On Mon, 13 Oct 1997, Karl V. Koenig wrote:
> The point of another contributor about salary is always a major area of
> contention. There's nothing subjective about the number of dollars on a
> check. And one would like to think there's some relationship between
> one's job performance and the pay for it. The solution here is a relative
> ranking (subjective) of individuals within an engineering specialty. That
> provides at least a starting point for debate over what my salary should
> be and what kind of pay raise it would take to get there.
The problem with measurement based on relative ranking is that it doesn't
measure how well one performs. Instead it measures how well one performs
*compared to one's co-workers* which is something else entirely. A
compensation system based on ranking rewards an achiever surrounded by
slackers more than it does an achiever surrounded by an achiever. It
creates a situation where it's counter to your own interests for your
colleagues to perform well. It encourages employees to compete against
each other for shares of a fixed salary budget; this leads to
credit-grabbing, peer pressure not to "pull up the curve" and sometimes
even sabotage. It forces supervisors to give negative evaluations to
people in order to stay in budget.
By mathematical definition, 10% of a workgroups will *always* be in the
top 10%, and 10% will always be in the bottom 10%. Is being in the top
10% a "good" performance? Is being in the bottom 10% a "bad" performance?
If so, then the supervisor has the ability to know how many employees did
good work and how many did bad work in a particular period *before*
anybody does their work!
Keep in mind that in a work group made up of highly educated, bright
people like engineers, there's going to be a lot less difference between
the top and bottom member in absolute performance than there would be in a
less highly-selected group. The vast majority of incompetents have been
weeded out by the selection process. It's just like the old saw about how
one out of 10 doctors graduated in the bottom 10% of his med school class.
Since I know something about math, I would have not the slightest qualm
about such a doctor performing extremely critical surgery on me, and might
well have very good reasons to prefer him over someone who graduated in
the top 10%. Graduating in the bottom 10% of one's med school class is a
tremendous academic achievement; the problem is that most people's
experience of the "bottom 10%" is with students in non-selective
environments like high school.
In a typical work group, ranking ties pay to performance about as well as
using random numbers does.
--Eric Bohlman <ebohlman@netcom.com>
Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>