Biased, Broken and Full of Bullsh*t: The Case for Killing CVs (Backed by Science)

At Applied, we set out to build a fair, predictive hiring platform. The first thing we did was rethink how and when we use CVs. Here’s why YOU should do the same... ‍

CVs have been knocking about since the 50’s.

And behavioural science has come a long way since then. 

We no longer communicate via telegraph. 

We don’t work on typewriters anymore.

But the way we screen candidates has remained largely unchanged.

Very much unlike a fine wine, they’re NOT getting better with age.

Our case for rethinking the CV is pretty simple really: CVs are biased and they’re not predictive of skills.

You might feel like a lost, dummy-less infant without CVs to rely on.

But, isn’t it about time we severed ties with a primitive hiring method that doesn’t do candidates OR employers any justice?

So fasten your seatbelt and hold on tight, here’s why you should change the way you review CVs.

‍

Exhibit A: CVs have been proven to invite bias

Are personal details actually necessary?

CVs allow reviewers to discriminate against candidates, both consciously and unconsciously.

Maybe without even knowing it, you make snap judgments based on information like:

• Gender
• Age
• Name
• Home address
  

The truth is, details on a CV are used to find reasons to EXCLUDE candidates.

Dismissing people because they didn’t work at a big company you’ve heard of is bad.... very bad.

BUT, candidates are being overlooked because of details that have nothing to do with their experience or skills.

I know what you’re thinking: this sounds very pessimistic. However, this is the reality of what's happening.

When sending off a job application, women and people of minority races are less likely to be contacted, despite having the same qualifications as those with Caucasian and male names. (Carlsson & Rooth, 2007; Oreopoulos, 2009; Pager et al, 2009).

Research has shown that simply changing the name on an application can significantly alter call-back rates, even for employers with a public commitment to equal opportunity. 

Experiments in the US, Canada and Australia which involved sending fake resumes to real employers reveal that the response rates to otherwise identical applications can be up to 50% higher for ‘white sounding’ names. In the UK, that figure is just over 75%.

Studies have also shown that even when organisations share explicit diversity statements on their job descriptions/website, this still doesn't result in any better outcomes.

In fact, it may even have the opposite effect.

A study from 2016 tested whether or not employers with pro-diversity language in their job ads showed any less recruitment bias.

Researchers found that when applying to an organisation that presents itself as valuing diversity, minority background candidates were less likely to ‘whiten’ their CV.

However, these diversity statements were not actually associated with reduced discrimination.

And so ethnic minority candidates were less likely to 'whiten' their CVs - resulting in more overall discrimination.

It isn't just ethnic minority candidates that face unfair discrimination, your gender can also dictate your chances of being hired.

A US study found that university science faculties perceived women as being less competent and hireable than their male candidates - despite all other factors being identical.

So why don’t we just blank out names, addresses and other identifying information on CVs?

Well, some companies are doing just that - manually blacking out personal details.

Think about that for a second… our current CV process is so broken that people are going to town on them with permanent markers.

‍

Are you snobby about education and experience?

Bias isn’t just about discriminating based on physical differences, it comes in many forms. Here’s just a few kinds of bias that CVs perpetuate:

Affinity bias: the tendency to like people who are similar to us or remind us of someone we like. 

Maybe a candidate has a similar background to you, maybe they went to your university or held past roles like yours.

Confirmation bias: the tendency to search for, or interpret information that confirms one’s preconceptions.

You take a liking to a candidate and so you think everything they’re saying in their CV is spot on (whether it is or not). 

Halo effect: describes how judgements about some aspects of an object may influence how other aspects of an object are judged. 

Oh, this person went to Loughborough Uni and did a cool sounding degree, they must also be very good at their job too. 

‍

You might think you’re immune to these biases.

But chances are, if you're human, you're biased!

Why? Most of your biases are likely UNCONSCIOUS. 

Think about this: the average time spent reviewing a CV is 7.4 seconds.

How long do you take to browse a menu at a new restaurant?

More than 7.4 seconds I bet?

Why are you spending longer choosing between the cheeseburger or fish and chips than reviewing applications for your new developer or salesperson?

Perhaps it’s because 7.4 measly seconds is all it takes to scan a CV for disqualifying information and move onto the next.

‍

The order in which you look at CVs affects your judgment

It’s not just about the content on someone's CV. Even time itself can influence biases.

Even when using our own CV-less assessment method (more on that to come), we found that the order in which applications are assessed affected their ranking.

‍

Candidates scored first performed better.

Psst... that’s why the Applied platform randomises the order in which applications are reviewed.

‍

Exhibit B: CVs aren’t predictive of skills

In the words of former head of HR at Google Laszlo Bock, “It (the CV) doesn’t capture the whole person. At best, they tell you what someone has done in the past and not what they’re capable of doing in the future.”

And he’s right.

It’s a list of what you have already done. Not what you can do.

In a nutshell, this is why CVs aren’t predictive of future job performance.

Don’t believe us? 

Take a long, hard look at these findings from Schmidtt & Hunter's landmark meta-study…

(The numbers along the bottom represent the predictive power of each assessment method). 

Predictive validity is a measure of accuracy used in science and psychology.

It's a way of determining how well a given test or assessment can predict future outcomes.

Predictive validity studies can take years to complete, but the meta-study above gives us an overview of 100+ years of research.

The results are clear: education and experience aren't very accurate at telling us who will perform best, should they get the job.