What will the growing enthusiasm for open data do for research on the voluntary and community sector? There’s a lot of hope, but I wonder whether there’s also a fair bit of hype; the results may be opaque as much as they are open.
There’s growing support for 360 Degree Giving’s ambition that 80% of grants awarded by UK charitable foundations and other funders are reported as open data. The Nominet Trust and NCVO are enthusiastically backing this. Some funders are publishing their award data already. What can we learn from their grants data? And what do we need to know?
A list of grants from a funder will tell you who got how much when, for what purpose, and possibly for what timescale, from that funder. With the same information from other funders you can then begin to find out who gets what from more than one source. But without comprehensive data about grantmaking from all funders, you’re going to have to make some assumptions about the other funders and some informed guesses about the other sources of funding available to them.
Here’s an illustration of the partial picture we could end up with. The graph links one set of awards data to information about the distribution of registered charities by region and by level of deprivation (as measured by the English Index of Multiple Deprivation). I’m interested in the general lessons we can learn from this, not the identity of the funder. The vertical axis shows the probability that charities received an award . The horizontal axis shows the level of deprivation in the area in which the charity is based, divided into ten classes ranked in descending order of deprivation – in other words, the left-hand class being the most deprived and the rightmost one scoring lowest on this index. Charities are allocated to geographical areas using their administrative address which might not reflect exactly where the charity operates, though in most cases it is a reasonably good approximation. Each line represents one of the government office regions.
The figures are scaled so that they represent ratios which compare the probability of receiving funding in any given area against the ‘base’ probability – i.e. the chances of receiving funding in the area where fewest charities obtained grants. A figure of 1 shows that the chances of a charity in that area having received an award is the lowest in the country. So there’s a sevenfold range – from the 6th decile of the deprivation index in London, to the top decile (most disadvantaged) of North East England.
If you think it’s appropriate to target disadvantage in some way, then you might take heart, first of all, from the positioning of North East England at the top – it’s been one of England’s most disadvantaged regions for much of the last century and for a given level of deprivation, its charities are more likely than those elsewhere to have received funding.
It is also interesting to note that the chances of receiving funding decline steadily as the level of disadvantage in the local area decreases. In other words, as social needs decline, the likelihood of a charity having received funding from this source drops. In the most disadvantaged parts of all regions, charities were up to twice as likely as their counterparts elsewhere to receive funding. On the face of it, this is an encouraging outcome – insofar it suggests that the pattern is reasonably well-matched with need. But before we rush to conclusions about the meaning of these patterns, we should think about what explanations there might be for it, and about what other sources of evidence we would need to gain a full understanding of it.
I’ll pick those points up in another post – some are obvious, some less so; some will involve a lot of effort and analysis. I’m also sure there are many potential uses of such data for a range of users, which means careful thought about how to maximise its potential.