But in Sweden, there's a big difference between how people say they think, and the decisions they make when they’re hiring.
Diversity in Sweden: the landscape
In 2006, researchers from Åbo Akademie University, Per-Anders Edin and Jonas Lagerström wanted to see if there was any discrimination towards women and people of colour. They examined the frequency of contact between employers and candidates.
To identify possible discrimination, candidates had the opportunity to hide information about their name and gender.
The study showed that equally qualified women and men received the same number of callbacks - but only when the women hid their gender. When women declared their gender in their application, they received 30% fewer responses. Meaning?
For every ten applications, men received 5 call backs and women only 3
This means that job hunting takes longer and, potentially, is more dispiriting to women.
Ethnicity & bias in CV screening
How about looking at ethnicity? One of Sweden’s first studies to look at how someone’s name affects their hiring potential was run in 2007. The researchers from the University of Stockholm wanted to see if there is any effect: does a person’s name really make a difference?
Bias in names
As with the last study, the researches again allowed candidates to hide their names to see what effect this would have on the recruitment process - only this time, the two groups had either native Swedish names, or names suggesting an Arabic origin.
Again, researchers saw that, even where candidates had the same level of experience and qualifications, people with ethnic Swedish names were called to interview 50% more often than people who have ethnic Arabic names.
This means that people with ethnic Arab names have to apply for 100% more jobs to even reach the interview stage. With years of experience and qualifications equal, and the same people achieving the same hit rate when names were hidden, we can clearly see that recruiters are struggling to increase diversity and avoid bias.
Why is bias still a problem?
Application processes that rely on personal (name, age, address) and historical (education, work history) information are more prone to bias. We often unconsciously feel affinity to people who are like us, even if we think we’re good, rational and objective decision makers.
This is what’s known as Affinity Bias - where we unconsciously favour people like us over people who feel different. Harvard University researcher Mahzarin Banjai from HBR writes “more than two decades of research confirms that, in reality, most of us fall woefully short of our inflated self-perception.” Basically? We think we’re more objective than we are.
Achieving diversity whilst we continue to lean on CVs and applications is likely to make it harder for minority groups to achieve equity - and for organisations to reap the rewards of diversity.
Diversity is imperative to future success
McKinsey reports a positive correlation between gender diversity, financial performance and long term value creation. Meaning? When companies take action towards EDI, they earn more money.
McKinsey’s results also show that companies with higher ethnic and cultural diversity have a correlation of up to about 25% towards profitability when compared to companies with lower ethnic and cultural diversity.
Diversity and the pandemic
This rang true in the pandemic: the recently published report from Board Ready showed clear patterns of financial growth in S&P 500 companies with greater diversity:
- 54% of companies with higher gender diversity delivered positive year-over-year revenue in 2020
- This is in contrast with companies with lower gender diversity, who achieved only 45%
- Less than a third of companies in the S&P 500 have boards in which non-whites occupy at least 20 percent of their seats.
- Companies with 30 percent or more board seats occupied by non-white directors performed better than the less diverse counterparts with growth rates increasing from 3.0 percent in 2019 to 4.0 percent in 2020, a modest 1.0 percent increase in growth rate.
Where we are now
The ethical and financial imperatives are clear: diverse leadership creates a significant financial benefit. We can also see that CV screening is allowing bias to affect who reaches interview, compounding EDI problems and ignoring sizeable groups of our population.
But as we say in science, correlation is not causation: we cannot say why the link profitability and diversity occurs. But if we follow the money we see that our strongest and most profitable future is one that actively works towards increasing diversity at every level.
Change starts here
Kajsa Asplund, Lead People Science at Alva says “The science is clear. The biggest risk for bias is at the start of the recruiting process, why evidence-based methods have the largest impact right here. Basing hiring decisions on data rather than on personal opinions means recruiting with higher diversity and better accuracy.”
By removing the possibility of bias from the first stage of recruitment, machine-learning systems make it easier for organisations and people to find the skills and potential of candidates, rather than share a list of places, names and faces that we struggle to respond to with neutrality.
The steps that we need to take to increase inclusion need to affect every stage of our organisations. And they all start with recruitment.