HideShow resource information
  • Created by: Luke
  • Created on: 14-05-13 12:08



  • There were approximantly 4450 male and female participants that took part in the subway study.
  • The participants were on average 45% black and 55% white.
  • The average number of people in each carriage during the research was 43.
  • The amount of people in the 'critical' area of the carriage was 8.5.
  • The participants were kept in the carriage for about 7.5 minutes when the field experiment occureded because on the particular route chosen between Harlem and Bronx on the New York subway, there wasnt a stop between 57th street and 125th street.
1 of 15

Piliavin Aim

The Aim:

The aim of Piliavins study was to investigate factors that affect helping behaviour.

The factors they were interested in was:

  • The type of the victim, either drunk of ill (cane)
  • The race of the victim, either black or white
  • The frequency of helping
  • The race of the helper
  • The speed of helping

Also the field experiment also investigated the impact of the presence of a model, aswell as the relationship between the size of the group and the frequency of helping.

2 of 15

Piliavin Method and procedure part 1

Piliavin method and procedure: part 1

  • The method used in Piliavins study was a type of field experiment because the participants were not aware that they were taking part in the experiment but the IV was controlled and not naturally occuring (i.e. type of victim), but also a observation because the participants behaviour was observed by two female observers sitting in the adjacent area of the carriage.
  • The participants were approximently 4450 men and women traveling on the New York subway between Harlem and Bronx between 11am to 3pm on weekdays between April 15th to June 26th, 1968.
  • On each trial, a team of four students (2male + 2 female), boarded a train using different enterances. There were four of these teams of four who together collected data for 103 trials.
  • The female confedirents sat outside the critical area of the carriage (in the adjeacent area) and recorded data as unobtrusively (meaning without being noticed) as possible during the journey via a observation method.
  • The male part of the team were spilt given two jobs. One of the males became a model which would help the victim up if nobody helped him and the other male would be the victim. This is the person who falls to the ground standing next to a certral pole in the critcal carriage either acting at the drunk or ill victim.
3 of 15

Piliavin method and procedure part 2

Piliavin method and procedure: part 2

  • As the train past the first station after about 70 seconds, the vivtim staggered forward and collapsed on the floor either acting as the drunk or ill victim.
  • Until he recieved help, he stayed motionless on the floor looking at the ceiling. If he recieved no help by the time the train had slowed at the stop, the model would help the victim up onto his feet.
  • At the stop the tesm got off and waited seperatly until all the participants had got off and then proceeded as a group to the next platform to board the train going in the opposite direction for the next trial.
  • About 6-8 trials were run each day and the same type of victim was used for the whole day.  

About the models:

There were four different model conditions used for both the drunk and cane conditions.These were

4 of 15

Piliavin about the confederants

Piliavin: about the confederants


  • The four victims (one from each team) were males, aged between 26 and 35. There were three white and one black, and were all identically dressed.
  •  On 38 trials the victims smelled of alcohol and car­ried a bottle wrapped tightly in a brown bag (drunk condition),
  • On the remaining 65 trials they appeared sober and carried a black cane (cane condition).

The victims behaved identically whether in the drunken or ill conditon when falling and each one of the male confederants played both type of victim eventually.


  • The models were all male also and aged between 24-29. They were all white and cusually dressed but not wearing identical clothing.
  • They would either stay in the:
    • Critcal area and wait until they passed the fourth station 
5 of 15

Piliavin :about the confederants part 2

Piliavin about the confederants part 2


  • The models were all male also and aged between 24-29. They were all white and cusually dressed but not wearing identical clothing.
  • They would either stay in the:
    • Stand in the critcal area and wait until they passed the 4th station and then help the victim(after 70s)
    • Stand in the critcal area and wait until they passed the 6th station and then help(after 150s)
    • Stand in the adjacent area and wait until they passed the 4th station and then help(after 70s)
    • Stand in the adjacent area and wait until they passed the 6th station and then help(after150s)
  • When the model helped the victim, he would sit him upright and stay with him for the remander of the journey/trial.
6 of 15

Piliavin: observations

Piliavin: observations

When the two female students observed the participants from the adjacent area they recorded:

  • The total number of passengers/participants who came to the victims assistance
  • The race, sex and location in the carriage of the helping passengers/participants 
  • The race, sex, seating or standing position and location of all the passengers in the carriage.
  • A second observer recorded how long it took for the help to the victim to arive
  • Recorded comments from other passengers when the victim fell and afterwards and also comments from the people sat next to them by asking.


