Data quality management within the DOM ensures that data remains accurate, consistent, timely, and reliable across the organisation.
- Accuracy: Verification protocols to ensure data accurately reflects real-world conditions.
- Completeness: Data completeness checks ensure no gaps or missing values, maintaining dataset integrity.
- Consistency: Standardisation ensures data across sources and departments is aligned for comparison.
- Timeliness: Data is accessible as needed, with updates aligned to BMT’s decision cycles.
- Uniqueness: De-duplication efforts at all layers to avoid redundancy and ensure efficient data use.
- Usability: Ensures data is structured, accessible, and understandable for end-users across BMT.
Accuracy
How well does a piece of information reflect reality?
Accuracy refers to the degree in which the data correctly portrays the real-world situation in which it was originally designed to measure. Data must be meaningful and useful to allow for correct and accurate interpretation and analysis. For data to be accurate, it must also be valid, meaning it must conform to a defined format whilst implementing and adhering to specific business rules, which may be recorded in a metadata repository (a system or application where information about the data (metadata) is stored and managed).
Completeness
Does it fulfil our expectations of what’s comprehensive?
This dimension reflects the ability to determine what data is missing, and whether omissions are acceptable (for example, optional data). Departments must determine and understand whether a data asset contains unacceptable gaps, as these may place limitations on the data leading to an increased reliance on assumptions and estimations or preclude the asset for use altogether. It is also useful to note the level of completeness, particularly if achieving 100% completeness may not be necessary to fulfil the dataset’s intended purpose. Also, if the dataset is considered complete as at a particular point in time, e.g. beginning or end of month.
Consistency
Does information stored in one place match relevant data stored elsewhere?
Consistency of data means that the data is collected, grouped, structured, and stored in a consistent and standardised way. This requires standard concepts, definitions, and classifications to be implemented across departments, and agreed upon as to their meanings and interpretation.
Data must also be consistent in the context of its use. For example, data may appear similar but have different meanings or uses in different departments. Duplication, or different meanings for similar data, may result in confusion or misinterpretation of data and render such data unsuitable for comparison with related assets. Also, it may be unclear if trends are due to a true effect or due to problems with inconsistent data collection.
Timeliness
Is our information available when you need it
Timeliness refers to how quickly data can be made available when required, and the delay between the reference period (period to which data refers, such as a financial year) and the release of information. Factors that may impact this include collection method and processing. Data must be discoverable, available, and accessible throughout all stages of the data asset lifecycle from creation to retirement, to be available for greater internal use, external use (external partners, other government departs and researchers) and the public. If delays occur during the provision of data, currency and reliability may be impacted.
Uniqueness (integrity)
Can different data sets be joined correctly to reflect a larger picture?
There must be the capacity to make meaningful comparisons across multiple data assets. This is achieved through common data definitions and standards. Common data definitions should be agreed and shared across the department, and any inconsistencies should be managed.
Usability
Is our data structured and accessible for effective use?
Usability ensures that data is organised, accessible, and understandable for end-users, enabling them to efficiently obtain meaningful insights without unnecessary complexity. High usability means data is not only available but also intuitive and actionable, supporting quick decision-making and operational activities. This requires data to be accessible via clear interfaces and structured in formats that align with users’ knowledge and needs, helping them navigate data with minimal support.
Data usability also involves establishing appropriate metadata, user documentation, and visualisation aids to support seamless access and interpretation. By prioritising usability, BMT ensures that data is readily adoptable across departments and contributes effectively to business objectives.