How to report and considerations for onward use of the data
OSR’s expectations
As with data collection, users of statistics benefit when producers are clear about what is reported. Data about sex and gender identity should be explained and defined for the purpose of a particular set of statistics, and terms, including ‘gender’, should not be used interchangeably or as a substitute for each other.
Producers must clearly explain the sources of data about sex or gender identity that they are using and how these data are collected. This applies to both survey and administrative data sources. Where producers do not have a clear understanding of how this information has been collected, this should be communicated transparently to users. Producers should also provide information about how responses such as ‘prefer not to say’ are treated in data processing.
The quality and any limitations of the source data should be shared with users. For surveys, this may include information on response rates, sample size, response bias and mode effect. For administrative data sources, this may include information on the potential sources of bias and error in administrative systems and how complete the data are.
If changes have been made to the data collection, including the questions asked or the nature of the guidance provided, this should be explained clearly to users, including information on any potential impact on comparability over time.
Many surveys are weighted to increase the representativeness of the data as well as to mitigate the impact of non-response bias. Producers who are applying weights to their survey data should ensure that the concepts within the weighting dataset and the survey data are aligned as far as possible. Producers should clearly explain the nature and the source of the data and the methodology used.
Producers should carefully consider how they assure themselves of the quality of the statistics, particularly where questions are voluntary or there may be limited other sources of data to compare or triangulate with. Producers should clearly explain the extent and nature of any uncertainty in the statistics.
The strengths and limitations of the published statistics should be considered in relation to different uses and be clearly explained. This may be particularly relevant when considering data at small sub-group levels, for example for certain gender identities. Producers should support users in understanding the data and the conclusions that can be drawn.
Producers should apply appropriate statistical disclosure control methods to protect individuals’ data. This is especially important when producers are reporting on small populations. If producers are reporting on gender identity, they may find that some of the response options are too small to report on and the data may need to be combined or suppressed.
Onward use
We strongly champion the onward use of data including through data sharing, data linkage and the reuse of data where feasible. We consider that these activities greatly increase the value of the statistics and data.
When providing data for onward use, such as through Trusted Research Environments, producers should ensure that they provide clear information on how data about sex or gender identity have been collected and processed. This information will help analysts and researchers understand the approach taken for each data collection.
When using combined datasets that may allow individuals to be classified as transgender, producers should be mindful of ethical considerations, seeking advice from their Head of Profession for Statistics, Chief Statistician or Lead Official and the expertise of the UK Statistics Authority’s Centre for Applied Data Ethics if necessary.
Questions for statistics producers to consider
How well are you communicating information to users?
Producers should ensure that statistics, data and explanatory materials are presented in a clear, unambiguous way that supports their intended use. The strengths and limitations of the statistics and data should be clearly explained to users, and users should be made aware of the nature and extent of any changes to the data collection.
Ask yourself:
- Have you been clear about the definitions or terms that you have used?
- Have you communicated any changes to the data collection to users?
- Have you included information on how these data are collected and any associated limitations?
Have you included key information about your statistical methods?
Producers should ensure that they are transparent about their methods and provide key information to support users when interpreting the statistics or conducting their own analysis.
Ask yourself:
- Have you explained to users the extent to which your approach to collecting data about sex or gender identity is coherent with related statistics?
- If you are applying weights to your data, have you assessed how closely the source data align to the concepts within your survey? Have you also clearly explained your weighting methodology as well as any associated limitations?
- Have you applied appropriate statistical disclosure control methods?