|
|
Background: The murder of Kitty Genovese- psychologists wanted to know how she was killed with 38 witnesses. Theries such as diffusion of responsibility and the cost-reward arousal model were developed to try to explain why such a situation had occurred.
Theories:
IV:
Procedure:
Theories:
- Diffusion of responsibility= the responsibility for behaviour is shared between those present. Here the responsibility for helping the victim would be shared amongst all the passengers in the carriage. As the number of bystanders increases, the likelihood that any individual would help decreases.
- Cost-reward arousal model= where a person evaluates the pros and cons of doing something. In Piliavin the ps weighed out the pros and cons of helping the victim. In the case of Kitty Genovese, the cons must have outweighed the pros and the witnesses probably thought that if they went out to save her, they might be harmed too, and they feared this.
- Type of victim (drunk or with a cane)
- Race of victim (black or white)
- Whether people were more likely to help in an emergency situation if they have seen someone else displaying helping Behaviour (modelling)
- The relationship of group size (diffusion of responsibility)
IV:
- Type of victim (drunk or with a cane)
- Race of victim (black or white)
- Whether the model displays helping behaviour (modelling)
- Number of people in the carriage
- Ethnocentric helping
- Time it took for victim to get help
- Whether help was faster if a model helped
- Around 4450 men and women
- Traveled in New York subway, weekdays, between the hours of 11.00 am and 3.00 pm during the period from April 15 to June 26, 1968.
- The racial composition of typical train= 45% Black: 55% White.
- Mean number of people per car = 43.
- Mean no of people in the ‘critical area’ (where the incident took place) = 8.5.
Procedure:
- *On each trial a team of four students,2 M, 2 F, boarded the train using different doors.
- 4 different teams, whose members ALWAYS worked together, were used to collect data for 103 trials.
- Each team varied the location of the experimental car from trial to trial.
- *The female confederates took seats outside the critical area and recorded data as DISCRETELY as possible for the duration of the ride, while the male model and victim remained standing.
- *The victim always stood next to a pole in the centre of the critical area. As the train passed the first station (approximately 70 seconds after departing) the victim staggered forward and COLLAPSED.
- *Until receiving help, the victim remained on the floor looking at the ceiling.
- *If the victim received no assistance by the time the train slowed to a stop, the model helped him to his feet.
- *At the stop, the team left and waited separately until other riders had left the station. They then proceeded to another platform to board a train going in the opposite direction for the next trial.
- On average 6- 8 trials were run a given day. All trials on a given day were in the same ‘victim condition’.
Victim:
|
Model:
|
Observers:
|
|
Observer 1:
|
Observer 2:
|
Controls:
Results:
QUAL DATA:
Research Method/Design:
+ = the use of the New York subway, a natural environment, allowed observers to gather data in relation to how normal people show helping behaviour when in a restricted area, which is high in Ecol Valid.
+ = DC's unlikely (unless saw exp more than once) as didn't realise they were being observed for their helping behaviour.
+ = Lots of detailed qual data from observations can be gathered.
- = Not realistic, usually people can remove themselves from a situation they are uncomfortable with being in.
- = Less control than lab exp as field exp cant control all extraneous variables. (Hard to isolate the IV and keep all other variables that could affect the DV constant/controlled).
- = Not practical as:
+ = Balanced racial mix of 55% white: 45% black.
+ = Large no of ps, 4450=representative of the target population of NY.
- = Sample was drawn from passengers/individuals on a subway so results cannot be generalised to passengers/individuals on other forms of transport/in other situations whose helping behaviour may be different.
- = Sample was drawn from the New York area, so ethnocentric, so results cannot be generalised to individuals from other areas/countries whose helping behaviour may be different.
- = Ps were passengers between 11.00 am and 3.00pm between April 15 to June 26, 1968, so results cannot be generalised to passengers who travel at other times/between other dates, whose helping behaviour may be different.
Sampling Method:
+ = No order effects
+ = Unlikely ps saw exp before.
+ = Wide range of people from all walks of life.
- = Ps may have seen exp before/realise aim, and produce DCs.
- = Ps on subway may have similar characteristics as have similar lifestyles, so results only generalisable to people who travel on the NY subway.
Reliability:
+ = STANDARDISED PROCEDURE, ps go through the same thing- get consistent results and therefore REPLICABLE.
