AM I A GOOD STATISTICIAN?

Today, during the Statistics discussion, we had the project presentation. One of the groups was trying to find if there is any difference between the number of sex partners of females and males at the university of Michigan..

They reached the conclusion that on average the guys have 4 more sex partners than gals.. but their survey was conducted in a very biased way.. and their result was obviously biased.. (like they have someone who has 96 sex partners, usually this kind of outliers must be eliminated from the hypothesis testing)..

When Yang Yang asked us if we have any comments.. I said something which made the whole class laughed..

this is what I said:

If you define sex partners as the ones people has had sex with before,
=> if A had sex with B, then A is counted as a sex partner of B and B is counted as a sex partner of A
=> the total number of sex partners for females and for males should be equal..
=> if the mean number differs by 4 people, it shows that there are a lot more gals in UM than guys (which is definitely not true)..
=> to explain the huge difference.. we can get the conclusion that UM has many bi-sexual people..

the whole class fainted.. and Yang Yang concluded that I am that kind of statistician that governments will want- cos I can infer the data as I want..

posted by panda xuan @ 12:46 AM