Ithuta JV

Source - Article by Bob Andrew

Does your project team ever ask you why you never praise them when they do a really good job? Is it because you believe praise does not work? Every time you have praised a member of your team for some good work, the next piece of work you receive is always mediocre. Thus, you believe praise not only does not make the team member any better at his work, in fact it makes him worse.

On the contrary, you might also have noticed that berating a team member after a very poor performance improves their subsequent performance.

This type of management behaviour: avoidance of praise after a good performance and beratement after a poor performance, is often connected with a statistical tendency, known as ‘regression to the mean’, where extreme behaviour (or scores) often return toward the average. Examples are ‘the rookie slump’, where the level of an athlete or sportsman drops off after their initial year and the so-called ‘Sports Illustrated Jinx’ where a sporting team that has a great year with loads of publicity, performs significantly worse the next year. In the case of the rookie, the supposed slump in performance is not really a decrease in performance, but rather a move toward the average performance of the player. In the case of the sporting team, despite one exceptional year, in time, the performance of the team will return to their average level of play.

In his book, The Drunkard’s Walk: How Randomness Rules Our Lives, author Leonard Mlodinow describes how all manner of events, sporting events, gambling, betting, box office successes, and probably project team performance as well, follow a property of regressing to a mean. This means that although a single event or a small sample of events in succession, may appear to be unique, if you analyse these events over a long enough period of time you will note that aberrations (also called outliers) are quickly compensated for by other outcomes that return the pattern to some underlying mean. It also means that extraordinary events are most likely to be followed, due purely to chance and randomness, by a more ordinary one. In fact, if regression to the mean did not happen, everything would go out of control: extraordinary performances would be followed by more extraordinary ones and inferiority would escalate: we would live in a world of progressive extreme extremes.

Praise should always be levied where necessary, but exceptional performance should not necessarily be expected to follow exceptional performance. Poor performance should be commented on, in a mentoring environment, and when poor performance follows poor performance, there being no return to the average performance, there should be an intervention and this would indicate a problem.