+ = Because a total of 103 trials were conducted with similar results found in each trial condition results can be considered reliable.
+ = Because the 2 observers in each team always recorded findings for the same coding categories, interpretation of the categories will have been consistent so results can be considered reliable e.g. the same observer in each team recorded
race, sex and location of every rider seated or standing in the critical area.
+ = Because both observers recorded comments spontaneously made by nearby passengers, findings could be checked for reliability.
+ = Quant data recorded is easy to compare.
- = Because each of the 2 observers in each trial were recording different data, no checks were made to ensure inter-rater reliability e.g. timings of latency of first helper were not checked for accuracy.
- = Because there were 4 teams working separately application of the codings of one team may have differed from that of another e.g. gender is not always apparent!
- = No of people in the carriage can't be kept constant.
- = Low generalisability- not many places have no escape.
- = The situation, may on occasions, have resulted in the observers’ views being blocked leaving them unable to see the incident clearly so they could not record /missed key information/write incorrect info (e.g count ps twice), making findings unreliable.
Validity:
+ = Qual data- rich detailed data.
+ = High Ecol valid as study takes place in real life situation.
+ = Unlikely the ps realise the aim of the study unless they have seen it more than once.
- = Due to repetition of the study (between April 15 and June 26, 1968) there is a chance that some passengers may have witnessed the incident more than once and responded with DCs / SDB making the results invalid.
Usefulness:
+ = Study yielded a lot of detailed data.
+ = Highlights that differences can occur between lab and field studies.
+ = Proposes big model for the way people behave in emergency situations (Arousal cost-reward model)
- = Only tells us about a situation where people can't escape.
- = Low generalisability- not many places have no escape.
Ethical Issues:
+ = ps not physically harmed.
+ = Info on ps kept confidential- names and ages of ps not given, except for their race/sex.
- = No informed consent, invasion of privacy, stress, anxiety about seeing someone collapse.
- = Deception- ps thought the researches were fellow passengers, not people who were observing them for their helping behaviour.
- = Passengers who did not help/felt unable to help may have left the study feeling guilty or distressed.
- = No R to W- no escape for the ps, they are stuck on the carriage until the journey ends.
Improvements & Implications:
1a) Ethics:
Advertise on the New York subway prior to the study taking place, that an investigation on 'helping behaviour' is imminent, providing the dates and times that it will be conducted. The results of the study will be posted around the appropriate stations as a form of debrief.
1b) Ethics:
Conduct a study where the victim (drunk/ill) collapses in a more open space, where ps can escape from the situation if they wish. E.g a street
2a) Method:
Conduct the same experiment but with cameras set up in the carriages already. Permission would have to be granted by carriage owner. They would be placed in the 4 corners of the carriage to give the best visibility. After the experiment had finished the researchers would view the recordings and collect the results for the race, sex etc of the passengers and helpers in the carriage.
2b) Method:
Conduct the experiment in the same way as Piliavin, except at the end the observers ask any helpers/passengers Q's such as: 'Why did you decide to help/not help the man who collapsed?', as a form of an informal interview. The observer would use a tape recorder in their bag to record what the passengers say, as this wouldn't make it obvious to the passenger that they were being observed for a study (as writing this down in front of them would be obvious and it would not be practical to expect the observer to remember the replies the passengers gave to these Q's.
- The victims acted out their roles in a standardised way so all participants were exposed to the same behaviour to make it fair.
- Each scenario was acted out between the same stations on the 8th Avenue IND to maintain consistency and limit the influence of other environmental factors.
- The victims were always male so gender differences did not influence helping behaviour.
- The female observers always took seats outside the critical area so differences in proximity to the incident did not influence helping behaviour / data recording.
Results:
QUAL DATA:
- Race (black/white) of every passenger in critical and adjacent areas
- Sex of every passenger in critical and adjacent areas
- Location of every passenger in critical and adjacent areas
- Whether passengers were seated or standing in critical or adjacent areas
- Race of helper (black/white)
- *Sex of helper
- *Location of helper – critical or adjacent area
- Latency of first helper’s arrival
- Latency of first helper’s arrival after the model had intervened
- No diffusion of responsibility found.
- *The ill victim received spontaneous help 95% of the time (62/65 trials) whereas the drunk victim was spontaneously helped 50% of the time (19/38 trials).
