<p>^^ thats what i did...i figured if they gave us 12 tanks they would want 12 treatments.</p>
<p>oh grr..if we werent clear enough on that CI one how is that graded assuming that whole part was wrong. 2/4? 1/4? thanks</p>
<p>what I did:
A high, B high, C high,
A low, B low, C low,</p>
<p>A control, B control, C control,
control nutrient high, control nutrient low, control nutrient control salinity</p>
<p>I know it seems repetitive, repetitious, redundnant, and it repeats itself, but hey there were 12 tanks so...</p>
<p>I guess we can talk about it now.</p>
<p>My stat teacher says that it's definitely the 6 treatments with replication. According to him (and he's pretty great at all stat-related things, not just someone randomly assigned to teach AP Statistics), replication is crucial for any 2-variable experiment. I did the 12 treatments, too, though. D:</p>
<p>For 6e (maybe it wasn't e... the one with the histograms), though... I was talking with someone, and he's pretty convinced that the region you mark is the last 5% of the data, but I think everything after 16.92 (I think that's what it was) makes more sense. In his way, 5% of the samples result in rejection for all of the distributions, which, at an alpha of 5%, is expected even if the null is true. I thought it would make more sense if, given the fact that each of them has a variance over 1.52, more than 5% of the samples would produce rejections. This led me to the answer for the next one... The distribution with the least data in the "region of rejection" is the one with the highest variance, and the one with the least data there has the lowest variance. Unfortunately, I got I and III with this method, when everyone else says it's II and III, so I must have made a mistake somewhere. >_> So yeah, what did everyone do for these two?</p>
<p>You just mark on the graphs where the rejection would be (>16.92) then compare its shape with the actual chi-square distribution (its graph is in the back)</p>
<p>if i remember correctly you basically just marked the ends of every histogram. </p>
<p>also, what is the significance of the standard error of the slope for the linear regression equation?</p>
<p>That's what I thought (and said)... Not sure what the guy I was talking to was thinking.</p>
<p>So, did you get II and III (not necessarily in that order)?</p>
<p>Shirobot,
I agree with you on the whole marking everything after the cutoff (not simply marking off only five percent) point. However, the reason everybody else thought it was II and III, including my self, is probably because we got 12 point something as the cutoff not 16.92.</p>
<p>12.something was definitely the chi-square value for the data they gave you, but I thought they wanted the critical value for rejection at 5%, 16.92? Besides... Wouldn't I still get the same thing? I'm not sure, though... I haven't seen the histograms since the test.</p>
<p>Oh, never mind you're right. I don't know what to say about the histograms then. I just found merely by looking at it, that II and III had the least and highest proportion of data that was after the cutoff mark, though not necissarily in that order (I dont remember if II was the low one and III was the high one or vice versa).</p>
<p>what was the 's' in part b of the regression question?</p>
<p>i think it was the standard deviation of the slope, not 100% sure tho</p>
<p>I think the 's' was the standard deviation of the data, I had no idea what it meant by "interpret it" though.</p>
<p>So if you put that there were 6 treatments, you got full points?!?! :D :D :D</p>
<p>Why don't they post the scoring guide?! When does that come out?</p>
<p>When do they usually put up the solutions/grading rubric for the free response? I noticed that the 2006 ap eng lang fr is online but not the solutions.</p>
<p>Can any statistics gods post how to do 3 b) and c) please? I want to know if I got them right but don't remember the answers - just the procedures i used.</p>
<p>I had no clue what S and Se was. I typed in the data and two lists and did 2 variable stats and got like 1 and almost 11 for sample deviations. No where near 2, so I just told a story.</p>
<p>Do you think they will take off for saying use a base 12 number generator to assign the tanks. I didnt feel like writing out all the ranges and crap.</p>
<p>2 b) This is the standard error about the model.
c) SEb = .7553. This is the standard error(or maybe deviation?) of the slope, or the ammount that the y increases for every one x increases. </p>
<p>Not sure, but thats what i put.</p>
<p>slope is the ammount that the y increases for every one x increases, but they gave the standard dev. of that, and also wanted it in context...meaning it had to have the word soapsuds and such.</p>
<p>Would someone please share their general answers for questions 1-5 so I can feel better?</p>