No, you cannot. Say there are 40 men and 20 women in the company. If the males average $33,470, then the approximate total amount spend on male salaries is (33470 x 40) = $1,338,800. Likewise, if the females average $29,525, then the approximate total amount spent on female salaries is (29525 x 20) = $590,500.

Added together, the total amount spent on all salaries is $1,929,300. Divided by 60 employees, the average would be $32,155 per person.

But, if you take the average (mean) of the female mean and the male mean, you get $31,497.50. That doesn't match up.

What happens in this case is a weighted average. In this case, we need to weight the average salary of the males twice as much as the females, since there are twice as many males as females. If we do that, we get [(2 * 33470) + 29525] / 3 = $32,155. That's better. Note that the denominator is 3, because we count the men twice. This is the same as (40 * 33470) + (20 * 29535) / 60.

The only time the mean of the female and the mean of the male salaries would be the same is if the number of men who worked at the company was the same as the number of women who worked at the company.

View the original article here