Piliavin

• Pluralistic ignorance+ genuine ambiguity mean people may not help in an emergency because they diffuse responsibility
• Diffusion of responsibility- responsibility is spread between people- less on each individual
• Bystander apathy- believing someone else will do what is necessary so not helping
• Murder of Kitty Genovese- woman stabbed after being followed for 30 min infront of 38 people
• Darley + Latane- bystanders hearing an epileptic fit over earphones led to those who believed other witnesses to be present were less likely to help the victim than those who believed they were alone
• Latane + Rodin- response of a victim of a fall- suggested assistance was far less likely if people were strangers than acquaintances
• Bryan + Test- individuals more likely to be good Samaritans if they have just observed an individual doing a helpful act
• Previous work on victimisation been conducted in a lab using non-visual emergency situations
• AIM- to investigate the effect of variables on helping behaviour in real life conditions
• Field experiment- A+ D trains on 8th avenue- NY subway- 7.5 mins journey between same two stops
• 4 IVs

  -Type of victim- drunk/cane

  -Race of victim- black/white

  -effect of model- 70-150 seconds or -critical or adjacent area or no model

  -Size of group- naturally occuring
• DV- recorded by two female observers in adjacent area

  -frequency of help

  -speed of help

  -race of helper

  -sex of helper

  -movement away from critical area

  -verbal comments from bystanders
• opportunity sample- 4,450 men and women who used NY subway between 11-3 between April 15- June 26- 1968

  -45% black / 55% white
• 4 teams of 4 researchers- two female observers, male victim + male model

  • Victim- 3 white/ 1 black - general studies students- 26- 35- smelled of liquor with bottle in brown bag or sober + cane
  • model, all white 24-29- 4 conditions - critical area early/ late - adjacent area early/late
• observers-

  one recorded DV- race, sex, location of people in critical area+ total numbers of helpers- race, sex and location of each

  one recorded DV- race, sex, location of adjacent area, timing of first helper

  Both recorded comments + tried to elicit response from rider next tp them
• Victim stood near pole in critical area- after 70s staggered forward + collapsed, stayed on floor staring at ceiling, if no help received by time train stopped, model helped. Change of platform to repeat in other direction
• 6-8 trials on a given day 'victim condition

  -more cane than drunk trials- uneven due to victim unwilling to play drunk- student strikes did not allow to correct this
• cane received spontaneous help 95% of the time compared to drunk- 50%
• 100% of help for cane- 81% drunk

  Help offered quicker to cane- within 5s compared to drunk- 109s
• 90% first two helpers male

  slight tendency to help same race- especially if drunk

  • no diffusion of responsibility- faster in larger groups
  • more comments in drunk than cane - most given with no help within first 70s
• Concluded- ill individual more likely to receive help than drunk

  men more likely to help a male victim

  people more likely to help same race- particularly if they deem situation to be of own making
• Concluded- no strong relationship between no of bystanders + speed of helping when incident visible

  when no escape, bystanders more likely to help

  cost/ benefit analysis before helping
• concluded- spontaneous help irrespective of race or victim type