Anonymising data

  • Print

Anonymising data

Before data obtained from research with people can be shared with other researchers or archived, you may need to anonymise them so that individuals, organisations or businesses cannot be identified  

- UK Data Archive



Why do I need to anonymise my data? 

You may need to anonymise your data as a result of:

  • ethical 
  • legal (e.g. personal or sensitive data) 
  • or commercial reasons

This is generally to protect the identities of those involved in your research. 

It is considered good practice to anonymise your data early on in your project, for example, alongside gaining informed consent and when imposing any access restrictions. 

What about visual or audio data?

It is possible to alter the voices and distort the images of those that may be involved in such research. 

However, this can be both a lengthy and expensive process so a better solution maybe to ask participants for consent to keep such data in its original format. 

Special considerations

With data that is geo-referenced or relational, it may be better to impose access restrictions when sharing data so that no valuable information is lost i.e. reference points.

Alternatively, you can replace identifiers with other variables which still allow you to understand the data, but protect the identity of a location.

How do I anonymise my data? 

There are several ways to anonymise your data, which may differ depending on the nature of your research:

Quantitative Data

  • remove any identifiers such as names, addresses, or date of births
  • remove any indirect identifiers such as age, occupation, salary
  • alter direct and indirect identifiers to more generic information (i.e. use postcode areas instead of the whole postcode)

Qualitative Data

  • use pseudonyms
  • keep a record of any replacements / edits / aggregations just in case this information is required in the future 
  • store anonymised data separately from data in its original format to avoid confusion