Review of the use of race and ethnicity statistics in media articles
At a glance
The previous government’s Inclusive Britain was a government action plan that set out 74 actions to tackle entrenched ethnic disparities. One of its aims is to understand the way issues of race are reported in order to ‘foster social cohesion and remove stigmatisation, deliberate or otherwise, of any ethnic group’. In order to achieve these aims, the then Equality Hub commissioned IFF Research to build the evidence base around the use of race and ethnicity statistics in media articles reporting on particular ethnic disparities: stop and search, ethnicity pay, school exclusions and maternal health.
This work has been carried out in the context of and recognising the UK’s historic commitment to freedom of the press. The Government strongly supports editorial independence and does not interfere with what the media can or cannot broadcast or publish.
This work was commissioned under the previous Conservative government in 2023.
About the client
The then Equality Hub (now the Office for Equality and Opportunity) is part of the Cabinet Office and focuses on disability policy, race and ethnicity, sex and gender equality, LGBT rights and the overall framework of equality legislation for the UK.
Our approach
Our work built on previous research commissioned by the Equality Hub and the Office for National Statistics to understand what the public thinks about ethnic identity terminology. It also relied on the new Standards for Ethnicity Data that sets out how government departments and other public bodies should record, understand and communicate ethnicity data.
Based on the Standards for Ethnicity Data, we created a framework to understand how ethnicity data is being used. We sought to create a systematic and objective approach whilst recognising that any research in this space will incur an element of subjective analysis. For example, our framework sought to establish the wider social, economic or cultural context in which the statistics were framed.
Our framework also sought to identify how race and ethnicity statistics are contextualised based on the information provided by the primary statistical report. For example, the framework identified if limitations of the data set, necessary to accurately interpret the statistics, had been sufficiently acknowledged in the media articles.
A summary of the framework can be found here.
To focus the scope of our review, we conducted searches for media articles in Edge and Google and developed a step-process to enable us to search for relevant articles. In total the review covered 32 articles published online via 12 different media channels.
Key findings
1. Our review found that most media articles reported race and ethnicity statistics correctly. This means that the statistics quoted were the same as those presented in the source document. In the instances that data was incorrect, this appeared to be a genuine error (for example, a typo) on the part of the author.
2. Articles did not always provide the underlying detail of the data set. For example:
- Some articles reported disparities in rates but did not present the base size or the number of incidents within each ethnic group. This information can be useful context for the reader. For example maternal death statistics in England were sometimes reported in terms of the disparity rate between black women and white women, without referencing the total numbers of maternal deaths.
- Some articles focused only on disparities for a particular ethnic minority group (often the black ethnic group) but not others.
- Some authors derived their own ratios from the statistical report to present findings in a more comparative way, but did not provide additional context (such as the actual numbers) for the reader to interpret the statistical nuances.
3. Articles frequently included links to the relevant research or published statistics, but did not always provide detail on the methodology and limitations to the data. Instances of reporting that were more compliant with the Standards for Ethnicity Data reporting were found when the organisation publishing the statistical data provided an accessible summary of their research findings, methodology and limitations perhaps in their press release.
4. Articles often provided anecdotes, opinions and quotes interspersed with statistical data that did not always support the statements being made.
5. Overall, our review found that in these articles ethnicity and race statistics were communicated in a variety of ways and there was no ‘standardised’ approach or practice – effectively, instances of compliance and non-compliance with the Standards varied across and within the articles we examined. It is important to note that the purpose of our review was to assess the accuracy of the ethnicity data and the context in which it was presented, rather than the conclusions drawn by the author. No assessment has been made on the independent editorial practices of news outlets or the journalistic approach taken within individual articles.
Next steps
To improve how government statistics are communicated to the public, the Equality Hub (now the Office for Equality and Opportunity) will build on the government’s Standards for Ethnicity Data to provide guidance on how ethnicity data and analysis should be communicated by government departments and public bodies in press releases and media summaries. The guidance will help those who publish ethnicity data to communicate their findings in a consistent and accessible manner, and facilitate the accurate reporting of complex statistics.