Piliavin et Al SA Flashcards
Piliavin et al-subway samaritans-Psych being investigated-Bystander Apathy
Bystanders are ppl who are physically present at the scene of an incident but do not take an active part in helpings other in need. Interest in this field triggered by murder of American Kitty Genovese (1964). A NYT articles about her death claimed that 38 witnesses ignored her cries for help. Although there is controversy over the truth of events that evening, the article inspired much interest in the field of social psych, prompting development of many ingenious studies to explore bystander apathy, the term used to describe the lack of help offered in such situations.
Although some evidence suggests that bystanders may evaluate the character of a victim negatively and therefore lack compassion, a great deal of work suggests that the reason for ppls inactivity often has more to do w the behavior of ppl other than the victim.
Psychology being investigated-Diffusion of responsability.
When an incident is observed by 1+ person, there may be a diffusion of responsibility; the more ppl are present, the less personal responsibility each individual feels. This suggests that in situations w many ppl present the victim will be less likely to receive assistance as less responsibility means less guilt for ppl who choose not to help.
Background 1
Kitty Genovese’s death inspired a wave of studies initiated by Darley+Latanè, all aiming to understand not only the bystander effect but also factors that encourage ppl to play the “good Samaritan” and offer help to others need. E.g ppl who overheard an individual supposedly having an epileptic seizure via earphones were less likely to intervene when believing others were also available to help, compared with when they believed themselves alone (Darley+Latanè 1968).
Background 2
In addition, in 1969 Darley+Latané found that people in a group were less likely to help a person who had fallen over if the victim was a stranger compared with acquaintance. Furthermore, observing helpfulness of others (modeling) also appears to inspire samaritanims(Bryan+Test,1967).
Before the Piliavin study, the majority of social psychological research on helping behavior in emergency situations had taken place in lab settings (justification being higher degree if control, thus internal validity of research). However, Piliavin et al. wanted to see how bystanders responded to an incident in a real-world setting, leading to findings w higher eco validity.
In Brief
-mainly interested in how type+race of victims affected bystanders behavior in terms of frequency of helping responses, speed of help, and race of helper. Decided to stage an incident in which an ill/drunk victim (a confederate of the experimenter) would collapse on a subway train. They believed ill victims would receive more sympathy due to lack of personal responsibility for misfortune. Also believed risk of helping the drunk may be greater in terms of disgust or aggression possibility. In addition, they were interested in whether modelled help from a 2nd confederate would inspire more help from other passengers. Finally, they wanted to determine if grp size affected frequency + speed of help offered.
AIMS
Piliaving et al aimed to investigate factors affecting helping behavior on a NY subway train. Specifically, they wanted to see how the following factors affected help to a passenger who collapsed in carriage:
-type of victim (condition: drunk or ill)
-race of victim (black or white)
-modelled help provided by another passenger
-number of people in the carriage (grp size)
Research Methodology
The method used was a field exp as it took place on a 7.5 min express train ride between 2 NY stations (Bronx and Harlem) + researchers manipulated the variables.
Design and Variables
-Independent measures design (each scene was carried out in front of a different carriage full of passengers.
-Data collected aslo via covert observation:
–covert observer 1:recorded number of ppl in car+race, sex,location of every passenger in critical area. She also noted how many ppl assisted the victim, as well as race, sex,location.
–Observer 2: recorded race, sex, location of passengers in adjacent area, as well as time taken to assist victim.
–Both noted comments made by passengers + tried to elicit them.
Variables
The IVs were the condition of victim (drunk/ill), race of vitim (b or w), + how close the helpful model was to victim (critical and adjacent area), how quickly hep was offered, and grp size.
The DVs were: time taken for help to arrive (before+after modelled help) (keep in mind may not help), sex, race of 1st helper, movement of passengers out of critical area of carriage (i.e were collapse happened) and passengers spontaneous comments about the collapse.
Sample
Consisted of approx 4450 men + women. 45% of whom were black and 55% white, all whom were riding the 8th Avenue train in NYC on weekdays between 11am-3pm (15 april-26 June 1968). The mean num of people was 43, and in critical area where incident took place 8.5. Pps were an opportunity sample who did not give informed consent + were unaware they were taking part in a study.
Procedure
-A team of 4 students boarded a carriage via separate doors
-2 female observers took separate seats in adjacent area, while the 2 male students playing victim roles + model remained standing.
