Being the best, it is a topic I heard about a lot this week. At the Oscars, in the American presidential race and closer to home in a discussion on diversity at my research group.
Should we give an Oscar to a black actor? Should you vote for Hillary because she is woman? Should we not hire another white man in our group?
No!! Oscars, presidencies and jobs should go, simply, to the best candidate.
When I was younger, I used to do competition dancing, and, as in the other three categories, you win dancing by getting votes, in this case from judges. In the competitions, there were always those complaining that the judges were unfair. Someone got a vote, but they did not even have x or do y. My partner and me used to say: if you want a fair sport, go do long jump, so you can clearly measure who’s best. As soon as human judgement is involved, there is no “fair”. Some people have amazing moves but lack technique, others execute steps perfectly but are bit boring. Who is ‘better’? Matter of taste, or of when exactly the judges look your way.
There is no best
In matters so complex, you cannot just say: the job should go to the best candidate. There is no best. Bernie might have better ideas, but Hillary might have experience and get the job done. I guess we all agree Leonardo deserved an Oscar because it was ‘his turn’, but was his gritty bear fighting better than Damon cursing on Mars? Is a professor with high-impact projects but 0 patience better than a candidate that excels at teaching but has fewer grants?
My take is that how you value candidates at such a high level depends, by and large, on the things you value as a person. Not so much on the candidates.
Here are some properties a presidential candidate might need: Knowledge, Perseverance, Intelligence, Strategy and Wellspokenness. Two candidates of equal value (given that there even is such a thing!) might look like this:
Now, how I view them also depends on my personal values. Let’s make things mathematical and call this the value vector. I might value Intelligence and Strategy over Knowledge, so I view the candidates with vector: (0.5, 1, 1.2, 1.2, 1). You on the other hand value Knowledge highest, viewing with (2, 0.8, 0.8, 0.5, 0.8). On a neutral candidate (all values 50) our different views look like this:
Not just taste
And this is not even about how much you value a skill, but also in what bucket you put observed behaviour. Going a bit deeper, and no longer talking about hypothetical candidates, I might be able to see ‘Perseverance’ in Hillary, where you might see ‘Strategy’. Values, but also experiences makes people see certain skills in a certain light. If you are smart, you’ll like intelligence. If, like me, you are frequently living through annoying remarks on your gender, you might really really value Hillary’s perseverance after being ridiculed for pants suits.
Finally, your value vector is likely to be further colored by some implicit bias, see here and here and here and here. For the same behaviour, you might score a man 50 while you would give a woman 40.
Perception is reality
This is why a diverse representation of people voting matters so much. Given that this is not a game of long jump, there rarely is an objective better. Do you really think that Leonardo is ‘better’ than Matt, or Bernie is ‘better’ than Hillary? That we can even assign a number to how good they are? All have their own unique mix of skills, talents and weaknesses. I hope next time you say ‘the best candidate’ think of what your value vector is, and how it might be colored. Where possible (for example in a job interview) maybe people should firstly test their value vectors in a few neutral, fake candidates. I think that will greatly aid a discussion of who’s ‘best’.