Hiring Biases

We all have our biases, the problem is that our biases limit our options and that’s not a good thing.

Think Progress had an article title: “People can’t be trusted to make unbiased hiring decisions.”  (http://thinkprogress.org/economy/2015/05/13/3657911/unitive-hiring-bias/) This is pretty much true. Our baseline for who we trust – is ourselves so we look for other people like us. And that’s not good for business.

Diversity in the workplace is challenging, but it brings many benefits. The most important benefit being that your decision making processes are improved when you allow diverse viewpoints to weigh in. Diversity isn’t a cure for doing stupid things, but it helps. People who don’t think like you are more likely to tell you when you are being stupid, if you let them.

And to let them, you need to hire them. You have to word your advertisements in such a way as to not discourage certain individuals from applying. Your hiring process needs to not take into account someone’s protected status, which is actually really hard to do. And then you try to be equal and still – your workplace is not as diverse as it should be.

So we use tools to help us avoid our biases. One of these tools is the personality assessment, which is designed to help us eliminate bias. But the tests are biased. In fact, they are generally rated as a good test if their bias level is moderate instead of severe.  The argument for such flawed tests is that if the tests are less biased then regular human bias, it’s a move in the right direction towards less bias.  (See: http://www.slate.com/articles/business/the_dismal_science/2009/03/testing_testing.html)

Companies who use personality tests should at least make sure that their tests are scientifically validated to reduce not only the bias, but to ensure that what you are testing for are actual personality traits. (https://humanistlearning.com/personalityassessments/)

So what is the solution to hiring bias?  How about adding in random selection. Once the basic criteria for hiring are met – choosing between equally qualified candidates, when done by humans, reintroduces human bias. There is research that shows having bad reasons for making a choice yields worse results than making a random choice. (http://humanisthappiness.blogspot.com/2014/11/randomness.html) So why not chose randomly if everything else is equal.

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...