Back to Listing

I see colour - the insidious nature of unconscious bias

01 August 2012

So, how do I know I’m not perfect in the liberty, equality, fraternity stakes? I know because when I answer a question about do I see someone’s colour, or have I got any prejudices about what boys and girls are capable of doing, or should older people driving on the highway and pulling out in front of me without indicating be allowed a license, I look at my audience and judge what answer shows me in the best light – even when that audience is me.

So what’s a bit of self-delusion among friends? 

A lot it seems, and I now have ‘irrefutable’ proof I need to work on quite a few hidden biases. My results in the Implicit Association Tests (IATs) I took online recently, tell me so.

(I love a test. They push my blood pressure up way beyond what is safe and when I finish, I feel I’ve done a good day’s research work. Our Ruby of the Month, EOWA Director Helen Conway, also admits a guilty pleasure in doing ‘personality’-style tests – the ones that reveal something about the way you tick. She admits she has a certain bias toward working with people who are keen to find ways of working flexibly. She says that means she has tended to favour women at the expense of men, because it is women who most want to change the workplace to suit the way we live and drag it kicking and screaming out of the dark male ages.)

My IAT ‘research’ all began when a friend sent me something she thought I should read on leadership – it’s what friends do when they know you’re a home-alone worker on a website devoted to ‘diversity issues’. 

It was a report on positive leadership strategies included in a Deloitte Diversity Inclusion newsletter and in it was a link to the US-based online entity, ‘Project Implicit’. 

The name itself was alluring enough. So, I clicked on for more.

According to the Project Implicit site, this “non-profit organization and international collaborative network of researchers investigates implicit social cognition – thoughts and feelings outside of conscious awareness and control.” The tests uncover “attitudes, stereotypes and other hidden biases that influence perception, judgment, and action”. The results, when analysed, can lead to the formulation of practical strategies for “addressing diversity, improving decision-making, and increasing the likelihood that practices are aligned with personal and organizational values”.

At my last count there were 14 different Implicit Association Tests. They assess attitudes toward everything from race, colour, weight, age, homosexuality, even weapons, to US Presidents.

I completed the “Fat–Thin IAT” with a result that was eye-opening, and unflattering.

I had a “slight automatic preference for Thin People compared to Fat People”.

According to my response speed and accuracy: I “responded faster when Thin faces and Good words were classified with the same keyboard letter than when Fat faces and Good words were classified with the same key.” 

Apparently: “Most respondents find it easier to associate Fat People with Bad and Thin People with Good compared to the reverse.” My data suggested, “a slight automatic preference for Thin People compared to Fat People.” 

18% more likely it seems.

What was even more worrying as well as interesting in a very unappealing way was my result for the “Gay – Straight IAT”.

The data suggested a “moderate automatic preference for Straight People compared to Gay People”.

27% more likely: internalised homophobia remains rampant in us all. 

Worse than my result though is the fact that lab studies, according to the IAT guys, “show that the social acceptability of negative attitudes toward gays has changed relatively little in recent years. We also know that anti-gay attitudes are observed on measures of implicit attitude such as the IAT, and that a person’s conscious and implicit attitudes toward gays are more often in agreement with each other than they are for some other socially significant domains.”  

In other words, people aren’t even attempting to be positive or covering their negative feelings for gay people in society or the workplace.

My result in the “Four-Category Race (brief) IAT” offered a tiny bit of light at the end of the tunnel.

I was more positive about people in this order: Hispanic, Asian, White, Black.

Were my test results in this IAT similar to my conscious beliefs? 

In some cases I’d like to think so and in others not. It would have been better if I’d seen no colour at all and I was equally positive about all people.

But I am not.

However, I am now well aware I have a bias or three. My ongoing strategy: align practices with values more honestly and find ways to mitigate those biases.

Check your own implicit associations at

also have a look at this link on unconscious bias in Australian workplaces and gender