the team
cape town

This article is taken from the August 2004 Phatlalatsa newsletter


A nation of givers?

Strategy & Tactics was commissioned by the Centre for Civil Society (CCS), National Development Agency (NDA) and SA Grantmakers Association (SAGA) to undertake a representative national sample survey to measure both attitudes to giving and giving behaviour itself. The survey is one component of a larger project focusing on resource mobilisation for poverty and development that covers state expenditure, overseas development assistance, corporate giving, as well as individual-level giving focusing on religious groups and poor communities.


Strategy & Tactics was commissioned by the Centre for Civil Society (CCS), National Development Agency (NDA) and SA Grantmakers Association (SAGA) to undertake a representative national sample survey to measure both attitudes to giving and giving behaviour itself. The survey is one component of a larger project focusing on resource mobilisation for poverty and development that covers state expenditure, overseas development assistance, corporate giving, as well as individual-level giving focusing on religious groups and poor communities.


‘Giving' can be a slippery notion, defined by the mind of the giver. For example, some people give money to car guards not because they believe their cars to be any safer, but as a way of helping poor people who are trying to help themselves – but they are in fact paying for a service. For these and other reasons, the survey design was a lengthy and complex process. First, a set of 15 focus groups were staged across the country in order to inform questionnaire design, and to reveal different understandings of what people give, why they do so, and to whom – as well as the reverse, namely what people do not give to and why. A full 100 respondent punched pilot study was important in helping finalise the survey instrument.

A random stratified survey sample was drawn by S&T. The sample was stratified by race and province at the first level, and then by area (rural/urban/etc.) at the second level. The sample frame comprised 3 000 respondents, yielding an error bar of 1.8%. The results are representative of all South Africans aged 18 and above, in all parts of the country, including formal and informal dwellings. Unlike many surveys, the project partners ensured that the rural component of the sample (commonly the most expensive for logistical reasons) was large and did not require heavy weighting (where a small number of respondents have to represent the views of a far larger community). Randomness was built into the selection of starting points and the selection of respondents. Three call-backs were undertaken to interview the selected respondent; if s/he was unavailable, the household was substituted.

A nation of givers

South Africa appears to be a nation of givers: in the month prior to being interviewed., over half of respondents (54%) gave money to charities or other causes, a third (31%) gave food or goods to charities or other causes, while slightly less than a fifth (17%) volunteered time for a charity or cause.

In addition to giving to formalised institutions or causes, slightly less than half of respondents told us they gave money and/or goods (45% respectively) not to formal charities but directly to the poor – street children, people begging on the street and so on.

Thinking about the last month, have you personally %

Given money to a charity or other cause?


Given goods, food or clothes to a charity or other cause?


Given time (i.e. volunteered) to a charity or other cause?


Given money o a beggar/street child/someone asking for help?


Given food, goods or time to a beggar/street child/someone asking for help? 45

Respondents appear to be more comfortable giving to formal structures than to the poor directly – if we measure this by the amounts given to formal structures rather than directly to the poor. That said, however, there are still very high levels of direct transactional giving to people in need in the form of cash and/or goods.

9/10 South Africans give

If we combine these different forms and methods of giving, we find that a massive 93% of respondents gave (time, money or goods, to a cause or individual) in the month before being interviewed. This is calculated by counting all respondents who gave money to a charity or gave goods to a charity or volunteered or gave money to the poor or gave goods to the poor. We deliberately cast the net as wide as possible: these figures include respondents who made monthly financial contributions to a charity as well as those (for example) who gave a sandwich or cold-drink to a street child begging at a traffic light.

Giving seems to be ingrained in respondents. Even among those scoring high on ‘alienation' variables, 92% gave in the month prior to being interviewed, rising to 94% among those with low levels of alienation. Similarly, we found that poor and non-poor respondents (measured using an adaptation of Statistics SA's household poverty measure) were equally likely to have given in the month prior to being interviewed. ‘Giving' is not the domain of the wealthy: it is part of everyday life for all South Africans, rich and poor alike.

How much do we give?

The three items that have been quantified are money given to a charity, cause or organisation; money given directly to the poor; and the amount of time given to a charity, cause or organisation (volunteering). The table below provides the mean or average across respondents (in the middle column) and the sum total mobilised (in the column on the right).

Item Mean Sum

Money given to a charity/cause/organisation


R80 781

Money given directly to the poor


R19 790

Time given to a charity/cause/organisation

11 hours

5 807 hours

Total money given (formal and informal)


R100 571

Charities and organisations received considerably more money than that given directly to the poor. Respondents who gave money to formal structures gave an average of R49 each in the month before being interviewed; taken together, these respondents gave a total of R80 781 to charities, causes and organisations. Respondents who gave to poor people (45% of the sample) tended to give substantially smaller amounts – an average of R14 each in the month before being interviewed – and gave just less than R20 000 to the poor.

