When we try to find out whether an organisation is diverse or not, we often look at the numbers – the relative representation of different social groups in their ranks. For example, since 2014, Apple has publicly shared data about the percentages of its workforce broken down by gender and race. In recent years, many large companies, like HSBC and BP have followed suit.
In many regards, this increased transparency is something to celebrate. It shows that companies are taking diversity seriously and tracking their progress. However, this practice can raise some tricky questions. For example, people may look at some statistics and ask themselves, “this is clearly not diverse, but what would a sufficiently diverse organisation actually look like?”
The answer, it turns out, depends on who you ask.
The ‘diversity line’
My colleague Miguel Unzueta from UCLA Anderson School of Management and I recently conducted a series of studies to figure out where people set the threshold between an organisation that is not sufficiently diverse, and one that is.
For example, just about everyone would agree that an organisation that is 0 per cent women and 100 per cent men is not diverse in terms of gender. Similarly, almost everyone would agree that an organisation that is 50 per cent women and 50 per cent men is diverse in terms of gender. Somewhere between these two points, then, must lie a threshold, a point where people will say sufficient diversity has been achieved. We called this threshold the ‘diversity line’.
Although numerical demographics may seem to provide an objective measure of diversity, the ambiguity around what sufficient diversity looks like allows for a subjective interpretation of these numbers. What is really significant then, and often overlooked by organisations, is how diversity statistics compare to different people’s ’diversity line’. Importantly, because of the groups they belong to, some people may be motivated to draw the diversity line at a different point than others.
Different groups define diversity differently
A long history of research shows that groups people belong to, and the relative standing of these groups in society, plays a key role in how they want to see the world. Typically, groups are divided between dominant groups (those who, on average, have the most access to power and resources) and non-dominant groups (those who, on average, have fewer advantages and face greater discrimination).
For example, in the US context where we conducted this research, men are the dominant group in terms of gender and women are the non-dominant group. On average, members of dominant groups typically like to see the world in ways that preserve their advantaged position, whereas members of non-dominant groups like to challenge the existing hierarchy.
Because drawing the ‘diversity line’ sets a normative standard for how strongly groups would be represented, we predicted that members of dominant groups would set relatively low thresholds of non-dominant group representation relative to members of non-dominant groups themselves. This is in line with other research I’ve conducted suggesting that members of dominant groups may feel threatened by diversity efforts.
To see what this would look like in action, imagine a company where the representation of women is increasing from 0 per cent to 50 per cent. We proposed that men tend to set a lower threshold for sufficient diversity than women. As a result, these men would be faster to declare their company’s gender diversity efforts a success than their female colleagues would.
Using novel empirical techniques, my research found support for this prediction over seven studies. Members of dominant groups (looking at both gender and race) consistently drew the diversity line at a lower level than members of non-dominant groups. In other words, this research shows that members of different groups may look at the same organisation and come to conflicting definitions about whether sufficient diversity has been attained and whether efforts to increase diversity within that organisation are still needed. Even when the numbers are clear, the groups we belong to can shape what we do with them.
This could mean that when members of dominant groups are in charge of evaluating diversity efforts, they may set their ambitions lower than members of non-dominant groups do, or declare their efforts a success at a point that members of non-dominant groups would deem premature. This may help explain why there has been such slow progress toward diversity in many organisations.
Looking beyond the numbers
It is important to stress that this research doesn’t mean we should stop monitoring diversity data, or throw diversity targets out of the window entirely. It’s practical and beneficial to have goals, and such targets allow organisations to be held accountable.
But it does show that there is a need to move beyond raw demographics to assess diversity within organisations. For those setting diversity targets, there is a need to recognise what may seem like a reasonable target to you may not seem reasonable to others. This means asking around your organisation when it comes to evaluating the effectiveness of your diversity efforts to get a representative range of perspectives, and also ensuring that you don’t make assumptions about how other people are interpreting your diversity statistics.
Finally, demographic representation isn’t everything. As most organisations are now aware, it’s essential to pair diversity efforts with investments in equity and inclusion. Looking beyond the numbers helps organisations ensure they have the tools they need to build a more comprehensive and sustainable approach to diversity.
Felix Danbold is an assistant professor at University College London School of Management