- *The median latency for ill victim trials (non-model condition) was 5 seconds whereas it was 109 seconds for drunk trials.
- *On 49/81 (60%) of the trials, spontaneous help came from more than one helper.
- *90% of first helpers were males.
- *Black victims received help less quickly than white victims, especially in the drunk condition.
- *In the drunk condition there was slight ‘same race’ helping.
- On 21/103 (20%) of the trials (with and without a model) 34 people left the critical area (where the victim was) after the victim collapsed.
- Help was slower to be forthcoming in the drunk condition: only 17% of the drunk victims were helped before the model stepped in, whereas 87% of the cane victims were helped before the model acted.
- Piliavin proposes the arousal: cost-reward model, a way of predicting the helping behaviour in emergency situations.
- Arousal is reduced by helping (directly/indirectly), leaving the scene of the emergency or rejecting the victim as undeserving of help.
- Ps were in an enclosed space so couldn’t escape.
- The proximity of other passengers made ps feel morally obliged to help.
- There were more potential helpers.
- The costs of helping were low because few risks were involved with others around to help as well.
Research Method/Design:
+ = the use of the New York subway, a natural environment, allowed observers to gather data in relation to how normal people show helping behaviour when in a restricted area, which is high in Ecol Valid.
+ = DC's unlikely (unless saw exp more than once) as didn't realise they were being observed for their helping behaviour.
+ = Lots of detailed qual data from observations can be gathered.
- = Not realistic, usually people can remove themselves from a situation they are uncomfortable with being in.
- = Less control than lab exp as field exp cant control all extraneous variables. (Hard to isolate the IV and keep all other variables that could affect the DV constant/controlled).
- = Not practical as:
- If the carriage where the incident occurred was particularly crowded/a passenger got in the way, the view of the observers may have been blocked so they could not see properly what was happening so relevant data could have been missed.
- There were too many passengers in the carriage so the experiment could not take place.
- Another emergency incident occurred so the experiment could not take place.
+ = Balanced racial mix of 55% white: 45% black.
+ = Large no of ps, 4450=representative of the target population of NY.
- = Sample was drawn from passengers/individuals on a subway so results cannot be generalised to passengers/individuals on other forms of transport/in other situations whose helping behaviour may be different.
- = Sample was drawn from the New York area, so ethnocentric, so results cannot be generalised to individuals from other areas/countries whose helping behaviour may be different.
- = Ps were passengers between 11.00 am and 3.00pm between April 15 to June 26, 1968, so results cannot be generalised to passengers who travel at other times/between other dates, whose helping behaviour may be different.
Sampling Method:
+ = No order effects
+ = Unlikely ps saw exp before.
+ = Wide range of people from all walks of life.
- = Ps may have seen exp before/realise aim, and produce DCs.
- = Ps on subway may have similar characteristics as have similar lifestyles, so results only generalisable to people who travel on the NY subway.
Reliability:
+ = STANDARDISED PROCEDURE, ps go through the same thing- get consistent results and therefore REPLICABLE.
+ = Because a total of 103 trials were conducted with similar results found in each trial condition results can be considered reliable.
+ = Because the 2 observers in each team always recorded findings for the same coding categories, interpretation of the categories will have been consistent so results can be considered reliable e.g. the same observer in each team recorded
race, sex and location of every rider seated or standing in the critical area.
+ = Because both observers recorded comments spontaneously made by nearby passengers, findings could be checked for reliability.
+ = Quant data recorded is easy to compare.
- = Because each of the 2 observers in each trial were recording different data, no checks were made to ensure inter-rater reliability e.g. timings of latency of first helper were not checked for accuracy.
- = Because there were 4 teams working separately application of the codings of one team may have differed from that of another e.g. gender is not always apparent!
- = No of people in the carriage can't be kept constant.
- = Low generalisability- not many places have no escape.
- = The situation, may on occasions, have resulted in the observers’ views being blocked leaving them unable to see the incident clearly so they could not record /missed key information/write incorrect info (e.g count ps twice), making findings unreliable.
Validity:
+ = Qual data- rich detailed data.
+ = High Ecol valid as study takes place in real life situation.
+ = Unlikely the ps realise the aim of the study unless they have seen it more than once.