Who gives?

Men tended to give more money than women: men (who gave) gave an average of R53 to charities and organisations, and an average of R15 to poor people asking for help. Women gave an average of R46 to charities and R14 to the poor. Racial differences also emerged: African respondents who gave money gave an average of R30 to charities and R11 to poor people; coloured respondents gave an average of R94 to charities and R15 to the poor; Indians gave an average of R85 to charities and R29 to the poor; and white respondents gave an average of R125 to charities and R123 to poor people asking for help.

It is notable that while white and Indian communities have commensurate levels of wealth, their giving behaviour differs markedly. Indian respondents were the most active givers; but where Indian respondents tended to give more to organisations than directly to the poor, their white counterparts gave almost equal amounts to organisations and directly to the poor.

In all, 77% of respondents told us they gave money (any amount) directly to charities, causes or organisations or to poor people directly. If we add up the amounts given to organisations and to the poor, we find that respondents who gave money, gave a total of R100 571 at an average of R44 per respondent who gave money.

As a nationally representative sample, we can extrapolate these findings to the population as a whole. South African citizens mobilise almost R930m in an average month for development and anti-poverty work. From one perspective, this is a massive amount of money. Seen in context, it amounts to 2.2% of the total monthly income for the working age population (as measured by Census 2001).

Volunteering 1

In all, 17% of respondents volunteered time in the month before being interviewed; during that month, they gave an average of 11 hours each, totalling nearly 6 000 hours.

Women volunteered slightly more time than men; African volunteers gave the most time, averaging 11 hours each in the month before being interviewed; they were followed by coloured respondents (10 hours), Indians (9 hours) and lastly whites (5 hours). It is important to note that the average amount of time volunteered is constant among youth and adults (between 10 and 11 hours) and only rises among those aged over 60 years of age (to an average of 12 hours). Poor respondents (23%) were more likely to have volunteered than non-poor (17%).

Volunteering, in South Africa, is not the preserve of the middle-class with time and resources at their disposal, which we also saw was true of other types of giving.

Provincial differences?

Means of money and time given by province

Provinces show very different giving behaviours. Respondents from the Eastern Cape – one of the poorest provinces in the country – show consistently high levels of giving both money and time. Their neighbours in the far wealthier Western Cape have the highest average levels of giving money (R88), but low levels of volunteering. The data shows the levels of giving that can be attained through hard work by non-profit organisations in the different provinces: the high levels of giving in the Eastern Cape indicate that provincial poverty is no hindrance to giving; it may be an incentive, given its inescapability.

People also give to non-household family members. This was most common in Eastern Cape and Limpopo, two of the poorest provinces in the country. It was least common in the Western Cape and Mpumalanga. Women (58%) were more likely than men (53%) to give money, goods or other items to non-household family members, although when men did so they were more likely to give money (75%) than women (66%). On other items, such as food and drink, clothing, medical supplies and so on, women predominated. Helping non-household family members differs considerably by race. African (59%) and Indian (58%) respondents were considerably more likely to do so than coloured (42%) or white (39%) respondents. It was also influenced by religion, with two-thirds (64%) of non-Christians helping non-family members, dipping slightly among Christians, while atheists were less (49%) rather than more likely (50%) to do so.

The table below summarises all forms of giving by province.

  W Cape E Cape Mpum KZN N Cape Limpopo Free State N West Gauteng
% gave money
to organisation
75 71 64 60 56 49 47 42 36
% gave goods
to organisation
19 59 19 29 27 35 26 17 30
% volunteered 15 33 5 17 17 26 14 8 13
% gave money to poor 42 55 32 36 34 48 63 46 51
% gave goods
to poor
66 53 21 25 27 49 68 44 55
% gave to non-hh family members 34 75 38 44 52 73 57 47 64
Giving to non-household family members by province

Most deserving causes?

The most deserving causes, according to respondents, are dominated by three categories: those associated with children or youth (22%), followed by HIV/AIDS (21%) and ‘the poor' (20%). These three are followed by a set of smaller categories, including people living with disabilities (8%) and the elderly (5%). We asked respondents what they thought was “the most deserving cause that you would support if you could?” (emphasis in questionnaire). This was an open-ended question where respondents could give any answer they wished; answers were later categorised and given numeric codes. It was phrased so as to allow all respondents to tell us what they would support regardless of whether or not they currently either do so or are in a position to do so.

‘What do you think is the most deserving cause, that you would support if you could?'

The high scores for children and youth, HIV/AIDS and poverty suggest that respondents have a progressive and pro-poor understanding of South Africa and societal priorities.

Local or international?

South Africans are highly motivated to give to local causes, but significantly less so to international ones. Less than one in ten respondents (8%) told us they had ever given money specifically to international causes. We also tested respondents' attitudes to giving to local and international causes. The majority (65%) of respondents regard domestic causes as more deserving than international causes, while a fifth (20%) regard local and international as equally deserving. A further one in ten (9%) had no response, while a tiny 4% told us international causes were more deserving than local causes.

