Thursday, September 09, 2021

Bad Sorting

A few days after my post on The Corporate Sorting Hat, I found myself arguing on Twitter about the ethics of using the Myers-Briggs Type Indicator (MBTI). 

@swardley and @Jon_Ayre argued that MBTI was not merely unscientific, based on dubious theory, but positively harmful. Jon made the following points.

1. Using the instrument to select people for management positions results in incompetent managers being appointed.

2. If candidates know which MBTI type is preferred, they have an incentive to lie in order to improve their chances of selection.

 

As I see it, bad candidates can lie through their teeth, and bad managers can be appointed, regardless of what selection process is used. And there are undoubtedly organizations whose management is dominated by liars and cheats, without having used MBTI at all. So these arguments seem to depend on MBTI producing significantly worse results than any of the commonly used alternatives.

However, if MBTI is so unreliable, it may occasionally be capable of producing better results than any alternative. It's worth noting that good candidates sometimes fail to present themselves in the best possible light, and a process of this kind might conceivably give them a second chance.


Jon also argued that it was unethical to make important decisions about people based on an unscientific and unreliable instrument.

If you have a scientifically credible and reliable alternative, then perhaps it would be unethical to use anything else. But the reality is that most selection processes are just as unfair and unreliable as this one, if not worse, so how much difference does it actually make?

 

And then there's Facebook.



Rory Cellan-Jones, Facebook accused of allowing sexist job advertising (BBC News, 9 September 2021)

Victor Lipman, The Disturbing Link Between Psychopathy And Leadership (Forbes, 25 April 2013)

David Robson, How narcissists climb the career ladder quickly (BBC Worklife, 1 September 2021)

 


No comments:

Post a Comment