Classifying the charity register

Christopher Damm |

To anyone doing research on the voluntary sector, or building services that help voluntary organisations, the size and diversity of activities that charities undertake is always staggering. From a large international charity building relations between Britain and Japan to a small trust maintaining a village playing field, the term “voluntary sector” hides a huge amount of complexity.

We are a small team of voluntary sector data specialists, supported by the Esmée Fairburn foundation to improve the classification of charitable activities in the UK. The project is still in its early stages, but we wanted to share our plans with the wider voluntary sector research community. 

Many data research projects on UK charities use the various national registers of charities as their main source of data. The breadth of activities that these organisations undertake, however, can mean that research on the sector as a whole hides the complexity, nuance and variation between these organisations. 

Voluntary Sector researchers may have come across the International Classification of Non-Profit / Third Sector Organisations (ICNPTSO). These classifications are widely used and were originally designed to help enable international comparisons of non-profit  organisations, particularly when preparing national accounts. As with any system of classification, they have their advantages and disadvantages depending on the user and their goals. 

A first potential concern for UK users of the ICNPTSO categories is that they are not recorded as part of charities’ registration process. They have to be allocated retrospectively; no small task given that there are over 160 thousand active charities registered in England and Wales alone. The last attempt to do so was over 10 years ago, meaning the number of unclassified charities continues to grow. 

Second, the ICNPTSO classification system is also sometimes a poor fit for common areas of charitable activity in the UK. For example, there are no categories for food banks, drug addiction services, or domestic violence refuges. 

Third, charities are also generally assigned to a single category. This has advantages for comparative analysis, but can make it even harder to capture the purpose and activities of some organisations. 

To help solve these problems, the research team has been funded to undertake a number of tasks. 

First, the team has created its own UK focussed classification scheme. This scheme is built from the ground up – we have started with a list of charities and then tried to attach “tags” to them based on their activities. It is more granular than the ICNPTSO system and designed specifically for use with UK charity data. 

Second, we are in the process of creating a manually classified sub-sample of the charity registers for England and Wales, Scotland and Northern Ireland. In line with previous efforts, we are applying a single ICNPTSO category for each charity, whereas for our own UK focussed tagging system, charities can be ‘tagged’ with multiple categories.  We then plan to use this sub-sample to create a set of automatic rules to classify the remaining charities based on keywords in their register entry. 

The task at hand is challenging but important. No classification system will be able to capture everything that anyone might be interested in (or it wouldn’t really be a classification system at all). Nor will we get every decision correct, especially with a limited view of organisations based on their written entries in the registers of charities. But we hope that our efforts will result in both a more up-to-date application of the ICNPTSO categories, as well as an entirely new way of breaking down and exploring charitable activities in the UK. 

Crucially, we want the outputs from this project to be used as an open classification scheme that can be used and owned by the whole voluntary sector. We don’t expect to get the answers right the first time, so the scheme will improve as people use and feedback on it.

We’re excited to share developments with you as the project progresses, and if you’re interested, please do get in touch or have a look at a more detailed outline of our project. We would love to receive your feedback or discuss our plans further. 

Chris Damm (Sheffield Hallam University), Oliver Chan (National Council for Voluntary Organisations); and David Kane (independent Data Scientist).