In recent years, there has been a number of surveys claiming to establish what Britons give to charity. We have just published our report ‘Give and Take’ that looks at the market research, statistical and practical challenges of these approaches to measuring giving. Here are some of the highlights from the report.
Asking people what they give is not easy
Asking people how much they give is fraught with two big difficulties. The first is people’s memories. If I ask you to think about how much you have given in the last month or longer are you confident you can remember everything: the coins in a collecting tin, the direct debits, the round-ups on a bill, the sponsored friends and so on. Now can you add it up correctly? I doubt you can. Apart from the frailty of memory, giving is something that people want to be seen to be doing. Along with issues such as how much alcohol people drink and how much sex they have, giving has what market researchers call the ‘social desirability bias’. Human nature means that we tend to give the answers that make us appear to be good citizens (or who get a lot of sex or who drink in moderation). These reasons alone mean self-reported answers on giving need to be treated with profound caution.
Three surveys, three different conclusions about which age group gives the most
We have a case study in the report of three surveys by CAF, Blackbaud and Barclays Bank each of which suggests that a different age group gives the most to charity. CAF says it’s the traditional older age group of 65+, Barclays say it’s the 35-54 years old age group and Blackbaud says its Millennials and Gen Z (broadly speaking the under 35s). Of course, they can’t all be right. Our report digs into the research and shows that the differing conclusions come from differing samples, different methodologies and different ways of extrapolating the results to a national population. For example how many people there are in the Millennials or Baby Boomers group is critical if extrapolating to create a national picture.
Calculating the national picture from individual surveys is really dodgy
At nfpSynergy, we have been fairly careful over the years to avoid using our surveys to derive a total figure for individual giving in the UK. This is because we think the market research constraints make this too unreliable. For example, if you extrapolate the two variables of ‘percentage who gives’ and the ‘amount they give’ the confidence limits are squared. For example, scaling up our data to the UK population means a 95% confidence interval of £7.5 billion to £8.4 billion! In other words, a ‘drop’ or ‘increase’ of the order of £400m can be completely insignificant statistically.
More importantly, any changes in giving are impossible to determine to a particular cause. Yet CAF cites falls in trust as the reason for a decline in the number of people giving. We simple don’t know that – their research certainly doesn’t show it.
Sadly, using SORP-based accounts is no more reliable
It would be easy to imagine that an alternative to using public surveys would be to look at the formal accounts of charities. Sadly not. Recent changes to SORP have made it all but impossible to determine what money is ‘donated’ to charities by individuals. This is because the SORP ‘donations and legacies’ category includes grants from government or foundations. So, if you go to the Charity Commission website it would be easy to imagine that British Council is the most amazing fundraiser – it raises £168 million from ‘donations and legacies’ while spending ‘£0.000B’ on ‘raising funds’. In reality, this income is just the grant from the Foreign Office. The problem is that disentangling grants from real donations requires forensic analysis of detailed accounts – and for some organisations it just won’t be possible.
So, charity accounts are a shaky foundation for any analysis of giving patterns. Which is very sad.
We should start with what fundraisers want
Perhaps a better way of looking at the challenge of public giving is to ask what is actually useful for charities. I have yet to meet a fundraiser who finds a macro analysis of total giving levels useful (or the equally simplistic global generosity league).
At nfpSynergy, we get a steady stream of requests for benchmarking data on different types of giving. What are people raising from events? Or from lotteries? Or major donors? And what is each type of giving costing? How many staff are employed? This kind of data for fundraisers is like gold dust – but there isn’t an authoritative benchmarking survey with a high level of participants in the sector.
However, the best starting place would be to ask fundraisers and charities what information about giving and fundraising they would like to help them do a better job
A panel-based approach may be the best solution
One solution to the problem of measuring giving in a way that’s useful could be to have a panel of charities, representative of the different sizes and types of charities. This type of granular approach could allow a rapid, accurate and useful analysis of individual giving levels and the success of different types of fundraising. It would allow charities to know whether the good or bad results for their recent appeals, events or activities were reflected in the wider sector.
One thing is for sure – fundraisers deserve and need better data on giving than they get at the moment. And the sector overall deserves a more accurate and reliable approach to measuring public giving than self-reported public surveys.
Read our 'Give and Take' report on the challenges of measuring public generosity in the UK here.