7 of 15

Piliavin: Results and findings

Piliavin: Results and findings

Quantitative data:

  • The cane victim recieved help on 62/65 trials
  • The Drunken victim recied help on 19/38 trials
  • 60% of the 81 trials where help was given, more than one person provided help.
  • Once one person started to help, the victim conditions (black/white/drunken/cane) had no effect on additonal people helping.
  • The race of the victim made no significant difference but it was seen that in the drunken condition, people of the same race of the victim tended to help more than different races.
  • 90% of the helpers were male
  • 64% of the helpers in total were white

Qualitative data:

  • COMMENTS SUCH AS "its for men to help" " i wish i could help him - i'm not stong enough"
8 of 15

Piliavin: Results and findings - Diffusion of resp

Piliavin: Results and findings - Diffusion of responsibility

Diffusiuon of responceability is the idea that in an emergency, if there are lots of people about people dont feel as much that it is them who have to do something because the responsibility is almost shared with the other people around them or bystanders.

Diffusion of responsability was not found in Piliavin's study. The evidence goes against it because they found that in the larger groups of people in the carriages, help was recieved at the quickest rate.

9 of 15

Piliavin: Results and findings - Reasons

  • During the experiments, nobody left the situation in order to get rid of the arrousel.This was because the route choosen had no stops between 59th street and 125th street meaning the participants couldnt leave the carriage. 
  • Although in 21 of the trials, a total of 34 people left the critical area after the victim had fallen, particually when the drunken victim was used.
  • More comments were recorded on the drunken trials than the cane when nobody helped after 70 seconds because od dicomfort  and the hope that others would confirm that inaction  was appropriate.
10 of 15

Piliavin: Explanation of behaviour via the Cost re

Piliavin: Explanation of behaviour via the Cost reward model

Piliavin developed a model to explain the results gathered called the "cost reward model". This is the idea that when a person observes an emergency situation, it creates an emotional responce which creates arousal.Arousal may be percieved as fear, disgust or sympathy depending on the situation.

The state of arousal can be increased by a number of factors including:

  • Empathy with the victim (can you see yourself in the victims situation)
  • being close to the emergency
  • The length of time the emergency goes on for.

Decreased by:

  • Helping
  • Seeking help from another source
  • leaving the scene
  • deciding theperson doesnt need or deserve your help
11 of 15

Piliavin: Explanation of behaviour via the Cost re

Piliavin: Explanation of behaviour via the Cost reward model part 2

Cost reward model: This is the idea that we weigh up the costs and rewards of helping people in situtations. These can be:

  • Costs of helping - effort, embarrassment , physical harm
  • Costs of not helping - Self blame, percieved censure from others
  • Rewards of helping - self phraise, help for the victim
  • Rewards of not helpng - not getting involved, not risking the possible costs of helping.

Piliavin believes that in his study people weighed up the above and this helped them to decide whether to help or not. This can be seen where:

    • The drunken victim gets helped less at only 19/38 trials due to the cost of helping possibly being harm to passenger.
    • 10 % of helpers being women because they may not want to help because of the cost of effort because they are too weak or embarrassed if they cannot help if they can.
12 of 15

Piliavin: evaluation - Ad's disad's


  • The study has high inter-rater reliability because two observers were used , whose findings were compared against each other.
  • large sample size of 4450 makes it more representative and easier to generalise.


  • Low ecological validity because the study was done in one location which was the subway journey between New York between Harlem and Bronx and back. This means that helping behaviour cannot be generlised easily because only train passengers were used which could be different to people in shopping centres for example. Also only a small area of the world was used which makes generalising the findings to the wider population hard because not all types of people were observed (i.e. from diferent countries).
  • By using male victims only, it makes the research gender bias to males because the behaviour shown when males need help is the only one recorded, but behaviour for females in need of help could be different, making it harder to generlise.  
13 of 15

Piliavin: evaluation -

Piliavin: evaluation - 

Usefullness - the findings can be applied to a real life situation because the study was performed in a real life situation  and the participants did not know they were being observed because of the field experiments, which means that they would show natural behavour and the data collected from them would be more valid because it is more accurate to helping behaviour because it is uninfluenced.

Ethical issues-

Protection from harm - The participant could be physically hurt when the victim falls on the floor because in busy carriages, the victim could fall on the participant. The particiapnt could also become mentally harmed by the stress of the situation also

Right to withdraw - since the research was a field experiment, the participant would not know that they were taking part so did not know that they had the right tom withdraw at any time. Also because of the Harlem to Bronx route was chosen, the pariticipant was not able to leave the carriage even if they knew their right because they wasnt a stop from 59th street to 125th street.

Consent - the Participants did not gibve consent before the experiment because if they did they would have, they would know they were being observed and may chnage their behaviour causing less valid results.

14 of 15

Piliavin: Changes and implications

Piliavin: Changes and implications

  • Increase the number of places the experiment would be done in. For example, 20 trials of both the drunken and cane victims could be done in 5 different countries. They could be the: London, Paris, German, New York and Moscow subways. By doing this is would make the results kore representative because they would be easier to generalise to the wider population because different types of people have been observed with their helping behaviour.
  • More observers.This could be done by increasing the observers from 2 females to 2 females and 2 males observing from the same area. This would mean that the findings would have higher inter-rater reliability because more observations were being compared against each other, but also the reuslts may be more accurate because the different genders may pick up things that would not have been discovered with the single gender observers.
15 of 15


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »