KIND Community Standards: Data collection
This is a draft set of community standards for data collection. These will be reviewed by the KIND community before publication during August 2024.
Basic standard for data collection
You need to be absolutely sure that you are allowed to collect the data you’re planning to collect. If there is any doubt, you should contact your Information Governance and/or Caldicott teams before beginning collection.
Note that this guide does not deal with data collection for research purposes, which is governed by different legislative and ethical standards. There can be some difficulty in deciding whether a piece of data collection would count as research, and so you might find the HRA Is my study research? tool helpful.
- only collect data you are going to use
- collect the absolute minimum information you can. This is both good practice, and a legal obligation
- make sure that your information has not already been collected
- different types of information (especially personal data) will require different standards. This can be complex, so please make contact with your colleagues in Information Governance early in the process
Enhanced standard for data collection
- do you need to make sure that data will be collected consistently? This is especially important if data collection will be lengthy, or large-scale. Consider writing an SOP for data collection
Why think about data collection?
“We are often asked to collect, but many of us lack official training and support for the kinds of complex questions thrown up by data collection. Using data and collecting data are really different” (anonymous KIND member)
Why collect data?
Some example purposes, drawn from a small-scale survey of 40 analysts from NHS boards performed during 2023:
- for clinical audit / quality improvement
- for statutory data collection / official and national statistics
- establishing new services / requirements
- user research
- organisational consultation
- for performance and evaluation purposes
Tools used for data collection
In a 2023 small-scale survey of 40 analysts, the main tools identified were as follows:
| Tool | n |
|---|---|
| Excel | 32 |
| Microsoft Forms | 16 |
| Other survey tools (e.g. Surveymonkey, SmartSurvey) | 15 |
| RedCap | 2 |
(note that respondents were able to identify >1 tool used)
That simplicity of tools used conceals enormous complexity. A sample of factors identified as influencing tool choice were as follows:
- availability / ubiquity amongst those collecting
- flexibility / integration
- audit trail
- downstream ETL
- commissioning and training requirements
- ease of use / skills available
- IG / robustness / confidentiality
- cost
- stakeholder mandate
- technical features (e.g., limited response types)
- speed of build
- institutional approval and support
Many staff have restrictions on the tools available. For example, looking at Microsoft Forms, a wide variety of different possibilities are found. A survey of 64 members of the KIND network during July 2024 gained responses from 23 organisations and NHS boards.
Q: Can you use Forms?
| response | n |
|---|---|
| Yes - for anything | 17 |
| Yes - but not for personally-identifiable information | 36 |
| No | 2 |
| Other | 9 |