-The victim always stood centrally (by pole) in critical area. 70s into journey the victim collapsed n floor + laid on back looking at ceiling until received help.
-If no help was given by the time the train reached the next station, the model helped the victim to feet. The team would then re-board a train alongside route in opp direction, completing around 6-8 trials per day.
Victim part a
There was 4 teams of students, each with a victim aged 26 and 35 who dressed identically with the same old trousers, Eisenhower jacket, no tie.
-One victim was black, and 3 were white.
-38 trials of drunk condition + 65 ill.
-ill victim carried a black cane+ drunk carried a liquor bottle in a brown paper bag + smelt of alcohol.
-each student acting as a victim played both roles during course of study (yet one white drunk player felt too embrassed?)
Victim part b
All models were white males aged between 24-29. There were 4 conditions in which the model helped the victim to a sitting position+ stayed with them until the train stopped. A no model condition too. Order of conditions was randomised, model conditions were:
-1) critical area-early (stood in critical area; gave help after 70s
-2)Critical area-late: stood in critical area; gave help after 150s
-3)adjacent area-early: stood adjacent to critical; gave help after 70s
-4)adjacent area-late: stood adjacent critical area; gave help after 150s
Results 1
Main results for each aim of exp:
-Type of victim-A person appearing ill is more likely to receive help than one drunk (62/65 VS 19/38 of trials). The ill victim was helped 100% of time in condition with no model.
-Race of Victim: No tendency for same-race helping unless victim was drunk. The drunk white victim was helped 100% of time, while drunk black only 73%; and more frequently by black helpers than white. On 9% of trials with black victim, people left critical area, compared to 5% of trials with a white one.
-Modelled help-early models were more likely to elicit additional help than late models (4vs2).
-Group size-There was a weak positive correlation between group size+helping behaviour. Groups of 7 or more were faster to respons than groups of 3.
Results 2
Summary of quantitative results (% of trials on which help was given, by race ,condition of victim, and total num of trials in each condition:
White Victims
Cane Condition
No model: 100%
Number of trials run: 54
Model trials: 100%
Number of trials run (model): 3
Total number of trials: 57
Drunk Condition
No model: 100%
Number of trials run: 11
Model trials: 77%
Number of trials run (model): 13
Total number of trials: 24
Black Victim
Cane Condition
No model: 100%
Number of trials run: 8
Model trials: Not applicable (–)
Number of trials run (model): 0
Total number of trials: 8
Drunk Condition
No model: 73%
Number of trials run: 11
Model trials: 67%
Number of trials run (model): 3
Total number of trials: 14
Results 3
Adittionally, males were more likely to help than females. 90% of first helpers were male and 68% white (of helpers). In 20% of trials, ppl left critical area and a total of 34 ppl left.
Researchers reported few interesting findings from bystanders comments: Female comments include: “It’s for men to help him”, “I wish I could help him-I’m just not strong enough” “I never saw this kind of tihng bfore, Idk where to look” “You feel so bad that you don’t know what to do”:
Conclusions
Research did not support Darley+Latanès diffusion of responsibility hypothesis in a real world setting. Despite collapses taking place on a busy train, help was offered quite frequently & quickly.
-Piliavin et al. proposed that witnessing an emergency situation leads to the creation of an emotional arousal state, which a person then wishes to rid themselves of.
-The srength + nature of this arousal depends on factors such as the level of empathy levels a person feels for a victim, distance from situation, and amount passed without intervention.
To reduce this arousal, a person can offer help directly, fetch help, leave the scene, or decide that the victim is undeserving of help. Piliavin et al. suggested that people weigh up these options using a cost-reward matrix.
Costs may be related to either helping (e.g effort, disgust) or not helping (feeling) of self-blame or negative perception of others), as well as received rewards.
Evaluation-ethical issues-informed consent -
One key criticism is the lack of informed consent. Pps were unaware that the study was taking place so this was unable to be obtained. Therefore pps couldn’t give permission to take part in the potentially distressing situation of a human collapsing in a subway carriage (might have causes psych harm). Furthermore, this also meant that they weren’t given opportunity to consent to their data being used in study. Some comments were recorded and published in the journal article-might have caused some embarrassment.
Ethical concerns: Participants were not aware they were in an experiment, meaning they could not give consent or withdraw, reducing the ethical validity.
