Creating new datasets

This strand is focused on developing cutting-edge new tools, techniques and cohorts which will enrich data research and analysis.

It consists of two core areas:

  1. Methodological Innovation
  2. Historical Longitudinal Data

Methodological Innovation

Under this theme we are focusing on the creation of new datasets (using data from present-day administrative records), either by enabling new linkages, or accessing newly available data. In doing so we aim to be as flexible as possible, responding to new opportunities that arise from data controllers’ appetite to share data (and what they expect to get from this) and emerging needs or interests from policymakers in government and beyond.

Current areas being explored under this theme include:

  • creation of a linked data set on veterans in Scotland;
  • linkage of the Scottish population, using individual's CHI number to a house (UPRN number)

Historical longitudinal data

The second theme involves analysing longitudinal data, which is gathered by following particular individuals over prolonged periods of time—often years or decades. We aim to demonstrate the value of this approach by developing individual-level ‘longitudinal’ data to support the substantive programmes of research.

Analysing administrative data across the complete lifespan of individuals can tell us a great deal about the impact factors in people’s lives, and how early impact factors have an effect in later lives.

The aim of this work is to promote innovation in methodological approaches in linking and coding, and run illustrative projects to show the quality of the data available. The opportunity to analyse complete-life histories of individuals from birth to death is crucial to understanding the mechanisms linking early and later life, supporting the current development of strategies for improving the outcomes of Scottish citizens.

Current areas being explored under this theme include:

  • enhancing the Scottish Historic Population Database (SHPD) through the application of data science, for example auto-coding causes of death in Scotland;
  • enhancing birth cohort datasets through the application of record linkage and data science, for example Enhanced SMS47 Cohort standing dataset project.

Categories and tags

eCohorts
datasets
birthcohort
population database
Veterans