Assessing the accuracy of your data

Knowing the quality of the input sources is important to ensure resulting information is fit-for-purpose. Without this knowledge the wrong assumptions could be made when transforming the data into information, jeopardising the quality of the output.

Impact on data quality   Accuracy   Relevance

Learning outcomes

  • Know your data sources and their quality dimensions.
  • Understand potential causes of error in each data source.
  • Understand the mitigations to reduce each type of risk for each data source.

To understand the quality of source data you need to:

  • Know what your input data sources are
  • Understand the quality of your data sources across all 12 quality dimensions
  • Understand the assumptions made, and limitations of, each input data source
  • Recognise how to know these assumptions are being met and what to do if they aren’t.

Errors relating to source data include:

  • Specification error (concepts, frameworks, design feasibility)
  • Frame error (approximations to the design, imperfections, exclusions)
  • Non-response error (data or a component that’s not provided)
  • Measurement error (inability to provide precise and relevant data)
  • Sampling error (applicable to any sample survey components of the product)

 

Top