<p>Self-studying here.</p>
<p>Have gotten 3s on both practice tests -_-</p>
<p>Calculator memory will NOT be deleted before the test…I’m pretty sure they don’t do that for Stat.
Btw, if you’re aiming for a 5, just try to get 10 wrong out of 35…</p>
<p>Great little sheet that summarizes all the assumptions for each test we need to know:
<a href=“Supporting Students from Day One to Exam Day – AP Central | College Board”>Supporting Students from Day One to Exam Day – AP Central | College Board;
<p>wow!! thanks for that assumptions sheet! you saved me so much time 
!!!</p>
<p>alright.
and is anyone familiar with the formula that barrons gives you to find the sample size of z test?</p>
<p>.5root2/root n -> do you guys think this formula will be helpful</p>
<p>You should know that formula; a question about minimum sample size usually shows up on the multiple choice.</p>
<p>n≥((z*θ)/(margin of error))²</p>
<p>for proportions:</p>
<p>n≥((z*)/(margin of error))² x p(1-p)</p>
<p>when we don’t know p, use .5</p>
<p>And what if they don’t mention the significance level, we use .05 or .01?</p>
<p>They’ll say they want the margin of error to equal, say, .04 for the 95% confidence interval; It will always be given to you so you can find the z* value.</p>
<p>p = .5 usually (for proportions)</p>
<p>i always have trouble knowing what tests to use…any tips??</p>
<p>If they give you the mean or they give you data and ask you to statistically justify if there is a different between the means…for stuff like this you’d use a T-Test.
If they give you proportions instead of means, you’d use the Z-Tests.
If they compare samples to each other to determine association and stuff, its a chi-square test.
If they give you a scatterplot and some info and tell you to statistically interpret the correlation, its a T-Test for linear regression.</p>
<p>thanks so much!!</p>
<p>For anyone using a TI-83 or TI-83 plus this site has a program that’ll do the Chi-Square Goodness of Fit Tests for you: [CHISQGOF</a> (Chi-Square Goodness of Fit) Program for TI-83 or TI-83 PLUS](<a href=“http://staweb.sta.cathedral.org/departments/math/mhansen/public_html/01stat/programs/chisqgof.htm]CHISQGOF”>http://staweb.sta.cathedral.org/departments/math/mhansen/public_html/01stat/programs/chisqgof.htm)</p>
<p>I’ll throw out some advice just to review myself :)</p>
<p>Know also that with two-sample tests, you’re testing to see whether one sample is greater to, less than, or equal to another sample… Here you’re talking more about populations with the samples, because, for example, just because you get a sample with a bigger mean/proportion/center, it does NOT mean that its population has a greater mean/proportion/center, as it could be caused from variability (you could have just gotten a funky sample :S). This is different from pooled or nonpooled characteristics because those deal more with spread from the same population (pooled) rather than comparisons between the two. I hope that makes sense! :D</p>
<p>you never pool the stuff on the calc right? that’s what my teacher said…</p>
<p>oh also, are there a lot of I II III questions on the actual exam??</p>
<p>From the released multiple choice sections we took, I would say there is about 3 of those type for each, not too many.</p>
<p>okay.cuz barrons has a ton.and i’m kinda bad at finding all of them…i always miss one!!</p>
<p>when do we use a chi-goodness of fit test, and how is that different from the normal chi test?</p>
<p>Chi-square GoF - One Population/sample & data falls in line
Chi-square Independence - Bivariate data, one population
Chi-square Homogeneity - Univariate data, two populations</p>