|
So who else is ready for those AP tests coming up?
I got a stats HW review and here is the question:
A tennis racquet manufacturer has designed a new racquet. The manufacturer claims the new racquet will allow the user to return more serves than any currently sold. A group of male volunteers who play tennis and own their own racquets agree to participate in the study.
a. Design an experiment that would test the manufacturer's claim.
b. Suppose the volunteer group had included both men and women. Would you adjust the design of your experiment? If so, provide the new design.
And for stem plots, can you put decimals or do you round them?
Thanks
|
Oooh AP tests.. I can't help you on yours, but I could definitely use some help in preparing for my AP U.S. History test. :p
|
This one is really rough, partially because I play tennis and the other because I over think it.
IN this case a ball machine to simulate a serve would be best, make sure your sample size is large enough to accommodate the test say 100 returns or so per person. You should make sure that you pick volunteers of all skill levels so that you can get a comprehensive experiment. If you want to be really comprehensive pick different court types (clay, hard, grass). Make sure you do both duece and add court and then you should be good to go.
The issue is that the ball machine doesn't really facilitate the randomness that serves present. A ball machine can't really accurately hit a kick serve or a slice serve. Conversely a human being won't be consistent at all unless you somehow get a pro female player to do the experiment for you and even then... yea.
Women experiment need not be changed much except for serve speed, make it slightly slower.
|
On April 28 2009 12:43 Jayme wrote:
This one is really rough, partially because I play tennis and the other because I over think it.
IN this case a ball machine to simulate a serve would be best, make sure your sample size is large enough to accommodate the test say 100 returns or so per person. You should make sure that you pick volunteers of all skill levels so that you can get a comprehensive experiment. If you want to be really comprehensive pick different court types (clay, hard, grass). Make sure you do both duece and add court and then you should be good to go.
The issue is that the ball machine doesn't really facilitate the randomness that serves present. A ball machine can't really accurately hit a kick serve or a slice serve. Conversely a human being won't be consistent at all unless you somehow get a pro female player to do the experiment for you and even then... yea.
Women experiment need not be changed much except for serve speed, make it slightly slower.
Thanks, but I think after reading this, I somehow figured out it's a blind or double blind experiment! Not sure how I remembered it but thanks
|
On April 28 2009 12:43 Jayme wrote:
This one is really rough, partially because I play tennis and the other because I over think it.
IN this case a ball machine to simulate a serve would be best, make sure your sample size is large enough to accommodate the test say 100 returns or so per person. You should make sure that you pick volunteers of all skill levels so that you can get a comprehensive experiment. If you want to be really comprehensive pick different court types (clay, hard, grass). Make sure you do both duece and add court and then you should be good to go.
The issue is that the ball machine doesn't really facilitate the randomness that serves present. A ball machine can't really accurately hit a kick serve or a slice serve. Conversely a human being won't be consistent at all unless you somehow get a pro female player to do the experiment for you and even then... yea.
Women experiment need not be changed much except for serve speed, make it slightly slower.
You totally missed the point of this question.
Nobody cares about machines throwing kick or slice serves.
What is important is to make this a (double) blind experiment.
So, for example, collect all rackets from the testers (say 100 testers). spraypaint them black and do the same with the company's rackets which should be tested so they are indishtinguisable to the testers. Now, provide 50 rackets from the company and 50 pre-owned rackets to the group and let them play, say 100 returns vs machine. Of course somebody needs to keep track of which players use the company's rackets, but the players themselves should not know if they use a pre-owned or company racket during the test.
|
LaptopLegacy, I think that's a good answer but what about fatigue? Should I randomly make half of the volunteers use their own racquets first and the other half use the manufacturer's and vice versa?
|
If the mix of volunteers is truly random, then they will on average become tired at the same rate, isn't it? so i think it shouldn't be a factor in the test.
what you also can do is run a test where every volunteer has to play two times, once with the company racket and once with a pre-owned racket. but he can not know in which test he is using the company racket.
if you set it up this way, then let 50 volunteers play with the company racket first and 50 with the pre-owned racket first.
|
Making it a double blind experiment is definetly key here. From my reading of the question the 'proper' way to do the test would be with real people serving it since that is how the racket is going to be judged; this of course complicates things. You can easily go into a lot of detail with how to set this up. I think the key would be double blind, large enough sample size, and explain how you would try to eliminate the affect of any other factors (such as fatigue, skill level, etc...) I wouldn't go into too much detail, maybe if you have time at the end of the test. Hope that helps 
|
It's been a while since I took an AP test, and I never took AP statistics, so while I won't comment on the question directly, what I will say is this: these tests are graded incredibly quickly, so you really want to just hit the high points, drive them home, and move on. So here you want to right away state that you would do a double blind experiment and then just explain how you would pull it off. I might even underline the key parts or terminology. That way the teacher grading the exam, who is working on his/her summer vacation for very little pay looking at AP exams for 12 hours/day, will skim through, see the keywords and important ideas and immediately think "okay this guy knows what he's talking about."
Remember that it's the kids who didn't study who are making stuff up about ball machines and all sorts of other information that isn't really pertinent. The question is about statistics, not tennis.
|
|
|
|
|
|
EPT
