02 September 2020

I Am Prejudiced

So are you.

By Richard Pooley

Before you “cancel” me for such an outrageous accusation, ask yourself this: are you really so saintly that you are not prejudiced against any group of people, whatever the colour of their skin, their sex, their sexual orientation, their culture, their religious or political beliefs, even their accent or the clothes they wear?

Assuming that, like me, you admit that you are no saint, let’s take a look at the prejudice which has been dominating so much of the news these last few months, including the coverage of the coronavirus pandemic: racism.  A warning though before we continue: this article is in two parts.  There is too much to cover in one piece.  Besides which I am hoping you will take a short test in the days between this week’s Part One and next week’s Part Two.

Racism is only one form of prejudice, albeit one of the most pernicious ones.  It’s a word which is applied far too liberally these days to many forms of prejudice.  The Concise Oxford English Dictionary defines it as “Belief in superiority of a particular race; antagonism between different races.”  Which begs the question; what is “race”?  Several dictionaries give the same answer to that one: “A group of people identified as distinct from other groups because of supposed physical or genetic traits shared by the group.”  They go on to say that most biologists “do not recognize race as a biologically valid classification”.  Most people however, including biologists, would agree that in common parlance race refers to one thing: what you look like, particularly the colour of your skin.

I’m a 68-year old white English man who not only went to an independent school but could afford to send my children to such a school.  I am hugely privileged.  I know my pinko-white skin has been one of many factors which has made it easier for me to succeed in life.  Even so, I have always liked to believe that though I do have prejudices, being racist is not one of them.  I spent much of my teenage years and twenties living and working in southern and eastern Africa.  I witnessed the appalling racism of South African whites towards their black, brown and mixed-race compatriots at first hand and subsequently was an active member of the Anti-Apartheid Society.  As a teacher in Botswana I also experienced the racism of my black pupils from the local Bakwena tribe towards their lighter-skinned and smaller San (or Bushmen) classmates.  I have spent much of my working life training businesspeople to understand the values and behaviour of people from different cultures in order that they could live and work successfully in those cultures.  This has meant tackling head-on the negative stereotypes, often racist, that those businesspeople had of people who looked and behaved differently from themselves.  I have felt the cuts of racism myself.  The subtle prejudice shown by the Japanese towards white Gaijin is as nothing compared to the daily, overt racism experienced by black-skinned people in Japan; but it can still hurt.

To find out if I am racist, I decided last week to do something I have been putting off for years: take the Implicit Association Test.  Not heard of it?  It’s a test originally devised by a trio of US psychologists in 1998 to determine someone’s level of subconscious racism.  It has since been expanded to check how subconsciously prejudicial a person is in seventeen other areas – eg sexual orientation, young v old, skinny v fat, even “presidential popularity” (one for you,  J.R.Thomas, though you would have to sign in as an American to do it?).

Around twenty million people have taken at least one of these tests in the past twenty-two years.  The test which has had by far the biggest influence on politicians, business executives and educators, especially in the USA,  is the original one, the one I took: the race IAT or ‘Black – White’ IAT.  3.3 million Americans completed the US version of this test online between 2003 and 2015, many of them while undergoing Diversity training, designed ostensibly to rid them of their subconscious (or implicit) racism.

Project Implicit is the organisation which runs the tests –

https://implicit.harvard.edu/implicit.

Yes, Harvard.  The Head of Psychology at Harvard University, Professor Mahzarin Banaji, has been a vocal backer of these tests from the start and gives no quarter to those who question their efficacy.  So, it was natural that the IAT’s creators partnered with her and Harvard to get the IAT more widely known, used and accepted.

If you want to take the race IAT yourself, click on that link and set aside 25 minutes – 5 to understand the instructions, 3 to decide how much you want to reveal about yourself and your opinions in answering the first two questions, 2 to check whether you have really understood the instructions when you come to the first “question” in the test proper, 5 to complete the test proper, 2 to read and absorb your results, 3 to decide whether you are going to compete the incomprehensible post-test questions, and 5 to pour yourself a strong drink.  Don’t do as I did and have the strong drink first.  Have your results ready by the time you read the second part of this article.  By the way, this is a test supposed to be taken by anyone of any skin colour.  It does not discriminate.  At least, not explicitly.

I have to admit that I did not approach the test with a mind consciously cleared of bias towards it.  I am a long-term sceptic when it comes to psychometric tests.  If you have ever taken the trouble to investigate the origins and research basis of some of them – the Myers-Briggs Type Indicator, for example – you will know just how flaky many of them are.  MBTI, beloved of so many otherwise sound companies and used in training, recruitment and personnel assessment, has been described by social scientists as “bullshit” and “little more than a Chinese fortune cookie”.

The IAT has been under attack by psychologists and other academics almost from its inception.  The best layman’s evaluation of the IAT I have come across was written by science reporter (and white, male) Jesse Singal in the New York Magazine in 2017 –

https://www.thecut.com/2017/01/psychologys-racism-measuring-tool-isnt-up-to-the-job.html.

It’s very long but well worth reading.  Singal’s conclusion?  That the race IAT is dangerous pseudoscience.

A pseudoscience because it clearly does not meet the two most important criteria in any test: high levels of reliability and of validity.  If such a test were reliable, you would expect someone who takes it, waits a bit and then takes it again would get a similar result.  Not so.  Even the research of the race IAT’s proponents concludes that a person is more likely than not to get a completely different result second time round (perhaps still showing implicit racism but in the opposite direction!).  To be valid, the race IAT should be shown to be measuring what it is claimed to be measuring: people’s subconscious prejudice towards those of a different skin colour.  In other words is the test able to predict a person’s discriminatory behaviour?  The answer from nearly all the studies and every single meta-study is No.  Again, even its supporters accept this to be true.  Yet participants on Diversity courses continue to take a one-off race IAT, which four times out of five shows that they are at least mildly racist, whatever they have said to their colleagues before they took the test (over half say they are not racist).  This is the aspect of the IAT which Singal considers dangerous, something I want to come on to next week.

So, what did I have to do and what were the results?

The preamble to the test gave me the instructions and a warning that I must be prepared for results that I did not like.  Off I went.  I had to press either ‘e’ or ‘i’ when a word or image of a person came up, depending on whether it was deemed ‘good’ or ‘bad’.  Each time there was only one correct answer.  I was not being tested on whether I was correct or not.  It was my reaction time that was being measured and the results passed on to an algorithm.  The algorithm generated my score.  And here are my results:

“Your data suggests: Moderate automatic preference for Black people over White people.”

So, all that exposure to black Africans has made me prefer to be in the company of black-skinned people than people who look like me?  Hmm.  Does this mean I subconsciously discriminate against white people?  Hmm again.

I took the test again yesterday, four days after the first.  Here are the results:

“Your data suggests: Little or no automatic preference between White people and Black people.”

I hardly dare take a third test but I will.

Have a go yourself and join me next week for part two.