THE BIG IDEA
I really wasn’t planning on doing back-to-back editions on recruitment
(this one was due to be about the split personality of the CHRO – so, sorry, you’ll have to wait until next time!)
but some fascinating data was published this week that:
- is really important for the value we are able to add; and…
- provides some good news for HR
So, with that explanation in place, let’s dive in.
I saw this article on LinkedIn: Recruiting bias shown to exist, and read a few of the onward shares – finding great commentary on the article.
At my core, when it comes to transforming HR and the value we may be able to add, you’ll always find me starting at the data.
In this case, I clicked through to the original research brief which I’m going to link in 2 versions:
- Online – A Discrimination Report Card (Kline, Rose, Walters, 2024)
- PDF – A Discrimination Report Card (Kline, Rose, Walters, 2024)
(the first is a really well presented online interactive version of the findings – well played web team, well played! – the second a more traditional PDF, good for sharing in static form)
I’m actually going to pivot from the data fairly quickly here – being an applied scientist, after all – but here are the key findings.
- A HUGE data-set: 83,000 fictitious applications, 11,000 entry level job openings, 100+ Fortune 500 companies!
- A small number of companies are responsible for a substantial amount of contact discrimination
- Firms were estimated to favor white applicants over black applicants by 3%-24% (compared to up to 50% as measured in the early 2000s)
- Over half of the variation in racial discrimination is explained by industry
- Gender disparity is less widespread, however a small handful of firms favor either male or female names
All interesting stuff.
But the nugget of info that we’re going to be talking about in this edition of Improving HR comes from the NYT article which includes interviews with the researchers:
Several common measures — like employing a chief diversity officer, offering diversity training or having a diverse board — were not correlated with decreased discrimination in entry-level hiring, the researchers found.
But one thing strongly predicted less discrimination: a centralized HR operation.
The researchers recorded the voice mail messages that the fake applicants received. When a company’s calls came from fewer individual phone numbers, suggesting that they were originating from a central office, there tended to be less bias. When they came from individual hiring managers at local stores or warehouses, there was more. These messages often sounded frantic and informal, asking if an applicant could start the next day, for example.
“That’s when implicit biases kick in,” Professor Kline said. A more formalized hiring process helps overcome this, he said: “Just thinking about things, which steps to take, having to run something by someone for approval, can be quite important in mitigating bias.”
We in HR should find this really, really reassuring; a sign that we can, and indeed do, make a difference.
So, let’s pick it apart to see what how we may go further.
TRY THIS
Maybe surprisingly, I’m not starting with HR. This is simply because there is one sentence that just jumps out at me:
When they came from individual hiring managers at local stores or warehouses, there was more. These messages often sounded frantic and informal, asking if an applicant could start the next day, for example.
Now, I’m not going to pile in on managers – they are, after all, doing a pressured job; not to mention that this research was happening in the aftermath of the pandemic shut-down and re-gearing of the work-force
(they don’t get a pass, BTW, it’s just that this edition is not about them as the issue – well, directly at least)
But there is something here: a chaotic approach to recruitment.
What I read in this section is not a JIT supply-chain-esque process – instead I hear panic mode, and I hear people left to do whatever they think is the right way to recruit.
In other words, left to discretion, in a significant proportion of situations, standards will inevitably drift downwards
(and I might suggest that this drift is back to organizational cultural norms, which may be why industry is the major a contributing factor in racial contact discrimination?)
So now we get to HR. And we have to note what the researchers tell us.
The smoothing effect of centralized recruitment operations appears to be from a neutralized process, not mandated compliance.
In other words, in this case, HR was adding value by doing things right up-front, not by simply protecting against bad apples.
This is really important for anything we may choose to consider doing based upon these research findings.
Do The Right Things… Do Them Well
(now, why does that sound familiar? 😊)
We can fix a lot of this through process!
What can we be sure of? Well, we know that every human being has implicit biases, and that it’s largely impossible to fully eradicate them.
We also know that, sadly, many companies are still stuck in the loop of believing that isn’t the case; that there is some halcyon state where no pre-judgement occurs, where everything is objective and non-emotional.
In such companies, where the aim is a post-discrimination ideal state, DEI is often positioned as a panacea – more often than not by executives than by the DEI/HR professionals doing the work.
As the researchers say, though, the presence of DEI doesn’t appear to correlate with the findings, and these companies are unlikely to see the shift they desire.
The answer here is process – simple, straight-forward process.
And the question we need to ask is:
At each stage of our hiring process, how do we reduce the impact of implicit biases?
USE THIS
Now that we’ve narrowed down the question we’re asking, let’s break the hiring process down into “modules”:
1 – Role Definition
The first step to reducing bias here is to have a clear role definition
(from my work with candidates, I still find a lot of postings that appear to be based on a Job Title alone)
so back to basics here:
Job Specification – Role Summary, Key Responsibilities, Working Conditions
Person Specification – Skills, Capabilities, Experience, Motivation
2 – Attraction
- Inclusive channels and advertising
- Monitor employee referrals for variance (i.e. that we are not just recommending more of the same)
- Sharing diverse success stories
As an aside, if your recruitment team aren’t working with/learning from your marketing team, you are missing a huge opportunity here.
3 – Application
Given that the research didn’t find technological/system correlation, you might think there’s nothing to lean on here. However, consider access to application routes, language-of-commerce within system, ability to enter non-local qualifications, etc. Is you application system truly accessible?
4 – Pre-Screening
Increase objectivity, decrease human interpretation. Be careful of the algorithms that are scanning applications – have they been built upon implicitly biased data-sets?
5 – Screening and Interviewing
Once we hit this stage, we’re beyond contact discrimination, and into a whole other area of bias-impact.
Of course, bias awareness is critical, and monitoring of decisions for biased conclusions.
Beyond these modules, though, as I was thinking about the article and data here – which is, after all, about pre-judgements made upon ethnic inference of names – I got to thinking about how the written name provides the greatest opening for implicit bias.
Given we work with what are considered complex
(nearly intelligent)
systems, I have to ask:
Is there a reason that any anyone in the interview process needs to know a candidate’s name until the candidate enters the first interview (whether in person or via Zoom)?
Surely a Candidate A, B, C approach serves the same purpose, right up until the person is in front of us?
I’ve sat with this question overnight and any devil’s advocate answer sounds like a justification for allowing implicit (or explicit) bias to have a seat at the table.
So, given that, we really are talking about how to systematically hold back names of candidates until interview.
(and, while we’re at it, why don’t we hold back date of birth as a hedge against age bias?)
That’s doable in an ATS, if the will to demand the functionality is there.
(full disclosure, I haven’t been on a deep dive through current ATS, so the functionality may be there – though the research suggests that, if it is, it’s far from the norm)
How about the next conversation with your ATS provider starts with: “How do we anonymize candidates thru screening?”
FOOD FOR THOUGHT
“Systems do not maintain themselves; even our lack of intervention is an act of maintenance.” ~Sonya Renee Taylor
