OPINION29 April 2021

Being more censu-tive to diversity in research

Inclusion Opinion UK

As we expand our understanding of society, research design must also evolve, says Jordan McLaren. 

Diversity difference ideas nature_crop

As you’ll be aware, research recently had its 10-yearly moment of nationwide fame – the census (in England and Wales).

Over its 200-year lifespan, the census has adapted its design to be more inclusive of the population it measures.

This ongoing adaption led me to ask: Are we – commercial researchers – adapting ourselves in the same way to ensure we’re ‘censu-tive’ enough towards the increasingly diverse nature of the society we investigate?

The current approach to measuring ethnicity in the census is newer than you think

The census first measured ethnicity in 1991. Pre-1991, ethnicity was measured alongside nationality, using old, new and African Commonwealth identification as a proxy for non-white ethnicities. This is a method we now know to be inaccurate and not reflective of the times.

In 1991 when the census modernised its approach to measuring ethnicity, so too did the Labour Force Survey. This measured an ‘increase’ in the UK’s ethnic minority population. The more inclusive and accurate method of measuring ethnic minority identities in the UK is credited with causing this ‘increase’. This is evidence that minor adjustments to how we approach research design over time can, and does, improve research effectiveness.

A small step to improve representation regarding gender identity

2021 was the first year the census in England and Wales asked: "Is the gender you identify with the same as your sex registered at birth?"

This allowed for better representation of the transgender and gender non-binary communities. It is a crucial step in recognising these communities and producing insights that can impact distribution and access to resources such as NHS funding for gender related healthcare.

Such edits allow for the transgender and gender non-binary communities to see themselves reflected in the census. This may seem a small thing, but it provides people in these communities a sense of inclusion and recognition.

Conversely, if the census hadn’t made this edit, it’d risk transgender and gender non-binary communities feeling isolated and not included in policy maker’s thinking.

Research design isn’t neatly separated from the politics of identity

As a young researcher, I find it refreshing to see the census updated in ways like the above.

However, ideas of accurate research are continually being contested. Currently, court rulings are in place for amendments to the confirmation processes of a respondent’s sex. These conversations have sparked heated debate on the ‘accuracy’ of such processes and evidence the way that the census’ design reflects both our societal beliefs and how they evolve.  

As we continue to expand our understanding of society, research design needs to incorporate this evolving knowledge. Research design is not devoid of ideological intent. In fact, research design’s language and categorisations are a reflective snapshot into society similarly to the outputs themselves.

Why is diversity and inclusion important in research design?

I’ve used the census as a springboard introduction to diversity and inclusion in research design as we’re all aware of its purpose.

But is the research design used to make business decisions on communication and innovation as ‘censu-tive’ to diversity and inclusion as it should be?

It’s no secret that our industry has long been centred on the white, heteronormative, male experience. However, by pursuing a commitment to increasing research’s diversity we can gain a range of insight that informs brands how to better position, communicate and appeal to today and tomorrow’s diverse population.

How do we design research to be more censu-tive to inclusion?

It is a big question, and not one a single article can answer in full. However, here are three things we do to design diverse and inclusive research: 

  1. Make diverse sampling the default mindset for research design

Think about minority participants from the beginning. When sampling, start specifying a diverse recruit, explicitly for ethnicity, gender identities beyond the binary, sexuality, those with disabilities or neurodivergence.

Often these people are harder to find. But as gatekeepers ensuring brands hear the market’s voice, we’re responsible for ensuring everyone’s voice is heard. Practically, this means prioritising their recruitment early on to avoid running out of time to include them.

  1. Incorporate evolved language

We now have an expansive vocabulary to signal identity and difference. It’s important to stay in tune with this and use it in research design. Take gender identity for example: offering participant criteria beyond the male/female sex binary is vital in signalling to these audiences that their voice is important.

  1. Diverse research team = diverse research design

We can’t get it right every time. We’re human and socialised into all kinds of biases and perspectives we may not even consider. However, this is why it’s important to surround ourselves with people who can offer different considerations and perspectives. That’s why having a diverse team will equate to diverse research design.

Most importantly, we as researchers need to shift from being aware of these issues to taking meaningful actions that address these challenges. Fostering an inclusive environment committed to making change is key towards a successful and empathetic future. There’s only one way to do so: by being ‘censu-tive’.

Jordan McLaren is research executive at Northstar Research