variance and performance based evaluations

[chicheebee]

so a friend of mine works in sales, and is coming to the end of his probation period. whether he gets kept on is based solely on his "conversion rate", which is the proportion of calls he takes which result in actual sales, and is calculated at the end of each month - he needs 17.5% or higher to pass, and is currently on around 16%. now, from my poker background, i'm inclined to regard the number of calls he takes in a month as statistically insignificant (almost trivial). so i'm wondering a couple of things:

1. is it reasonable for them to select employees on this basis, or are they virtually picking people out of a hat?

2. if my friend doesn't get kept on because he just misses his target, is it worth kicking off on that basis, telling them they can't fire someone based on these results, cause they aren't worth [censored]. i have a degree in math, and my father is a professor of statistics, so if it came to that i'm confident we could "out math" them. but ultimately, we can't actually make them do anything, so it would just be a case of trying to embarrass them into changing their minds.worth a shot, or totally pointless?

cheers,
cheech

[pzhon]

[ QUOTE ]
is it reasonable for them to select employees on this basis, or are they virtually picking people out of a hat?

[/ QUOTE ]
It's not perfect, but it is a reasonable test. It might be that they should fire people who miss the target by a lot earlier, and let people who barely miss try longer, but that might actually be the policy. It could be that the real acceptable threshold is 25% conversions, and those who are between 17.5% and 25% are watched more closely later. Even if that is not the case, it's still better than random.

If you want to determine how much better than random it is, you need to describe both the number of calls and the amount of difference skill can make. Perfect informaition is not necessary, but are we talking about 40 calls during the probation period or 400? Is the difference between a good salesman and a bad one 25% vs. 15%, or is it 40% vs. 10%?