Quality guidelines
Principles
Statistical quality in HSE is defined as meeting user needs with particular reference to the six European Statistical Service dimensions of quality developed by Eurostat:
- relevance
- accuracy
- timeliness
- accessibility
- comparability
- coherence
Practice
The table below demonstrates how we deliver quality statistics in HSE based on the principles above.
|
Quality principle
|
How our statistics deliver
|
|
Relevance |
-
We will periodically review user needs for health and safety statistics and the relevance and use of the data we provide.
-
We will consult with both internal and external users
|
|
Accuracy |
-
all Statistical publications will contain a full description of data quality issues relating to the release. Any impact that has on analysis of changes over time or comparisons between different groups will be clear to both expert and lay audience.
-
For survey data, this description will include information about sample size, confidence intervals and response rates.
-
For administrative data the description will include reference to under-reporting and accuracy of coding as well as providing background information about how the data was collected.
-
Where a 'provisional' statistic is materially different from final' one, a full explanation will be given.
|
|
Timeliness |
-
we will publish statistical releases as soon as possible after the reference period (subject to production schedules, and to the needs of the "coherence" principle below)
|
|
Accessibility |
- A wide range of detailed tables, charts and maps, with interpretation, are available, free of charge, on the HSE statistics website.
- Many of the data tables have been designed as interactive excel tables which allows users to tailor their analysis and outputs.
- The HSE website aims to be as accessible as possible and complies with all web and UK government guidelines regarding accessibility[26].
- In accordance with National Statistics Protocols, all release dates are pre-announced.
|
|
Comparability |
-
We make every effort to ensure consistency of data over time. Where this is not possible, discontinuity lines will be presented and users will be informed of the reason and likely scale of effect.
-
Where reliable employment data are available, data for sub-populations such as by region, industry or occupation will always be presented as rates (often in addition to frequency estimates) to enable comparability across groups.
-
Where it is possible and sensible to do so, internationally agreed definitions will be used.
|
|
Coherence |
- As standard practice, we will release related statistical publications on the same day in order to aid user understanding unless:
- This would mean significant delay to one set of published figures in order to present the coherent set of releases (or)
- User Engagement suggests that separate releases on separate days would be preferable
- Where related measures are published across several publications we will clearly indicate to users where the related information can be found.
- Our statistics website will include analysis and interpretation of data for key health and safety topics drawing on information from the full range of available sources.
- Where different data sources measure similar aspects, we will provide guidance to users on the use and interpretation of these sources.
|