- = Due to repetition of the study (between April 15 and June 26, 1968) there is a chance that some passengers may have witnessed the incident more than once and responded with DCs / SDB making the results invalid.
Usefulness:
+ = Study yielded a lot of detailed data.
+ = Highlights that differences can occur between lab and field studies.
+ = Proposes big model for the way people behave in emergency situations (Arousal cost-reward model)
- = Only tells us about a situation where people can't escape.
- = Low generalisability- not many places have no escape.
Ethical Issues:
+ = ps not physically harmed.
+ = Info on ps kept confidential- names and ages of ps not given, except for their race/sex.
- = No informed consent, invasion of privacy, stress, anxiety about seeing someone collapse.
- = Deception- ps thought the researches were fellow passengers, not people who were observing them for their helping behaviour.
- = Passengers who did not help/felt unable to help may have left the study feeling guilty or distressed.
- = No R to W- no escape for the ps, they are stuck on the carriage until the journey ends.
Improvements & Implications:
1a) Ethics:
Advertise on the New York subway prior to the study taking place, that an investigation on 'helping behaviour' is imminent, providing the dates and times that it will be conducted. The results of the study will be posted around the appropriate stations as a form of debrief.
- + = Debriefing may make ps feel better/salve their consciousness.
- - = Getting informed consent will reduce stress but may lead to invalid responses.
- - = Getting informed consent will lead to DCs/ SDB
1b) Ethics:
Conduct a study where the victim (drunk/ill) collapses in a more open space, where ps can escape from the situation if they wish. E.g a street
- + = Ps have R to W as they are not in a confined space and are able to escape from the situation of a person collapsing, they are able to leave it for other people t help, which is less distressing.
- -/+ = Diffusion of responsibility may be found, as passersby may think the victim WILL get help by someone else, even if they don't help.
- - = Still have DECEPTION- the passersby thinks someone is really collapsing, which may cause them DISTRESS also if they walk away.
- - = Practical problems are worse than Piliavin's as passersby are on the move (whereas in Piliavin they were stationary for the 7 min journey), so need to make their observations on their race, sex and location (in approx. meters away from the victim) a fast as they can.
2a) Method:
Conduct the same experiment but with cameras set up in the carriages already. Permission would have to be granted by carriage owner. They would be placed in the 4 corners of the carriage to give the best visibility. After the experiment had finished the researchers would view the recordings and collect the results for the race, sex etc of the passengers and helpers in the carriage.
- + = Will ensure nothing is missed by the observers and will be less affected by an increase in the no of people in the carriage (but in Piliavin's exp this would have been a practical problem as the observers might not have been able to see the ALL the passengers).
- + = The observers are not required, so less researchers are needed to collect the research.
- + = Less likely to get DC's as observers are not there to act suspiciously, so passengers not likely to realise the aim of the study, thus more likely to act normally, so natural helping behaviour observed (Piliavins exp was already high in ecol valid, this is just more).
- + = No longer need to rely on the momentary interpretations of the observers on the sex and race of the passengers as the recordings can be watched over to clarify the race and sex of passengers is noted correctly.
- + = Inter-rater reliability can be ensured as the recordings can be checked by as many researchers as needed.
- - = It is still possible that the cameras will not be able to give the details of the passengers, if for e.g they have their hood up- the camera wont be able to 'see' their race or sex.
- - = Ethical issues- no informed consent to be observed and recorded on camera. Ps also have no right to withdraw as they are not aware they are taking part in the study as it wouldn't be feasible to inform all ps that they have taken part in a study.
2b) Method:
Conduct the experiment in the same way as Piliavin, except at the end the observers ask any helpers/passengers Q's such as: 'Why did you decide to help/not help the man who collapsed?', as a form of an informal interview. The observer would use a tape recorder in their bag to record what the passengers say, as this wouldn't make it obvious to the passenger that they were being observed for a study (as writing this down in front of them would be obvious and it would not be practical to expect the observer to remember the replies the passengers gave to these Q's.
- + = You would obtain detailed, meaningful qual data on the reasons why the passengers did/did not help. This will increase VALIDITY.
- - = ps could lie about why they helped/didn't help in a socially desirably way that they would feel makes them look better/accepted for helping/not helping. This would decrease validity.
- - = Time-consuming to compare and analyse all responses.