Evaluation-Ethical issues-Debrief-
Another weakness is that, due to subways nature + large num of pps leaving train at their stop, no debrief could take place. This meant that the pps could not have aims of the exp explained to them, or have any qs answered to ensure they left the train un the same psych state in which they arrived. Unfortunately; the lack of debrief meant that some pps may have left the train feeling shocked at collapse of ill/drunk man, or potentially guilt at not having offered help.
Evaluation-Methodological issues-Reliability-Standardisation+
A strength is that researchers controlled several aspects of the procedure. The victims always wore the same clothing + the collapse always took place when the train passed the 1st station on the subway route. This ensured that the procedure could potentially be repeated to test for whether helping behaviors towards the ill and drunk victims were reliable from trial to trial.
Evaluation-Methodological issues-Validity-Confounding Variables-
A weakness is that only 1 black victim was used in all trials. With only 1 person acting in this condition of the study, it’s diff to conclude that helping behavior towards this individual was due to race as opposed to personality factors, or diffs in his acting behavior in cane/drunk conditions (some students felt embarrassed to act drunk-especially a white one). This meant tat findings ab helping behavior for the IV of race may lack validity.
Evaluation-Methodological issues-Validity-Lack of controls-
Another weakness is that the location + activity of ppl in the carriage wasn’t controlled. Some had been reading a newspaper, or distracted in another way, not seeing victim collapse. This would have affected validity of study’s findings, as their reason for not helping may not have been a conscious choice.
Evaluation-Methodological issues-Validity-Qualitative data
A strength was that qualitative data was also collected in this study, allowing for further insights into bystanders’ justifications for lack of helping behavior (i.e the comments made+ reactions). This qualitative data allowed for Piliavin et al. to draw more meaningful conclusions abt the justifications ppl made when deciding to not help the victim (i.e through their cost-reward matrix).
Evaluation-Objectivity + subjectivity-Quantitative data
Strength was collection of mainly quantitative data, allowing researchers to objectively compare helping behavior between drunk + ill(caned) condition (e.g time taken to receive help was recorded. which is a measure free from researcher bias + need for subjective interpretation, allowing for valid comparison to be made between conditions.
Evaluation-Generalizations + Ecological validity-generalizing beyond the sample
Strength was large sample size (estimated 4450 ppl witnessed the victim incidents over time, comprising ab 45% black and 55%white individuals-pretty balanced. It could be argued that such a large sample may be representative of NY in general and the findings may be applied externally to city’s population.
Yet, a weakness is that research also lacks population validity, as pps were likely to consist predominantly of commuters on a specific NY subway route (and they may have even encounter the exp 1+ times-order effects, demand charasteristics, pointeless…). Urban residents may experience more deindividuation (loss of individuality + personal responsability) due to their busy surroundings + individualist culture. Therefore this limits the extent to which findings on helping behavior can be applied to people living in rural areas, smaller cities /collectivist cultures.
Evaluation-Generalizations + Ecological validity-Generalising to everyday life+
A strength was that it was a field exp, taking place on a real life subway route. The passengers witnessed an accident on theiur everyday commute, thus unaware of observations. hence demand characteristics avoided + pps showed genuine responses to supposed real life situations. This gives it a high eco validity, allowing researchers to draw accurate conclusions about their real life behavior in emergency sits (+ more cause effect relationship can be established.
Issues and Debates- Situational explanations
There was evidence to support a sit. explanation for the helping behavior in the study-the situation of an ill person falling appeared to rigger helping behavior in nearby observers. On every trial without model, someone offered to help the il victims, thus demonstrating that the specific situation a person found themselves in was he main motivator behind the subsequent behavior.
Applications to everyday life
Training + Education
Educating students on the cognitive processes ppl go through in emergency sits may help them to change their (pre-defined?) responses + make them more likely to help. Understanding the cost reward matrix would help ppl recognize their own bias in interpreting heightened emotions (e.g labelling an emotion as disgust `when a drunk falls over). This understanding may help them override negative or non-altruistic feelings in order to offer help to a victim, instead of ignoring them!
Demographics
Victims were only men.
Issues and Debates- Individual explanations
On the other hand, the majority of first helpers (90%) of any victims were male, supporting individual explanation for being behavior. Piliavin et al. suggested that people use a cost-reward matrix to determine whether to help: the perceived cost of helping may be lower for males and diff in each person, depending on personality and personal circumstances, highlighting an individual explanation for diffs in helping behavior.
Same race helping was mainly seen in….
the black drunk condition
