Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let's say I have 10,000 dollars I want divided among 10 people. With simple division each person gets $1,000. Easy enough.

Now suppose each person has a score on a test from 0 to 200. Now I want to divide the money among all the people but weighted by their score on the test, such that people who scored higher will get more money. How would I do that?

share|improve this question
1  
$W=\frac{Individual~ test~ score}{sum~ of~ all~ test~ scores}$ –  Simon Hayward Dec 7 '12 at 23:07

1 Answer 1

up vote 4 down vote accepted

Suppose that the test scores are $x_1,x_2,\dots,x_{10}$. Let $t$ be the sum of the test scores; then the first person’s share of the total is $\frac{x_1}t$, the second’s is $\frac{x_2}t$, and so on. These ten fractions add up to $1$, so just give person $k$ (for $k=1,2,\dots,10$)

$$10000\cdot\frac{x_k}t\text{ dollars}\;.$$

share|improve this answer
2  
Damn you have to get in quick to get ahead of Brian M. Scott. –  Simon Hayward Dec 7 '12 at 23:09
    
Is there a way to tweak the weighting such that even with a large disparity in scores there will not be as large a disparity in the money amount? –  User Dec 11 '12 at 7:40
1  
@User: I used a linear weighting of the test scores, but other weightings are possible. You could, for example, replace $x_k$ by $y_k=\log x_k$ and apply the same algorithm to the numbers $y_k$; this would greatly flatten the top end, since multiplying a score by $10$ (if you’re using logs base $10$) would only add $1$ to the modified score. Or you could replace $x_k$ by $\sqrt{x_k}$, which would have a similar but less dramatic effect. You might try replacing $x_k$ by $y_k=x_k^\alpha$ for different values of $\alpha$ between $0$ and $1$ and choosing an $\alpha$ that gives ... –  Brian M. Scott Dec 11 '12 at 7:59
    
... satisfactory results. –  Brian M. Scott Dec 11 '12 at 8:02

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.