Best Practices in Recruitment and Selection (According to Science)

Make sure referred candidates are still assessed 

Employee referrals don’t promote diversity.

Why? Because referred candidates tend to be demographically similar to the hirer.

They also benefit caucasian males over both women and minority background candidates.

This doesn’t mean you should scrap them entirely.

Referrals are 55% faster than career sites, and cheaper too.

However, you should be assessing these referred candidates in the same way as any other candidate.

Since referred candidates are generally similar to the hirer, you could use this to source more candidates from minority backgrounds.

Try getting referrals from people who are from these backgrounds themselves to improve the diversity of your candidate pool.

‍

Track your job boards like marketing campaigns 

Job boards can be expensive.

And not all job boards will bring you the same quality or diversity of candidates.

To make sure you’re getting decent ROI, there’s a pretty simple solution…

Set up tracking for your job boards.

Use UTM links when posting jobs, and see which job boards bring you the best candidates.

If you’re looking to boost diversity, you can also use specialist job boards that target certain demographics.

‍

CV screening

Don’t over-index on education and experience

When it comes to the CV screen, we’ve got some earth-shattering news for you…

The best practice is not to use CVs at all.

We’ve written extensively about what CVs have to go here, but we’ll quickly break it down again for you.

CVs rely on education and experience, but these are poor at predicting ability. 

The single most predictive form of assessment is known as a work sample.

They’re essentially like taking small chunks of the role and getting candidates to simulate them.

Work samples usually take the form of hypothetical questions/ tasks - you pick out a task that the candidate would be doing in the role, and then ask them how they’d manage it.

‍

Anonymize applications

The traditional CV process not only fails at predicting ability, but it also disadvantages candidates from minority backgrounds and women.

Take the results of Nuffield College's Centre for Social Investigation’s research: 

When identical resumes were sent out en masse, 24% of white British applicants received a call back from employers, compared to just 15% for ethnic minority candidates... 

• Pakistani heritage: needed 70% more applications
• Nigerian and South Asian heritage: needed 80% more applications
• Middle Eastern and North African heritage: needed 90% more applications
  

Another study involving science faculties found that gender bias is also at play when using CVs. Even though applications were again identical, female candidates were perceived as being less competent and hireable than their male counterparts.

This is why we’d strongly suggest anonymizing applications and adopting a blind hiring process. The easiest way to do this is by using a platform like ours, which you can now try for free.

‍

Interview best practices

Try structured interviews

Due to our unavoidable unconscious biases, most interviews are far from objective assessments.

'Gut feeling’ is essentially bias.

Although when meeting a candidate face to face (or via Zoom), anonymisation isn’t an option, there are a few steps you can take to mitigate a large amount of bias and make interviewing more predictive.

Our interview best practice is to use structured interviews.

This means asking all candidates the same questions.

Given that education and experience aren’t very useful when it comes to assessment, we use work sample-style interview questions.

You could also give candidates a task to perform or think through.

Are there any upcoming projects or past issues that you could ask them to work through?

If you’re stuck for ideas, we gave away a heap of interview questions we’ve used here.

To prevent the interview from veering off into unnecessary tangents and to ensure you actually come away with an idea of who will be progressing/hired, give yourself scoring criteria for each question.

If each question is tied to one of the skills you want to test for, then you’ll be able to build a skills map for each candidate.

It’s also worth keeping in mind that interviews are meant to be a chance for candidates to show off what they can do.

You want them to be at their best.

So, avoid trick questions or piling on any unnecessary pressure.

It’s an interview, not an interrogation.

‍

Have 3 interviewers

Generally speaking, the judgment of a group is more accurate than that of an individual.

This is known as ‘crowd wisdom’. 

This is why its best to have 3 interviewers (any more and you’ll see diminishing returns).

If you’ve got your criteria, each interviewer can independently score candidates question by question.

When you average out the scores and add them up for each candidate, you’ll have a much more accurate picture of their strengths and skills than if a single interviewer had conducted the interview.

Not everyone needs to be fully involved, but it’s best that all interviewers introduce themselves and ask at least one question so that they don’t come across as looming, silent judges that put candidates off.

‍

Diversity recruitment best practices

If improving diversity is a priority for you, then keep in mind these guidelines:

• Use the networks of minority background employees by asking them to refer candidates
• Use specialist job boards
• Be open about diversity initiatives on your careers page
• Anonymise applications
• Have 3 interviewers
• Track how candidates perform across progress to find any stages that disadvantage a particular group
• Look for culture add not a culture fit

‍

Here at Applied, we’re on a mission to make hiring fairer for candidates and more predictive for employers. We’ve replaced the biased hiring methods of yesteryear with the most predictive forms of assessment behavioural science has to offer. Find out more via our resources or start a free trial now.‍

‍