S - Levine Flashcards
Aim
To determine if a city’s tendency to offer non-emergency help to strangers is stable across situations over a wide range of cultures
Background
Studies conducted in several different countries (including the USA, Saudi Arabia and Sudan) have found that people living in urban areas tend to be less helpful than those in rural settings (Hedge & Yousif, 1992; Yousif & Korte, 1995).
Virtually all of the studies of community differences in helping have focused on the single variable of population size, most often testing the hypothesis that the tendency to help strangers declines as the size of the city increases. Steblay (1987) found general support for this hypothesis with the decline in helping rate beginning at populations of 300,000. She also found that urban environments of 300,000 people or more and rural environments of 5,000 people or less were the worst places if one was looking for help.
A major cultural difference in helping behaviour is the difference between collectivism and individualism. Collectivists attend more to the needs and goals of the group they belong to, and individualists focus on their own selves. Therefore, collectivists would be more likely to help in group members, but less frequent than individuals to help strangers (Triandis, 1991)
Method
This was a cross-cultural quasi experiment carried out in the field that used an independent measures design.
The field situation was 23 large cities around the world including Rio de Janeiro (Brazil), Calcutta (India), Madrid (Spain), Shanghai (China), Budapest (Hungary), Rome (Italy), New York (USA) and Kuala Lampur (Malaysia). The study should be considered as a quasi- experiment because the independent variable – the people in each city – was naturally occurring.
The experiment measured, through the use of a series of correlations of co-variables, helping behaviour in three non emergency situations: whether the victim dropped a pen whether the victim had a hurt/injured leg whether the victim was blind and trying to cross the street.
The dependent variable (DV) was the helping rate of the 23 individual cities (calculated to give each city an Overall Helping Index).
The three measures of helping were correlated with statistics reflecting population size, economic well-being, cultural values (individualism-collectivism, simpatia) and the pace of life for each of the 23 locations.
Sample
Participants in this study were large cities in each of 23 countries – in most cases the largest in each country i.e. individuals in each of these cities at the time of the experiment.
Each of the three helping measures and the walking speed measure were administered in two or more locations, in main downtown areas, during main business hours, on clear days, during the summer months of one or more years between 1992 and 1997.
For the dropped pen and hurt leg situations, only individuals walking alone were selected. Children (younger than 17 years old), and people who were physically disabled, very old, carrying packages etc (i.e. those who might not be fully capable or expected to help) were excluded.
Participants were selected by approaching the second potential person who crossed a predetermined line
Procedure
Data was collected by either interested, responsible students who were either travelling to foreign countries or returning to their home countries for the summer, or by cross-cultural psychologists and their students in other countries who volunteered to assist the authors.
All experimenters were college age and dressed neatly and casually. To control for experimenter gender effects and to avoid potential problems in some cities, all experimenters were men.
To ensure standardisation in scoring and to minimise experimenter effects:
all experimenters received both a detailed instruction sheet and on-site field training for acting their roles, learning the procedure for participant selection and scoring of participants the experimenters practised together no verbal communication was required of the experimenter.
The three helping measures were:
Dropped pen. Walking at a carefully practised, moderate pace (15 paces/10 seconds), experimenters walked toward a solitary pedestrian passing in the opposite direction. When 10 to 15 feet from the participant, the experimenter reached into his pocket and accidentally, without appearing to notice, dropped his pen behind him, in full view of the participant, and continued walking past the participant. A total of 214 men and 210 women were approached. Participants were scored as having helped if they called back to the experimenter that he had dropped the pen and/or picked up the pen and brought it to the experimenter.
Hurt leg. Walking with a heavy limp and wearing a large and clearly visible leg brace, experimenters accidentally dropped and unsuccessfully struggled to reach down for a pile of magazines as they came within 20 feet of a passing pedestrian. A total of 253 men and 240 women were approached. Helping was defined as offering to help and/or beginning to help without offering.
Helping a blind person across the street. Experimenters, dressed in dark glasses and carrying white canes, acted the role of a blind person needing help getting across the street. (The canes and training for the role were provided by the Fresno Friendship Centre for the Blind). Experimenters attempted to locate downtown corners with crosswalks, traffic signals, and moderate, steady pedestrian flow.
They stepped up to the corner just before the light turned green, held out their cane, and waited until someone offered help. A trial was terminated after 60 seconds or when the light turned red, whichever occurred first, after which the experimenter walked away from the corner. A total of 281 trials were conducted. Helping was scored if participants, at a minimum, informed the experimenter that the light was green.
