Eliminating unconscious bias in coding
Unconscious bias is a deeply ingrained, automatic thought process created from stereotypes which can affect decisions.
It can affect the decisions of employers and potentially put candidates at an immediate disadvantage before they have even had a chance to prove themselves.
Shockingly, a recent study by Nuffield College's Centre for Social Investigation (CSI) reveals that British citizens from ethnic minority backgrounds have to send 60 percent more job applications to get a positive response from employers compared to their white counterparts.
Even though many people would say they do not hold prejudices against groups of people who are different – whether that be in the race, weight, religion, gender, etc. – research from Project Implicit shows that the majority of people do hold unconscious biases.
DID YOU KNOW?
Until the 1950s, IT programming was a female-dominated industry. This was a time when maintaining computers involved extensive preparation, planning and learning from logical diagrams which were involved in maintaining computers. However, in the 1950s and ‘60s, employers began relying on aptitude tests and personality profiles that weeded- out women by prioritizing stereotypically masculine traits and, increasingly, anti-social ness, rather than an aptitude to do the job well.
Over the decades, perceptions have shifted and entrenched men as the de facto leaders of the IT industry. This persists to this day, perpetuated each time by stereotype and unconscious bias during the employment phase.
Unconscious bias training has been a go-to diversity solution for huge companies, such as Google, for a number of years. But it’s not a perfect solution. Training has only a limited effect on overall behavior and, although many recipients feel they will now go into interviews conscious of bias, unconscious bias is an unintentional process.
However, technology and AI could provide the solution needed to combat unconscious bias, whilst improving companies’ profitability, efficiency and reputation.
Logical hiring based on practical skills will also attract the best minds. With the best minds, comes the greatest innovation and development. A company is future-proofing itself as it works to remain competitive with/keep ahead of others in the market.
There are some examples of companies starting to use technology to combat bias. Furhat Robotics created Tengai, a robot designed to replace human interviewers. The robot sends a transcript to the HR team who can then decide based on it. This way, the human HR team has very limited interaction with the candidate.
However, this is still somewhat flawed as the employer would still need to sift through applications to see who they wanted to offer an interview.
Amazon also tried AI for hiring – but as men have typically dominated the tech industry, the algorithm soon concluded that women weren’t as good for the job as men were and automatically filtered them out.
The answer lies in simplicity. Instead of trying to re-wire the brain, or bypass it by using technological “third-parties”, start, and end, with the simplest of datasets: can an applicant do the job they’re applying for?
If we hold the tech equivalent of a blind audition, we are separating an applicant’s physicality or background from their ability to fulfill a role. Therefore, only those who can do the job are shortlisted.
The companies who can offer this method to employers are the companies who will raise the global labor market above the practice of unconscious bias, leading the way for a smarter, more efficient workforce.
As well as removing unconscious bias, this method can raise a company’s profitability. Not just due to the fact it's their new employee has the required skills which cut down on training costs – but also the fact a company it can hire someone in confidence that they can add value from day one.
The ability to hire someone with proven skills lowers the chance of having to re-advertise, re-interview and re-hire for the role -. which can put a business under productivity and financial pressure.
Technological “blind-auditions” could also improve reputation. Being confident that each every one employee has proven themselves to be competent will create consistency and reliability – something clients and customers should be able to depend upon.
After this becomes common and best practice, it will change the conversation around skills in the industry and change the defining characteristics of the word “who” in regards to recruitment. Instead of “who” being partially defined by someone’s social or racial background – the “who” becomes the level of talent and other factors such as cultural fit and aspirations within the business.