Build or manage data assets

Data assets include:

• Identifiers: unique labels that we assign to things. These are often considered the first building block of data infrastructure. They provide a clear point of reference for things we might want to collect and publish data about, which helps people to know when they are talking about the same thing, and join datasets together.

• Registers: lists of reference data that help to improve consistency and quality in how data is published and used. They help us to build confidence and trust in data by clarifying where different data stewards are referring to the same things in the same way.

• Datasets: can be data collected directly for research or monitoring purposes, a combination of data collected by others, or the output of a model or analysis. When we think about data assets, we are most likely to think about datasets. At the ODI, we think of data as existing on a spectrum, from closed to open.

The Data Spectrum. Image credit: ODI

Some example of data access initiatives building data assets include: Thomson Reuters’ data services go open – customers use Thomson Reuter’s open identifiers to gain new analytical insight and to build new products and services AMR Research Initiative – pharmaceutical companies share data from their global antibiotic surveillance studies to build a more complete picture of the antimicrobial resistance Drug Repurposing Hub – researchers access the Broad Institute’s dataset of thousands of market approved drugs to identify ways drugs can be repurposed to solve additional health problems