Our world has changed, but testing hasn’t.
By today’s standards, almost all ability
tests, personality questionnaires, assessment centres and structured interviews are unfair. Here’s why they are unjust; here’s why they are wrong…
What society considers fair isn’t fixed.
That’s the crucial point.
Once it was fair to commemorate slave
traders. Jokes about minorities were fair. Lower pay for women was fair. Ageist
discrimination was fair.
Today, none of those are fair, they are unacceptable.
Changing societal values have redefined fair. Many practices that used to be considered fair, aren’t today. Thankfully, much has changed.
But, our understanding of fair testing
and assessment has not changed. Organisations hang onto a definition of
fairness that’s entirely technical and insensitive. If majority and minority groups get roughly the
same average scores an assessment is considered statistically fair. If there are no
significant differences between samples - it is fair! And,
statistically fairness is, it seems, all that matters.
That concept of statistical fairness was decided upon decades ago by psychologists. Organisations and professional bodies have stuck with it. Sure, tests and questionnaires have been challenged and investigated, but always the questions have been about the psychometric technicalities of specific tests/questionnaires/centres. In other words, the question has always been is this or that assessment statistically fair?
The question has never been about the
appropriateness, relevance or value of “statistically fair”. Should we be
using “statistically fair” at all? Would a completely different understanding
of fairness fit today’s world much better?
Wrecks engagement, respect, listening, openness...
Let’s consider fairness from the person's - the "candidate's" - perspective...
Imagine you have applied for a job, and HR is “administering” tests. You’ve no idea of the tests' relevance - you are told only they are "valid" and “predictive”. Similarly, you’re clueless about how each test is constructed, scored and interpreted; you’re just told it’s statistically fair. And, told you don’t need to worry, the tests were developed by psychologists. However, you have heard people like you don’t do well on tests like these. Looking around, you can see applicants different to you who will do well.You are apprehensive; somewhat perplexed.
Imagine you have applied for a job, and HR is “administering” tests. You’ve no idea of the tests' relevance - you are told only they are "valid" and “predictive”. Similarly, you’re clueless about how each test is constructed, scored and interpreted; you’re just told it’s statistically fair. And, told you don’t need to worry, the tests were developed by psychologists. However, you have heard people like you don’t do well on tests like these. Looking around, you can see applicants different to you who will do well.You are apprehensive; somewhat perplexed.
Afterwards, you get a little “feedback” from HR, but there’s no helpful,
informative conversation about your results, or how the results have been used
to make decisions about you. There is no opportunity to challenge the tests, or
your tests' results, and you don’t have the know-how to do that anyway.
Throughout, no one seeks your views about what’s happening, or if its fair as you see it. One view of fairness - staticially fairness - has been imposed on you.
Throughout, no one seeks your views about what’s happening, or if its fair as you see it. One view of fairness - staticially fairness - has been imposed on you.
None of that testing process is
acceptable in today’s world. It rides roughshod over the things we value today.
It wrecks engagement, respect, listening, openness, honesty, trust and empowerment. A test may be statistically fair, but for the person the process is
horrible, demeaning and somewhat frightening. It’s “ethically unfair”. It
discriminates against vulnerable individuals irrespective of their ethnicity, gender, age... Yet, every day countless
individuals are subjected to exactly that injustice. Psychometric tests,
personality questionnaires, structured interviews, assessment centres… almost
all of them are unfair in a world that respects individuals.
I suspect many HR professionals and
psychologists will struggle to understand ethically fair. They won’t grasp
that assessment can be “ethically unfair” even for a white, young male from an
advantaged background. Ethical fairness is isn’t about statistical differences between
groups. It’s about the differences between what we do and our values. It is
about how we relate to people, and their personal experience of what we ask them to do.
Open Listening Simple Credible Responsible
An ethically fair process supports
and elaborates our core values. So, an ethically fair recruitment the process might be…
● Listening, respectful, collaborative - conversations, questions and curiosity are encouraged - there’s ample time for those.
● Simple, understandable, non-technical - so that everyone knows what’s happening and why.
● Credible, relevant, pertinent - so there’s absolutely no doubt about the value or authenticity of what's happening.
● Responsible, accountable, self-critical - the process goes out of its way to understand individuals’ experiences, fairness as individuals experience it, and what needs improvement.
I don’t know if that’s a complete or realistic outline of an “ethically fair” approach. (I’m new to this
too!) But, I’m pretty sure that’s what all of us want. To put it bluntly, we want to
be treated as a valued individuals, not as fodder for a technically determined
assessment process.
Note - I’m not suggesting statistical fairness is unimportant. It matters. No one should be using a process that discriminates. I am saying statistical fairness isn’t enough; we have to respect, value and work with others too.
Note - I’m not suggesting statistical fairness is unimportant. It matters. No one should be using a process that discriminates. I am saying statistical fairness isn’t enough; we have to respect, value and work with others too.
Get in touch!
Please contact me if you’d like to
explore this a little more. I think our present understanding of fairness - “statistically
fairness - is unwittingly discriminatory. Much of what we do today is wrong!
There's a quick "Contact me" form on the left. It would be good to hear from you.
Peter
© Copyright Peter Goodge, June 2020
