Data was collected from employees working for a large investment banking firm in Brisbane, Australia. 335 questionnaires were distributed to all the employees, but only 163 questionnaires were completed voluntarily and returned. The respondents were 129 females and 34 males, with a mean age of 37. 35 years (SD = 6. 20).
Measures
Respondents were required to report their age, then, their responses on three main variables were required as the following.
Perceived peer pressure. Three items from Peer Pressure about Humanitarian Behavior Scale (Rafferty, 2009) were used to examine the extent to which the presence of peer pressure influences a person’s behavior. It asks about the frequency of perceived peer pressure on a 5-point Likert scale (1 = never, 3 = sometimes, 5 = always), with statements indicating that the person who perceived peer pressure more often, he would contribute more in humanitarian behavior. Internal reliability of this scale was very high, Chronbach’s a = . 90.
Trait selfishness. Four items from Selfishness Scale (Loxton, 2009) were used to assess the extent of a person being self-absorb and having empathic concern for others. It asks about the extent of agreement on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree), indicating that person who rated higher, he would be less likely to participate in humanitarian behavior. Internal reliability of this scale was moderate high, Chronbach’s a = . 82.
Participation in humanitarian behavior. Three items from Participation in Humanitarian Behavior Scale (Iyer & Louis, 2010) were used to evaluate the extent to which the respondents participated in humanitarian activities and behaviors, such as donate to the poor’s and help communities affected by disaster voluntarily. It asks about the extent of agreement on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree), indicating how likely they took part in this behavior. Internal reliability of this scale was very high Chronbach’s a = . 91.
Procedure
Questionnaires were distributed to employees from an investment banking firm, only 163 surveys were voluntarily completed and returned. In the survey, respondents were asked to report their age and to respond 10-item with Likert scale about how likely they perceived peer pressure, to what extent they agreed with their selfishness, and how likely they participated in humanitarian behavior.
Results
A standard multiple regression analysis was conducted. The two predictors were perceived peer pressure and trait selfishness. Participation in humanitarian behavior was the criterion. Table 1 shows a negative but small correlation between selfishness and perceived peer pressure, r (161) = – . 30, p < . 001, where collinearity would not be a problem, indicating that as selfishness scores increase, perceived peer pressure scores will decrease. Descriptive statistics and intercorrelations between variables are shown in Table 1.
Table 1.
Descriptive statistics and intercorrelations between all variables in the model.
M (SD)
Participation in Humanitarian Behavior
Perceived Peer Pressure
Trait Selfishness
Age
Participation in Humanitarian Behavior
4. 06 ( . 57)
–
. 38***
– . 32***
. 65
Perceived Peer Pressure
3. 37 ( . 54)
–
– . 30***
– . 15*
Selfishness
3. 44 ( . 94)
–
– . 07
Age
37. 35 (6. 20)
–
*** p < . 001, ** p < . 01, * p < . 05
In a model that age was included as a control variable, high validity was found, perceived peer pressure and selfishness were both correlated with participation in humanitarian behavior. Positive correlation was found between perceived peer pressure and levels of participation in humanitarian behavior, r (161) = – . 38, p < . 001, indicating that as perceive peer pressure increase, levels of participation in humanitarian behavior increase. Oppositely, negative correlation were found between trait selfishness and level of participation in humanitarian behavior, r (161) = - . 32, p < . 001, indicating that as trait selfishness increase, levels of participation in humanitarian behavior decrease.
Perceived peer pressure and trait selfishness explained a significant 20% of the variability in participation in humanitarian behavior, R = . 45, R2 = . 20, R2adj = . 18, F (3, 159) = 13. 21, p < . 001. A significant 9. 73% of the variability in participation in humanitarian behavior was uniquely explained by perceived peer pressure, sr2 = . 10 β = . 33, b = . 35, SE = . 08, t (159) = 4. 39, p < . 001, and a significant 4% of the variability in participation in humanitarian behavior was uniquely explained by trait selfishness, sr2 = . 04 β = - . 21, b = - . 13, SE = . 05, t (159) = -2. 82, p = . 005. Perceived peer pressure explaining 9. 73% of the variability in participation in humanitarian behavior was the most important predictor. There was 6. 20% of shared variability in this model.
To test the third hypothesis, a moderated multiple regression analysis was conducted. The two original predictors, perceived peer pressure and trait selfishness were mean centered on zero. At step one, the mean centered scores on perceived peer pressure and trait selfishness were entered as predictors of participation in humanitarian behavior, the criterion.
An interaction variable was calculated by multiplying the two mean centered scores of the two predictors in model one. The interaction term was calculated using mean centered scores to reduce the multicollinearity between the original predictors and the interaction variable. At step two, the interaction was entered as an additional predictor to determine whether the interaction of the two predictors explained an additional variability in participation in humanitarian behavior, which was beyond what the two predictors could explain additively.
At step one, perceived peer pressure and traits selfishness explained a significant 19% of the variability in participation in humanitarian behavior, R = . 44, R2adj = . 18, F (2, 37) = 18. 78, p < . 001. At step two, after the interaction was added as a third predictor, an additional 3. 50% of the variability in the criterion was explained. This increment was significant, R2change = . 04, F (1, 159) = 7. 28, p = . 008, indicating the interaction of perceived peer pressure and trait selfishness improved the prediction of participation in humanitarian participation, which was beyond what the additive model found at step one. Generally, at step two, a significant 22. 60% of the variability in participation in humanitarian behavior was predicted by perceived peer pressure, trait selfishness and their interaction, R = . 48, R2adj = . 21, F (3, 159) = 15. 44, p < . 001.
