EMB
We recognize from long experience that a small sample of
athletic behavior is an imperfect guide to athletic performance.
In the corporate context, unless you interview lots of people
and keep careful track of how well the people you hire did in
the interview compared to how well they performed in the
actual job over time, you’re not going to have the data to show
that interviewing can be ineffective in predicting performance.
Incidentally, what we observe in interviews is often not the
behavior we’re interested in predicting – namely job behavior.
Instead, it is interview behavior, which is not at all the same
thing. Extroverts tend to ace interviews, but that may not be a
very important attribute for the position we’re looking to fill.
The problem is compounded when we interview someone from
a different culture, which may have self-presentation rules
different from those of our own. For example, Westerners (and
perhaps especially Americans!) are inclined to beat their own
drum, while people from East Asia and Southeast Asia tend to
behave more humbly.
2. Regression to the Mean
I’ve often heard people speculate about reasons for second
season slumps in sporting performance. In his or her debut
year, an athlete can rise to star status, yet the season later,
fail to wow the crowds.
A common remark is, “The guy gets too cocky and slacks off.”
But let’s think about how somebody becomes the very best
athlete over the course of a year. Having substantially more
than average talent among the other newcomers is surely a
big part of it. Yet, more often than not, everything else also
goes right for that guy: he gets particularly good coaching at
the beginning of the year, he happens to perform very well in
his first two or three games which builds his confidence, he
has no injuries, and he gets engaged to the girl of his dreams.
By contrast, the next year the great dice roller in the sky gives
the same player an elbow injury, and his fiancée jilts him.
Somebody else gets all the breaks next year.
The relevant principle here: regression to the mean. This is a
statistical principle which tells us that an observation that is
significantly higher or lower than the average will be closer to
the mean if measured a second time. This type of regression
is always to be expected for any behavior or event with a
chance component.
In sum, regression to the mean is the tendency for scores or
performance to average out when measured multiple times.
The term was coined by Sir Francis Galton when measuring the
heights of parents and children. He found that while there was
a tendency for taller parents to have taller children, and shorter
parents to have short children, at both ends of the spectrum, their
children’s heights were usually less extreme then their own.
In a workplace context, it is important for decision-makers
to consider this principle by focusing on the average or
“mean” performance of an employee rather than focusing on
individual achievements or mistakes, which could be isolated
occurrences. In a recruitment scenario, it can be a mistake
to offer a huge salary to hire someone who has had a brief if
meteoric career with another company. And on the other hand,
it may be wise to be patient with a new employee who isn’t
living up to standards set in a previous job.
3. Opportunity Cost
Many years ago, my wife and I bought a summer cottage, but
we didn’t have enough money to buy furniture. So, I decided to
make furniture myself. That would be a good idea for someone
with carpentry skills. Unfortunately, that didn’t apply to me, so
I took a woodworking class. Fifteen hours later I had produced
a box. It became clear that I was paying a huge opportunity
cost for attempting to make my own furniture.
Opportunity cost, economists tell us, is the cost of forgoing
the best alternative action. There were actually dozens of
alternatives that made more sense than making my own
furniture. For instance, it would have been far more economical
for me to buy inexpensive furniture at a big box store and eat
cheaply for a few months so I would have enough money to pay
for it, and then buy better furniture as money became available.
This path would have allowed me to do any number of other
things that would have been more valuable to me than making
(inferior) furniture.
EMBreview.org 73