AP stats discussion

<p>Since the error was based on a previous question (whether you rejected the null or not), if it matches with your conclusion and you justified it, you should get full credit.</p>

<p>So for the 2c…is it (130-125)/(2.9) as a normalcdf or as a t-test with df=4? Because that makes a huge difference. My rationale for t-test was because it was means and the original population SD was NOT given…only the sample SD.</p>

<p>And why did you divide by .5? Isnt it by 5?</p>

<p>And if the numbers were 130 = x, 125 = mean, and 6.5 = sd, with n=5, does anyone want to recheck to get like 0.08? Thanks</p>

<p>^ idk but using normcdf i got .0427
(130, 1000000000, 125, 6.5/sqrt(5))</p>

<p>I did 650-625 / 14.53 (multiplied it all out and found a new standard deviation)</p>

<p>I then got z = 1.72
P(z>1.72)= .0427</p>

<p>^ that was a typo, it is divided by sqrt(5), not .5. My bad.</p>

<p>Why on earth would you do a t-test?
I’m mostly sure this is a normal curve probability question.</p>

<p>I thought it only gave the sample SD? It didnt give the population SD? Ahh if I get this wrong, Ill be really mad cause I did do normalcdf first but changed it to t-test.</p>

<p>You are assuming that the 6.5 is a population st.dev because it didn’t say that it was from a sample (lol @ the reasoning but I think its right).</p>

<p>dang…i think i know where I messed up…the five cars were chosen from that sample…ahh</p>

<p>ok…and does anyone know that for the confidence interval question, if you forgot 1 condition (the nearly normal one), is it an automatic 3?</p>

<p>btw although i messed up 2c, i am almost positive 2a is 128ish because i just did invorm(0.7)=something that got me 128.</p>

<p>Actually im pretty sure u had to use a t-test. </p>

<p>U can never assume that they give you population S.D. If they don’t strictly say it, then they didn’t give it to you.</p>

<p>"
The probability that it is greater than the 70th percentile is .3. And it would be P(at least 2), so you wouldn’t need to find the complement if you did binomial cdf from 2 to 5.]</p>

<p>(5 C x)(.3)^x(.7)^(5-x) for x=2,3,4,5"</p>

<p>for this one did we have to state the conditions for binomial???
and if I didn’t is it failure? :(</p>

<p>I didnt state the conditions…I dont think it was a biomial test…it was just a ordinary probability problem. You didnt have to use binomcdf (although you could to make it quicker)</p>

<p>Anyone want to verify</p>

<ol>
<li><p>2a is 128ish? its just what I did with inovmr(0.7)=(x-125)/(6.5) right?</p></li>
<li><p>What was the answer for 4c where it said “can this come from a normal population or is the normal population skewed”…I didnt state CLT…but i said that as n increases variability decreases so it can still come from a normal population…as the first sample had a really low n, increasing its chances for variability. Is that right?</p></li>
</ol>

<p>I said that the 100 samples showed a normal distribution as they were centered around a certain range of numbers and that the occurances of a sample having a certain stat decreases as the stat goes farther away from the center</p>

<p>pretty sure its CLT.</p>

<p>HiPeople, aren’t you talking about question 6?</p>

<p>and for that interval question asking whether the mayors predictions were wrong or whatever, did you people say yes because zero was in the interval?</p>

<p>also one of the questions mentioned some type of error. Was it a type I error?</p>

<p>thanks peepz</p>

<p>Type II error. The one about patients or something. I stated that a Type II error could cause unnecessary deaths : /</p>

<p>Yes sorry I meant 6c. What did you get for that?</p>

<p>For the mayor one, I said it does not support the guys claim because 0 (the null) is in the interval.</p>

<p>Did the free response have to be in pen, or was it ok to use pencil?</p>

<p>Yeah, i said that the sample could have come from a normal distribution ;-)</p>

<p>aww i guess i missed the one about the error :-(</p>

<p>DoleWhip, the directions said a number 2 pencil or pen.</p>

<p>its funny that you asked that…cuz i did have of mine in pencil, half in pen LoL</p>

<p>Just checking, someone just told me it had to be in pen so I started freaking out</p>

<p>thanks :)</p>

<p>The type I or II depends on whether you rejected the null or not frm the previous question so you could still get full credit.</p>

<p>What was the reason for the normal population? If its CLT, Ill be kinda mad because I totally forgot to state it…I just talked about the SD.</p>