Our Research posts are about the latest academic research being done in the School of Economics. This week:
The characteristics of volunteers in South Africa
Volunteers are people who spend their time for the benefit of others. Their work is quite important in a time when social safety nets are weak and there are ever increasing demands on welfare organisations. It is estimated that approximately 1.2 million South Africans participate in volunteer activities. This amounts to millions of hours and billions of Rands of value had the activities been compensated. In a recent ERSA working paper Dr Ferdinand Niyimbanira and Prof Waldo Krugell asked, who are these good Samaritans? What are the characteristics of these volunteers and how are these related to the number of hours that they spend as volunteers? Are they looking for psychological reward, or business contacts and skills?
The paper uses data from the South African Volunteer Activities Survey and examines the links between individuals’ resources and volunteerism.
A quick description of the data already shows an interesting fact: Blacks volunteer on average almost double the number of hours that the other population groups do, but the analysis shows that this cannot be explained by individual ‘assets’ such as gender, level of education, work status or income.
A regression model of the predictors of the number of hours that people volunteer shows two predictors that are statistically significant. First, compared to the Black and Coloured population groups, being White is negatively and significantly associated with hours volunteered. In fact, Whites volunteer 8.5 per cent fewer hours compared to the Black and Coloured population groups. Second, when “others” received the benefit of the volunteer work, compared to household members, family or friends, there is a positive and significant association with hours volunteered – 9.4 per cent more hours are volunteered.
The other predictors are not statistically significant but might be considered indicative of the direction of relationships. Age is negatively associated with hours volunteered. People who are married or living together are compared to those that are single, divorced or widowed. Being married is negatively associated with hours volunteered, but living together has a positive association. Compared to people who did not complete secondary education, those that did, and those that also completed tertiary education, volunteer more hours. People who reported that they received something in return for their volunteer work (cash, expenses, or in-kind payments), volunteered more hours than those who did not receive anything in return. Higher incomes are positively associated with hours volunteered, but not significantly so.
If differences in individual resources and income do not matter for volunteering, is there any evidence of investment-type behaviour? A more detailed breakdown of the average number of hours volunteered per population group by employment status and what they reported they received in return for the volunteer work shows an interesting story for the Black population. The Coloured and White groups for the most part received nothing for their volunteer work. This is largely true for the Black population as well, but a small number of them reported receiving out of pocket expenses, food or transport. In these cases the unemployed volunteered significantly more hours than the average. However, the number of people involved is small and receiving rewards for volunteering do not drive the differences in volunteering rates between the different population groups. As for the investment model, there were some unemployed Black volunteers (46 respondents) that indicated that they gained experience and skills – they also volunteered substantially more hours (53 hours on average).
A final question may be whether the differences in volunteering rates can be explained by the nature of the volunteer work. The regression results showed that when “others” received the benefit of the volunteer work, compared to household members, family or friends, there is a positive and significant association with hours volunteered. It is however not the case that the relatively less educated, more unemployed, poorer Black population is mainly helping out friends and family in lieu of better social safety nets.
As such this paper shows the limits of large sample survey research and serves as a call for further research. There is a need to further examine the causes of volunteerism in terms of subjective dispositions, life course and social context using in-depth interviews, focus groups and life histories.