Results
- Although statistically insignificant, there was a small relationship between walking speed and overall helping, with participants in faster cities somewhat less likely to help.
- More individualistic countries showed somewhat less overall helping and less helping in the hurt leg situation than collectivist countries, but none of the correlations reached significance.
- There was no relationship between population size and helping behaviour.
- The two community variables of economic productivity and individualism-collectivism and walking speed were highly intercorrelated: economic productivity was positively correlated with individualism and negatively correlated with walking speed i.e. faster places had stronger economic productivity. Individualism was also negatively correlated with walking speed i.e. faster places were more individualistic.
- Simpatia countries (Brazil, Costa Rica, El Salvador, Mexico and Spain) were, on average, more helpful than non simpatia countries.
- Overall, a city’s helping rate was relatively stable across all three measures.
ping measures. Results showed that the most helpful cities/ countries were (1) Rio de Janeiro (Brazil), 93%, (2) San Jose (Costa Rica), 91% (3) Lilongwe (Malawi), 86%. The least helpful cities/countries were
(23) Kuala Lampur (Malaysia), 40% (22) New York (USA), 45%, (21) Singapore (Singapore), 48%
Conclusions
The helping of strangers is a cross-culturally meaningful characteristic of a place.
There are large cross-cultural variations in helping rates.
Helping across cultures is inversely related to a country’s economic productivity.
Countries with the cultural tradition of simpatia are, on average, more helpful than counties with no such tradition.
Although faster cities tend to be less helpful than slower cities, the link between economic health and helping is not a by-product of a fast pace of life in affluent societies.
The value of collectivism-individualism is unrelated to helping behaviours.
Method Evaluation
Able to compare levels of helping behaviour across diverse cultures; individualistic, Simpatia, and collectivist – helping to establish INTERNAL VALIDITY as measuring the aim; The rates of helping behaviour across 23 different countries.
Avoids ETHNOCENTRIC BIAS;
23 different cities are studied and rated on a 10- point scale (1 = the most collectivistic, 10 = the most individualistic), this helps to make the results generalisable to each type of culture.
High ECOLGOCAL VALIDITY- As the p’s fall naturally into each condition of the I.V. the level of helping behaviour displayed to each victim should be natural.
Reduced INTERNAL VALIDITY
Lacks control over extraneous variables such as individual differences.
Levine has no control over the individual differences of the participants in each city; these individual differences may effect their decision to help the victim not their cultural values.
Data
Easy to analyse and make comparisons between the level of helping behaviour across all three victim conditions and the 23 countries studied.
The correlation coefficient allows the type and strength of relationship to be established between the level of help and the community variables studied. This identifies which of the community variables has the strongest possible influence and can help to establish in which area to carry out further research.
This type of data does not provide insight into why each culture helped more or less.
However, the fact that a number of community variables were studied helps to provide insight into why some cultures help more than others.
Ethics
No protection from harm;
Witnessing someone in need may cause some degree of distress especially as p’s are close to the incident and so may feel obliged to do something.
No consent gained;
P’s were not aware that they were taking part in research as they were crossing a road in an down town area of a city.
Reliability
The procedure used by all the researchers was standardised ensuring that all p’s in every city were tested in the same way.
Experiments were trained;
1) all experimenters received both a detailed instruction sheet and on site field training for acting their roles, learning the procedure for participant selection and scoring of participants.
2) The experimenters practiced together.
3) No verbal communication was required from the experimenter.
Lack of control over the environment may have meant that experiments could not always act in the same way across the three victim conditions.
Therefore reducing consistency of procedure and measurement of help taking.
Validity
Using pace of walking as a measure of pace of life is innovative and probably as accurate a measure as any.
Extraneous variables, such as gender, were controlled by having all-male experimenters, therefore increasing the internal validity of the study.
Reduced INTERNAL VALIDITY
Lacks control over extraneous variables;
Environmental variables in the downtown areas of each of the 23 cities may have effected levels of helping behaviour, for example amount of traffic, amount of people around causing diffusion of responsibility.
Sampling Bias
s -23 countries from different cultures were used – making the sample less ethnocentric.
w- Possible bias within the sample;
There may have been more individualistic cultures than simpatia.
Ethnocentrism
To some extent could still be ethnocentric;
Helping behaviour was defined by westernised ideas of helping.
