Manipulation of Social Identification and Group Norms Variables
The total sample size was N= 60 health practitioners with mean age of 40.74 years (SD= 2.57); 25 males, 33 females and 2 non-binary. The participants were a staff of a large Australian government organization with more than 2500 employees. He staff were sent an email containing an invitation link to participate in the study. The research participants clicked through the link to read and respond to the manipulations and scenario online. A priori power analysis was performed and it suggested that a sample size of 60 would be appropriate to test the expected effects, and consequently the experiment stopped recruiting participants once 60 had completed the experiment. Participation was voluntary as per IRB regulations and the respondents were allocated randomly to one of the six experimental conditions.
The research used a 2 (social identification condition) × 3 (Group norms) between-groups factorial design. Manipulation of the group norms circumstances were through ‘social identification condition’ variables with two levels namely High/Low, and the ‘group norms’ variable with three levels namely Low/Control/High. The response variable was donation behavior and in such, the perceived group norms levels and social identification were experimentally manipulated and behavioral intention of donation was measured as a consequence.
All the variables in this case study included participants of the large Australian government organization responding to the questionnaire. We had Social Identification Manipulation in which the variables of interest were Low and High Social Identification. The respondents were to list 3-5 attributes (values, lifestyle, experiences, social concerns and relationships) that makes them different from an average health practitioner. We also had Group Giving Norms Manipulation in which respondents would be categorized as either low giving norm, control condition and high giving norm.
A 7-point Likert scale was used in manipulation check whereby a value of 7 indicated extremely highly health practitioner (high social identification) and a value of 1 indicated extremely low health practitioner (low social identification). Low social identification showed that the health practitioner distanced themselves from group norms while high social identification scores indicated that the health practitioner adhered to group norms.
A 7-point Likert scale was also used to measure the donation intention with the value of 7 being extremely a high donation intention while a value of 1 being extremely a low donation intention.
A consent form was sent to the participants to ask for their permission of voluntarily participation in the research. The consent form had instructions, an overview of the research as well as the questions they were going to complete. The research participants were then provided with a scenario of the six which was applicable to their experimental condition. The participants were the expected to respond to the two questions after reading the scenario presented to them.
Effect of Social Identification and Group Norms on Likeliness/Intention of Donation-Manipulation Check
A 2 (social identification: low, high) × (group norms: low, neutral, high) between-group factorial ANOVA analysis was performed for manipulation check of how they affect health practitioners’ intentions to make donations. It was found that social identification was insignificant F (1, 54) = .022, p = .881, = .017. The participants in the high social identification recorded a high donation intention (M= 3.70, SD= 2.09) compared to participants in the low social identification group (M=3.67, SD= 1.95).
Results of Manipulation Check
The results indicated that group norms was significant F (2, 54) = 130.37, p = .000, = 96.82. Participants in high group norm showed a higher donation intention (M= 6.05, SD= 0.76), followed by participants in the neutral group norm (M= 3.30, SD= 1.03) and lastly participants in the low group norm indicated the least donation intention (M= 1.70, SD= 0.80).
We also had the interaction effects and this is the effect of social identification combined with group norm on the health practitioners’ donation intention. The results indicated that that interaction effect was insignificant in determining the likeliness of donation F (2, 54) = 2.18, p = .12, = 1.62.
The manipulation check process was successful in determining how group norms and social identity likely influence the health practitioners’ donations to charity. It was found that social identification and the interaction between social identification and group norms were insignificant while group norms was significant at determining the donation intention.
Effect of Social Identification and Group Norms on Charity Donation.
A 2 (social identification: low, high) × (group norms: low, neutral, high) between-group factorial ANOVA analysis was also performed to determine how the two predictor variables affect charity donation among healthcare practitioners. It was found that social identification was significant F (1, 54) = 7.33, p = .009, = 153.60. The participants in the high social identification category recorded a high donation amount (M= 7.23, SD= 6.64) compared to participants in the low social identification group (M= 4.03, SD= 4.23).
The results again indicated that group norms was significant F (2, 54) = 12.41, p = .000, = 260.02. Participants in high group norm showed a higher donation amount (M= 9.35, SD= 6.61), followed by participants in the control group norm (M= 5.40, SD= 4.77) and lastly participants in the low group norm indicated the least donation amount (M= 2.15, SD= 2.98).
We also had the interaction effects and this is the effect of social identification combined with group norm on the health practitioners’ donation amount. The results indicated that that interaction effect was insignificant in determining the donation amount among healthcare practitioners F (2, 54) = 3.51, p = .037, = 73.55.
From this analysis, it was found that both social identification, group norms and the interaction between social identification and group norms were all significant in determining charity donation amount. This is unlike in the manipulation check results in which social identity and the interaction effect were not significant in determining donation intention.
The purpose of the study was to investigate the independent effect of social identification and group giving norms on the donation behavior of health practitioner. In addition, the study was also aimed at determining how the interaction between the two independent variables influences the donation behavior of health practitioners.
The research hypothesized that a higher giving norm tendency will lead to an increase in donation while consequently, a lower giving norm tendency will reduce donation. This hypothesis was supported by both the manipulation check and donation amount results whereby the high group norm results indicated both high donation amount and high donation intentions while low group norm results indicated both low donation amount and low donation intentions. This agrees with the findings of Kashif et al. (2015) who stipulated that group norms that are pro-giving are associated to high chances of donation.
Results of ANOVA Analysis for Charity Donation
Another hypothesis of the study was that the effect of group norms on the donation behavior of health practitioners will be significantly large especially when the participant had a higher social identification. This is about the interaction effect and the results confirmed to this hypothesis whereby high group norm had the highest mean value followed by control and then lastly low group norm. These findings agree with the work done by Terry & Hogg (1996) who found that persons with higher social identification confirms more to group norms. Therefore, high social identification will lead to high group norms (high confirmation to group norms) and hence health practitioners in high giving group norms will have high donation behavior.
The research also had some predictions and in this case, the study made a prediction that a higher donation amount would be made when the respondent has high social identification- highly health practitioner than when the participant has a low social identification- distanced from the group. This prediction was true because the results of the study supported it because high social identification had the highest mean donation amount. This is because health practitioners are considered generous and according to Louis et al. (2007), group members psychologically will define themselves in terms of that group norms.
Furthermore, the study made a prediction that the amount of donation will be higher when the group norm was depicted to be more supportive of giving compared to the control condition. On the other hand, it was also predicted that the donation amount will be low in cases where the group norm was depicted as less supportive of giving compared to the control condition. The results of the study confirmed the predictions to be true whereby high group norms indicated a higher donation amount while a low group norm indicated a lower donation amount. This again agrees with the work done by Louis et al. (2007).
The results also showed that the group giving norms, social identification and the interaction between group giving norms and social identification were significant at determining or explaining the donation behavior of the health practitioners. This agrees with the work done by (Martin & Randal, 2008; Louis & Blackwood, 2012) who stated that people have higher chances of donating to their respective groups when social identification is made salient.
The strength of this study was that it used appropriate participants. The research used health practitioners who are mostly involved with charity matters and donations and hence this played a significant role in ensuring that the results are valid and reliable. This strength also addressed the existing gap in the literature which had been using students as participants and as it is known, students are hardly defined by donation characteristics.
The limitation of the study lies in the sample size. A sample size of 60 is not large enough to have a fair representation of the population. The population of this study 2500 and hence a sample of 60 have higher chances of under-representing the population. This consequently has higher chances of leading to sample size error which in turn might affect the reliability and validity of the results.