly left the area in the 'drunk' victim condition.
   • Spontaneous Comments
     • Most comments was observed in trials in which no-one helped after 70 seconds
     • More comments were obtained during the drunk victims trial than the cane victims.
   • 60% of trials where the victim received help, more than one person helped.
   • The longer the emergency continued without the victim being offered help:
     • The less impact the model had on the helping behaviour of the people within the cart
     • The more likely it was that the individuals left the immediate area to avoid the siutation
     • The more likely it was that participants discussed the incident and it's implications for their behaviour

6. Conclusion

   • Piliavin & Co proposed a model of response to emergency situations: The Arousal Cost-Reward Model. A Heuristic device that could be used to predict the behaviour of any given emergency.
     • Arousal state is higher when:
       • -The more one can empathise with the victim (see themselves in the situation). -The closer one is to the emergency. - The longer the emergency continues without help being given.
     • Observation of an emergency creates emotional arousal in the bystander.
     • Arousal can be reduced by:
       • - Helping directly - Going to get help - Leaving the scene of emergency - Rejecting the victim as undeserving of (your) help
   • Situational explanation of bystander behaviour was supported/proved.
   • Bystander responses are determined by weighing up the costs and rewards of helping or not helping.
     • The drunk victim is helped less because they are seen as unpredicatable. There is seen to be more cost in hekping them than reward to ones person.

7. Usefulness

   • Regarding the existence of altruism*, the model created due to the results this study would suggest that our behaviour is calculated more on the terms of personal gain than any act of selflessness.
     • Alturism is where someone does something that has, in no shape or form, any benedit to oneself.
   • Gives us insight on why humans will help some people but avoid others.

8. Weaknesses

   • Unethical
     • No Debriefing
     • No informed Consent
     • Psychological Harm
       • Seeing the victim fall over may have caused people stress.
       • Seeing someone so drunk that they fall over and are unable to get up may be upsetting. The same goes for an ill person.
     • No chance to withdraw
   • Trials
     • An uneven amount of trials were conducted within each condition. There was more cane trials that drunk trisls, and they were distributed unevenly across black and white victims.
   • Validity
     • Only tells us how people will reaction in close quarters in such an acciedent. If it happened in the street people may reaction differently.
   • Low Controls
     • The fact it was a field experiment that not all extranuous variables could be controlled. There was always the chance that a participant noticed the observers and caught onto the hoax.
   • Sample
     • Only represantative for American people who take the subway.
   • Observers
     • The recorders recorded different information to each other.

9. Strengths

   • Ecological Validity
     • It was a covert field experiment meaning the experiment took place in a natural setting and the oarticipants were completely unaware that they were taking part in the study.
     • This means that the oarticipants are very unlikely to be subject to demand characteristics as they will be acting as how they would respond in a real life situation (as it's what the participants think it is).
   • Useful
     • The study is applicable to everyday life.
   • Inter-rater reliability
     • There were two observers in each group in an attempt to eatablsh a high inter-rater reliability.
   • Representivity
     • 4450 participants is a large sample, containing a multitude of people with a high chance of having different backgrounds, ages, ethnics and genders.
   • The Data
     • Both quantitative and qualitative data was collected meaning both statistical analysis could be made as well as the possible chance of cause and effect of those statistical analysises.
     • The quantitative data collected were the number and type of passengers who helped, their race and how long it took them to respond.
     • The qualitative data was the spontaneous comments made.

10. Background

    • Kitty Genovese was the victim of a thirty minutes stabbing ttackin 1964. 12 witnesses were counted but they took no action other than calling the police and yelling outside. No movement to go outside and stop the attack was made.
    • Pluralistic Ignorance
      • An individual seeks out a cue from another to knowwhat to do/before acting.
    • Diffusion of Responsibility
      • The more people who are present, the lower the likely hood of someone helping a victim there is.
    • Modelling Effect
      • People are more likely to help if they see someone else helping (modelling).