Figure 1. Predicted participation in humanitarian behavior scores as shown a function of selfishness and perceived peer pressure. NB: High and low values were +1 SD and -1 SD from the mean respectively.
The significant contribution of the interaction was followed up by performing a simple slopes analysis in Figure 1. High values of perceived peer pressure were calculated by subtracting one SD from the mean-centered scores on perceived peer pressure. Low values were calculated by adding one SD to mean-centered scores.
When perceived peer pressure was high, trait selfishness was not a significant predictor of participation of humanitarian behavior, β = – . 07, b = – . 04, SE = . 06, t (159) = – . 69, p = . 491. However, when perceived peer pressure was low, trait selfishness was a significant negative predictor of participation in humanitarian behavior, such that as trait selfishness scores increased, participation in humanitarian scores decreased, β = – . 23, b = – . 14, SE = . 04, t (159) = -3. 11, p = . 002.
Discussion
The aim of current study was to examine the predictive validity of the measure of perceived peer pressure, in relation to a measure of trait selfishness. According to the results, first hypothesis that selfishness is correlated negatively with perceived peer pressure was supported. People who often perceive peer pressure, they tend to know and aware of others’ difficulties and sufferings, being sympathy for others, they would be willing to help the communities without any pay back to help others feeling better. These people would not be selfish, because people who are selfish, they tend to ignore things happen around them and they only concern about themselves. Therefore, people who perceive more peer pressure are less likely to be people who are selfish.
Regarding to the second hypothesis that in a model included age as a control variable, selfishness and perceived peer pressure are associated with participation in humanitarian behavior, was supported. Specifically, there is a positive relationship between perceived peer pressure and participation in humanitarian behavior. People who perceived more peer pressure tends to pay more attention to humanitarian disaster, and they also tend to sympathize on the victims, so they are more likely to volunteer themselves to help the victims recover more quickly from disasters by donating money or clothes. This is consistent with the past research done by Cohen and Hoberman (1983) where the occurrence of appropriate peer social pressure can have certain influence on a person’s behavior. Besides, Lim (1997) also mentions that peer pressure is a factor that less likely to be assumed as a cause of lower levels of humanitarian behavior. So, peer pressure does have a certain influence on humanitarian behavior.
Furthermore, there is a negative relationship between selfishness and participation in humanitarian behavior. People who are selfish, are more likely to focus on their goals and desires, so they are less likely to pay attention on the other’s needs and interests (Eysenck, 1967), and so, they are less likely to be aware of humanitarian issues and disasters to participate in humanitarian behavior. This is consistent with the research by Eaves, Eysenck, and Martin (1989) where people scores higher on selfishness less likely engaging in humanitarian behavior, and by Furnham, Petrides, Jackson, and Cotter (2002) where selfishness has been consistently associated with low levels of humanitarian behavior.
Perceived peer pressure is found to be a more important predictor than trait selfishness is. Obviously, perceived peer pressure turns out to predict the humanitarian behavior better, it might due to humanitarian behavior could be seen based on how much does a person perceive peer pressure from others, while trait selfishness occurs depends on the situation. As previously identified, selfishness does not have the negative inferences for humanitarian behavior (Perkins & Corr, 2000; Smillie, Yeo, Furnham, & Jackson, 2001), the relationship between selfishness and humanitarian behavior would depend on different factors.
Concerning to the third hypothesis, perceived peer pressure moderated the effect of selfishness on humanitarian behavior was supported. The interaction of perceived peer pressure and selfishness improved the prediction of participation in humanitarian participation. When perceived peer pressure was low, trait selfishness predicted negatively the participation in humanitarian behavior, however, when perceived peer pressure was high, trait selfishness did not predict participation of humanitarian behavior, this means perceived peer pressure has an effect on the relationship between selfishness and humanitarian behavior, selfishness decreases humanitarian behavior only when the perceived peer pressure is low.
From this research, it found that peer pressure does have an effect on selfishness to predict humanitarian behavior. Selfishness is a stable personality trait which is hardly to be changed (Eaves, Eysenck, & Martin, 1989), but when it comes with peer pressure, it would be reduced or affected, and turn out people are selfish but often perceive peer pressure, they are likely to engage in humanitarian behavior as compared to those who perceive peer pressure lesser.
Questionnaires in this research was supported by past research but not a new formed questionnaires, it is good to check if the results in current study is consistent with past researches. Besides, as the participation of current study was in voluntarily basis, respondents tended to participate willingly, so their answers would be honest and frank compared to those participation involuntarily.
However, sample in current study was not randomized, data was collected from a certain group only, so results would not be able to be generalized to other sample group. Besides, questionnaires were too short, 10-item questionnaire for three variables would not enough to identify the relationships and effect between the variables. Therefore, more items should be added in the future research to identify the effects and relationships between variables efficiently. Besides, different items which is unrelated to the variables should be added to prevent from respondents being able to identify the aim of research, result would be affected.
