|
Tee Time is a tutorial designed to demonstrate how sampling size can effect the power of a t-test. The president of High School Golfers Association (fictional) has funded a study out of concern that SAT scores for high school golf teams were extremely low. The Association was not sure how men and women golfers differ on SAT scores on average. Your job is to figure out if the men and women differ significantly on SAT scores. To do so, select the means for both populations, then set the sample size you wish to draw from each population. Samples will be drawn 100 times from each population, with a t-test run on each pair of means. This will give you an idea of how likely you will reject the null hypothesis given the population means and the sample size you have selected. If you set different population values for the two groups, what happens to your chances of rejecting the null hypothesis as you increase sample size? If you set the two population means equal to each other, out of 100 tests, how many t-tests do you expect to turn out significant? |