Refining and embedding the Intersectional "MAIHDA" approach to intersectionality in quantitative social science research.

There are big inequalities in our society, in a range of different things. Regarding health, we know that, in the UK, white people are generally more healthy than non-white people, and rich people are more healthy than poorer people. Beyond health, there are similar inequalities everywhere - in rates of unemployment, in income, in education levels, and so on.

What is less well known is how these inequalities vary and interact. For instance, we know there is an ethnicity gap in health, but is that the same for men and women? Is it the same for rich women and poorer women? As we combine characteristics to produce smaller and smaller groups, it becomes clear that the inequalities in our society are complicated. And that complexity matters: it's important that we know who has worse health, so that we can target policies to the right people; it's important that we know who is being failed by our education system, so we can think about how the system could be changed to be more equitable. Fundamentally, in order to improve social justice, we need to first understand who is currently being let down by society. And that needs to be more than just a comparison of broad categories of people.

The MAIHDA method is a statistical model that allows us to do this: it lets us to see which combinations of characteristics are associated with advantage, and which are associated with disadvantage. This innovative method has been incrementally developed between 2016 and 2022, and has already informed us about the nature of inequalities in a range of health outcomes.

However, the method remains in its infancy, and this grant will refine the method to allow it to be used in situations that it has never been used before. This includes evaluating government policies to see which groups benefit from a particular policy, and which groups benefit less or even are harmed. The project will also allow us to consider how complex inequalities have changed over time, and how they vary from place to place. We will test our methodological refinements using real datasets which will act as exemplars in how these methods could be used going forward, relating to (among other things) obesity, covid-19, and environmental pollutants.

We are also keen to expand the use of the method beyond health inequalities. As such, we will collaborate with academics and researchers in academic research centres and non-academic organisations, to implement the MAIHDA methods in other social science subject areas. This will include the analysis of carers, and how being a carer affects different groups of people in different ways. It will include an analysis of water quality in the United States, to see how different groups of people, in different places, are affected by poor-quality, polluted drinking water. And we will consider how educational inequalities vary for different groups of people, to see which groups are being let down by the education sector.

Finally, a key part of this grant is training. We want to upskill researchers, in academia and beyond, to be able to use this method and our refinements to it in the future. As such we will produce a wide variety of online training materials, and run both online and in-person training courses, aimed at established academics, PhD students, and non-academic research organisations. These materials will be developed with other stakeholders to ensure that they meet the needs of those organisations and individuals.