2011 AP Statistics Thread

<p>Player A was the correct answer because you needed to calculate the z scores for their performance both for strength and speed. Then take the average z score. find the p-value. You find out that player a performs, on average, at the 97th percentile, while player b performs somewhere around the 75th.</p>

<p>I got that they weren’t equal, which should make sense since party preference should not depend on gender.</p>

<p>Um, actually if they weren’t equal it DOES depend on gender. Which is true in the real world.</p>

<p>radracquet95 - yes, for all but a z score to compare. I’ll discuss it more with you when allowed, but the answer is correct.</p>

<p>okay im pretty positive we can start discussing now, no??</p>

<p>In my stat class, we have to do an end of year project. we gotta use sampling and carry out a significant test. any ideas on a topic? (i want it to be easy to do, but the teacher also wants it to be something interesting)</p>

<p>do you actually have to do the sampling or can you use existing data?</p>

<p>no we have to do the sampling</p>

<p>My teacher is doing the same thing. I’m probably going to do something like the number of hours high school students get to sleep.</p>

<p>yeah my teacher is making us do an inference test but i dont think we have to sample. hmmm ill try to think of interesting ideas for you though! can we discuss frq answers noww? the questions have been uploaded to collegeboard!</p>

<p>Where are they uploaded?</p>

<p>They’re [url=&lt;a href=“AP Statistics Exam – AP Students | College Board”&gt;AP Statistics Exam – AP Students | College Board]Here[/url</a>] on the College Board website</p>

<p>For the question regarding the Parties and the Genders of people, specifically the part regarding whether or not the two events are independent, would it have been ok to say something along the lines of:</p>

<p>Since the two events ‘male’ and ‘party y’ rows/columns intersect on the given chart (by technical definition, independent events are those where P(A and B) = 0) such that P(A and B) = 48/500, the two events must not be independent, but rather dependent.</p>

<p>I dunno, it was something more common-sensical but I’m just hoping they won’t mark off just because I didn’t apply the formula.</p>

<p>So would that part get points?</p>

<p>Also, here’s how I think I did overall on the exam (so if anyone experienced can tell me what they think I got, please do! and by experienced I just mean use your own judgment and what you think the curve will be, because some say the curve this year will be harsh!):</p>

<ul>
<li>At least 30/40 MC questions answered right (I hope)</li>
<li>FRQ 1.
Part A: Said that it was normal because the minimum value fell within 2 SD of the mean (I guessed on this though).
Part B: I got this right for sure
Part C: I <em>think</em> I got this part right. I answered Player A (which I believe was right) but the method in which I answered the question was kind of odd. I just added the percentiles (since the two are ‘equal’) and said that one way to find out which, on average, was better was the higher one (Player A).</li>
<li>FRQ 2. Part A: got it right for sure. Part B: risky wording (already mentioned above). Part C: I definitely got this part wrong. I just copied the graph they gave for Franklin Township. Considering I said the events were dependent in part B, I kind of contradicted myself… which is worrying.</li>
<li>FRQ 3. Got it right for sure, except I mentioned that the children (part b) were a ‘factor’ that could change the result of the experiment (something more intelligent sounding though…) as opposed to a ‘lurking variable’ or some other statistical term.</li>
<li>FRQ 4. I feel like I got this right. I did mention that we are assuming that no values are outliers and that the spread is approximately normal though because the sample sizes were less than 40. (???)</li>
<li>FRQ 5. I also feel like I got this right. Was Part B: 2.4 Amps?</li>
<li>Investigative Task. I totally got this wrong… I managed to find the confidence interval for part A but only because I used a calculator. I tried to do it by hand but I kept getting an ‘off’ answer so I just crossed out all my work. Part B I got (3/8)k - (1/8) or something with a fraction of an eighth, and I didn’t get to answer part C because of timing :(!</li>
</ul>

<p>Sorry it was so long, but opinions anyone!!??</p>

<p>Does anyone know how to do the one in 2c with the graph. I don’t think it was the same but it was slightly different i got. For 1 part A i said the same as above totally guessed too. Part b was just z score and part c i found the z scores but I put player B which seems to be wrong. #2 to see if it was independent u had to use the formula P(A/B)=P(A), which it didnt equal so not independent. #3 i didn’t know how to design with the clusters, but I said the children were other variables and using stratified reduced variability. #4 was pretty straight forward and it was 2.4 amps difference and u concluded that wind and electrical output was associated cause p value was 0. #5 was just a confidence interval first then u had to fill in the tree chart and u should hv got .25-.25k and .75-.75k. It’s k+(.25-.25k) for the probability i believe then i got stuck. Anyone remember what they got for that part?</p>

<p>For #1 Part A, it isn’t normal because the minimum value is less than 2 SD away from the mean. If it were normal, we would expect the minimum value to be at least 3 SD’s away…</p>

<p>if it wasnt approximately normal, then we wouldnt be able to do the calculations in part c</p>

<p>I thought the test wasn’t too bad. =) Does anyone know what percentage we need to get right in order to get a 5?</p>

<p>So I think we’re allowed to discuss the Free Response now…</p>

<h1>1: Part A) I couldn’t get because I had no idea how to determine if the distribution was normal with just the mean and minimum value. My answer was a lot of BS for that one</h1>

<pre><code> Part B) I’m pretty sure I got this right. For the interpretation I just said it was __ standard deviations away from the mean
Part C) I said player A. I mentioned something about the standard deviations of the speed and weight distributions I think.
</code></pre>

<h1>2 Part A) I definitely got</h1>

<pre><code>Part B) I said they were dependent because the probabilities weren’t equal? I feel like that was enough but I also went on to do a confidence interval and stated that since it didn’t contain zero it was dependent. I feel like I shouldn’t have. But some people are saying that you were supposed to do chi square?
</code></pre>

<p>Part C) Pretty sure I got this. Males and females had the same proportions</p>

<h1>3 Part A and B) Easy! I love designing!</h1>

<h1>4 Was that the difference of 2 means test? or was that #5? Because I did that somewhere on the exam</h1>

<h1>6: I <em>think</em> I got part A. For the tree diagram I just did 1-k for the probability of guessing. .25 was the probability of guessing right and .75 for guessing wrong. Then I multiplied and came up with a final answer like .25 +.75K for the people who got it right. And then for part D) solving for k I got .04 (I set it equal to .28 from the first part) and then I set up the confidence interval. Is that even remotely right??</h1>

<p>Either way, I think I did pretty okay. I know I got at least a 4.</p>

<h1>4 was the diff. of two means (2-sample t-test)</h1>

<h1>5 was the linear regression line</h1>

<p>@loldanielol: “For #1 Part A, it isn’t normal because the minimum value is less than 2 SD away from the mean. If it were normal, we would expect the minimum value to be at least 3 SD’s away…”</p>

<p>YES!!! That is what I wrote in at the last second. Haha, that part actually stumped me for some reason when I first looked at it. o.O</p>

<p>^same here thats what i wrote. Then for part C, i put “assume that 40 yd dash times distribution is normal” and did the z-score thing</p>

<p>The segmented bar graphs were the same for males and females. Because, you discovered earlier in part b that gender and party are independent, therefore gender has no affect on registered party. So, the bar graphs for both were simply the proportion: party/total.</p>