Skip to content
Social Research

Making better use of existing data

Bookmark and Share

Data linking

Data linking is a technique for creating links between data sources so that anonymised information that is thought to relate to the same person, family, place or event can be connected for research purposes.

Data linking must be done in a manner that is secure, ethical and in line with the requirements of the Data Protection Act and other relevant legislation.
The data linking demonstration projects were funded jointly with the Economic and Social Research Council (ESRC) and were completed using the Welsh Government funded Secure Anonymised Information Linkage (SAIL) infrastructure established by Swansea University.

Key ways in which data linking can contribute to the evidence base

  • It can allow us to “collect (information) once, use many times”.
  • By linking administrative data from multiple sources, we can plot the ‘journeys’ of different kinds of service user to explore the different ‘pathways’ they follow through a range of related services. This can help us to identify where services could be improved.
  • Linking data is by its very nature ‘longitudinal’ i.e. it allows us to see how things change over time. This also allows us to disentangle cause and effect, something that is essential to the design of successful interventions and can otherwise only be achieved by funding expensive primary longitudinal research.
  • In some cases, we can reduce or even eliminate the data collection burden on citizens, saving money by using data we already hold rather than undertaking new primary research.
  • The linking of data sets relating to social policy interventions can allow us to create baselines retrospectively and to create detailed ‘control’ cases in order to support policy evaluation.
  • The linking of large-scale data sets can allow us to research rare groups or small groups not distributed homogenously in the population as well as to provide estimates for smaller geographical areas. For some small population groups, for example some of those with characteristics protected under the Equality Act 2010, linking administrative and/or survey data will be the only realistic way to deliver usable data.
  • Linked data allows us to analyse the relationships between various different issues that influence peoples’ lives, for example to research ‘wicked’ issues that have complex, crosscutting causes e.g. domestic violence, obesity, substance misuse.
  • Administrative records can be used to add value to existing social survey data sets, for example by comparing health service use with self-reported health and subjective well-being.
  • Using shared, harmonised variables (from survey or administrative datasets), we can start to develop richer population information for Wales.
Keep up to date @StatisticsWales@StatisticsWales Sign up to our newsletterSign up to our newsletter
Share