Why do we give?

Helping the poor is widely regarded as an important part of building the new South Africa – 93% agreed or strongly agreed with the statement – which may seem self-evident to some, but indicates the broad popular support for pro-poor policies and programmes. This is not mere altruism: there is an apocalyptic edge to the issue, indicated by the 57% of respondents who agreed or strongly agreed that “if we do not help the poor now we might lose everything later”; just a quarter (26%) rejected the notion. Helping the poor is both a moral act and (for some) linked to self-preservation.

Helping the poor is widely regarded as the responsibility of citizens, not (just) government. Almost two-thirds of respondents (61%) disagreed or strongly disagreed with the statement: “It is government's responsibility to help the poor, not mine”. A quarter (24%) of respondents agreed or strongly agreed with the notion that helping the poor was the sole responsibility of government. Intriguingly, agreement was highest among Indian respondents (at 38%), who also showed the highest levels of giving. Among other race groups, agreement with the statement ranged from 29% of whites and coloureds to 23% of Africans.

Attitudes to giving (all respondents, ‘neutral' not shown)

Giving to make SA a better place?

Respondents from the Western Cape were least likely to agree (66% did so) that they give to make the country a better place; at the other extreme, 95% of Free State respondents and 93% of Eastern Cape respondents agreed or strongly agreed. The provincial differences may deserve further detailed research.

“I give to make the country a better place” % agree
Western Cape 66
Limpopo 70
North West 73
Northern Cape 76
KwaZulu-Natal 77
Gauteng 81
Mpumalanga 82
Eastern Cape 93
Free State 95
Agreement with Likert item (by province)

Help the poor because...?

We asked respondents a pointed question about helping the poor, reflecting its importance to the project as a whole. Respondents had to complete the following sentence: “Help the poor because…” This was an open-ended question; answers are set out below.

“Help the poor because...” (all respondents)

For two-thirds (68%) of respondents, giving to the poor is motivated by feelings of human solidarity – we should give because the poor have nothing, or are suffering, or are in need, or deserve something from us. For others, it seems to be more of a rational decision to try and help tackle poverty (10%). Almost one in ten respondents answered the question in religious terms, with 3% telling us they gave because their God required it of them and 6% because by giving they will be blessed.

Charity vs. change?

A third (34%) of respondents told us they give to people in immediate need, and a fifth (21%) that both short-term need and long-term solutions deserve their support. The data suggest that both charity and development have a support base to draw on. Analysed across the 9 provinces, some interesting patterns emerge.


E Cape




Free State
















N Cape












W Cape




Charity, change or both? (by province)

If we analyse attitudes to giving across the three socio-economic status categories we find a third (33%) of respondents in the ‘low' socio-economic status category supported charity but 47% supported change, with a fifth (19%) supporting both. This was similar in the middle category, with 34% supporting charity, 48% change, and 19% both. When we look at those in the high socio-economic status category, we find 28% supporting both charity and change; while charity and change were supported by 36% respectively of respondents in the category. The poorer the respondent, the more likely they are to support causes that seek to change rather than ameliorate.

Overall, the dominant view remains clear: people prefer giving to causes that seek longer-term solutions to our problems than short-term charitable interventions, although a significant proportion see the value of the latter. There seems to be a solid support base for a wide range of organisations and campaigns, from social movements through to more traditional charities. This is a positive finding for the non-profit sector as a whole, particularly organisations that use advocacy, lobbying and local organisational work alongside developmental work.

Using the data

A key goal of the survey was to make the data available for use in the non-profit sector. We have prepared sets of tables, aimed at allowing non-profits to develop a targeting strategy for deepening their domestic support base, be it giving in any form or specifically targeting money, goods or time. The tables in two forms: firstly, the statistical tables provide a probability matrix which reveal the probability of giving across a range of demographic variables: province, dwelling/area, sex, race, education level, socio-economic status, age and so on. This is immediately followed by a second table that specifies which of these groups can be considered a potential target group and which not. A fairly simple algorithm is used: (a) if the level of giving across the different groups for a particular measure is statistically significant (p<0.05) and (b) respondents in any given group are more likely to give than the average then we assume this group to be a target group, and the second table specifies which groups are and are not targets.

The data will be made available, in user-friendly format, on the CCS, SAGA and NDA websites, allowing organisations or individuals to run the data according to their specific needs.


1 ‘Volunteer' appears throughout this report, reflecting the language of focus group participants which was translated into survey design. In South Africa, many unemployed people describe themselves as volunteers because they are not remunerated for work they do.

Giving' is not the domain of the wealthy: it is part of everyday life for all South Africans, rich and poor alike.


[top] [to ] [Previous page]
the team