Self Assessment Answer #
4
by Dr Jamie Love 2002 - 2005
|
I hope you understood that, if there was no link with sex, the colorblindness should be equally distributed between the males and females. You may have concluded that the expected numbers are ...
|
But this is WRONG!
Look at the number of colorblind people in this population. It has a total of 16 colorblind people and 16 normal people. True, they are equally distributed between the two sexes but that is not what we started with. We started with 10 colorblind people and 22 normal people. How did we suddenly get 16 colorblind people and 16 normal? Remember, when determining the expected ratio we redistribute the phenotypes to create the ratio we expect but we do NOT simply make up new data because that defeats the whole purpose of using real data to arrive at real conclusions. We would expect those 10 (not 16) colorblind people to be equally distributed between the sexes so we should expect 5 of them to be male and 5 of them to be female. (Think about that and convince yourself that what I am saying is correct.) |
Let's walk through this systematically.
We have a total of 32 people - 16 males and 16 females. By the way, we have a "perfect" sex ratio. The chi-square of the sex ratio (just males to females) is zero so the sex ratio itself is not skewed - but that is not our concern here.
We have 10 colorblind people. If there is no linkage with sex, we would expect 5 females and 5 males to be colorblind. That leaves us 11 normal females and 11 normal males. These should add up to 32 people, 10 colorblind and 22 normal (and, by the way, 16 males and 16 females).
This is the distribution we expect - an equal distribution of normal to colorblind people in both the male and female populations.
|
|
Return to the Question.