Dr Dharmi Kapadia, of the Centre on Dynamics of Ethnicity and panellist for #DataImpact2021, explores representation of ethnic minority people in data and introduces the Evidence for Equality National Survey.
The UK has some of the best national surveys in the world that enable robust research into racism and its consequences for ethnic minority people.
The Ethnic Minority Boost (EMB) sampling employed in many of these surveys, whereby ethnic minority people are over-sampled, is imperative in order to allow for large enough sub-groups for meaningful statistical analysis. Indeed, I have used many of these datasets in my own research investigating ethnic inequalities in access to mental health services and the links between poverty and social networks for ethnic minority people.
But there are many areas where our current data resources fall short.
One of our projects funded by The Nuffield Foundation, ‘Ethnic Inequalities in Later Life’ aims to understand why ethnic inequalities in health, life expectancy and socioeconomic circumstances exist for older people, and how they have changed over time. Ethnic minority older people are the one of the most disadvantaged groups in the UK but worryingly, quantitative data to investigate this group are sparse.
For example the most recent wave of the English Longitudinal Study of Ageing (ELSA) contains just over 500 ethnic minority older people (total N at Wave 9 = 8,736), and only approximately 160 of these were aged over 65. Even the largest survey in the UK, Understanding Society, is lacking in statistical power to examine ethnic inequalities in later life. The most recent data release (Wave 10) contains only 561 non-White ethnic minority people aged 65 and over (out of 6,467 people aged 65 or over).
These sample sizes do not allow for subgroup analysis to ascertain how older people in different ethnic minority groups are faring; for example there are only 36 Bangladeshi older people.
This minimal representation in nationally funded datasets really equates to exclusion from research and is a form of institutional racism leading to a socially unjust dearth of knowledge about the health and socioeconomic circumstances of an already disadvantaged part of the population.
To address some of these data problems, we have harmonised a number of survey datasets from 1993 to 2017 with suitable ethnic minority boost samples, to undertake analyses of how older ethnic minority people’s health has changed over time, relative to the White majority population. Emerging findings show persistent ethnic health inequalities over more than 20 years.
As well as thinking of creative methodological solutions for use with existing data, we are also undertaking primary data collection to understand the drivers of ethnic inequalities, and how these have manifested during the Covid-19 pandemic.
Evidence for Equality National Survey (EVENS)
This survey is the largest of its kind and aims to recruit 20,000 people in total to document and understand the experiences of ethnic and religious minorities during the pandemic. We are using cutting-edge methods to collect survey data online and via telephone. The non-probability nature of the sample requires statistical methods to compensate for the selection bias. These methods rely on probability reference samples that will be used to incorporate ‘randomisation’ into the non-probability sample and allow for statistical inference.
Participants will be recruited using a targeted digital campaign in partnership with major Voluntary, Community and Social Enterprise (VCSE) organisations such as Race Equality Foundation, the Muslim Council of Britain and Operation Black Vote. Participants are also given extra weblinks to the survey to pass on to family member and friends, allowing the research team to look at how people have been affected collectively by the pandemic.
The innovation of this survey does not lie only in its sampling design. We are also collecting ethnicity in a number ways, including a write-in box, and we will include sufficient numbers of the major ethnic minority groups in the UK, as well as including ethnic and religious minorities that have not been meaningfully captured previously.
For example, we will differentiate African people of sub-Saharan descent and those who are not, and aim for large samples of Jewish and Gypsy Traveller populations. EVENS has the potential to truly capture the experiences of ethnic and religious minorities during the pandemic and influence current debates and policy. Importantly, we will also deposit the new EVENS survey dataset with the UK Data Service, allowing other researchers to access these high quality data and conduct their own analyses.
How is data collection developing?
In recent years, advances have been made towards including and representing ethnic and religious minority groups in surveys. We have seen that the upcoming 2021 census of England and Wales will additionally include ‘Roma’ as one of its ethnic categories, in recognition of the inequalities that this group faces in the UK.
But these acts of commitment to racial equality seen in responsive data collection to produce evidence is required from all statutory organisations, as well as from the UK Research Councils that fund large scale surveys and research projects.
Additionally, to make real headway in addressing racism and its consequences, there needs to be a commitment from national government to end ethnic inequalities. There is still much work to be done in reaching that goal.
With expertise in social statistical methods, her main areas of research are ethnic inequalities in health and access to health services. She has also conducted research on ethnic inequalities in the labour market, and on the relationships between poverty, ethnicity and social networks. She has published in journals including the British Medical Journal and Ethnicity and Health.
Join us for #DataImpact2021: #IdentityInData – Who Counts? Visibility, voice and culture in data collection and use (24 February, 10.30-12.30), where Dharmi will be one of our panellists.