<p>since 48 hours have passed, and the questions are online we can safely discuss it.</p>
<p>Questions: <a href=“http://apcentral.collegeboard.com/apc/members/repository/ap07_stat_frq.pdf[/url]”>http://apcentral.collegeboard.com/apc/members/repository/ap07_stat_frq.pdf</a></p>
<li>2 sample-t-test right? </li>
</ol>
<p>6c the t-value is b/SE_b right?
and p is tcdf(-E99,t,38)</p>
<p>Now… what else?</p>
<ol>
<li>1 sample t mean of differences (matched pairs) because each piece of meat will have different levels of bacteria</li>
</ol>
<p>what did you say for 1a?</p>
<p>3) You can still use z because the population is big enough?</p>
<p>5) Its an experiement?</p>
<p>6) I did Ho B=0, so thats wrong.</p>
<p>4 is not matched pairs. there was a huge discussion on it.
It's nothing on before and after or anything that could satisfy matched pair.</p>
<p>a is jsut the normal itnerpertation that of every increase of one unit in x...blah</p>
<ol>
<li><p>You can still use z because the sample is big enough - Central Limit theorem states that all samples of large enough size are approximately normal, regardless of the shape of the population's distribution.</p></li>
<li><p>It is an experiment, the treatments are imposed on the subjects.</p></li>
<li><p>I did Ho B=0 as well. Why is this wrong</p></li>
</ol>
<p>8parks11? Does this mean that it is a two-sample t?</p>
<p>and for 6, i think it is that.. its Beta=0 for Ho..how is that wrong? Mine was just to find the t-value.</p>
<ol>
<li>Did people draw a stem plot or a histogram? (back-to-back or two histograms) Thats what my AP Stats teacher told us to do to show the shape of it (If it's normally distributed or not)</li>
</ol>
<ol>
<li>IT is not a 2-sample-t-test. It is a Linear Regression T Test. Notice that the linear regression test involves the t-value. So to find the t-value without any data you do b/SE_b.</li>
</ol>
<p>1a and 3a...what did you guys put?</p>
<p>1a. The more widely spread the values are, the larger the standard deviation is...blah blah.. I just mentioned the equation and the difference between mean and blah blah.</p>
<p>3a: 2nd one since as n increases, it will increase the actual mean (in this case 8) according to the law of large numbers.</p>
<p>IMO, 4 has to be matched pairs!! You use matched pair (for mean difference) when the study is conducted and measured on the SAME SAMPLE GROUP. You cannot use 2-sample because the 2 sets of data came from A SINGLE SAMPLE! If you insist, you can see the two sets of data as "before" and "after," and calculate the mean difference.</p>
<p>6.read the question again. the way the equation is written it is not possible for the slope to be less than 0. y^ is equal to the numerical value of the percieved distance, distance would never be negative. </p>
<p>4 is matched pairs. it does not necesarily have to be a before and after to be matched pairs. in this case you match to eliminate the variable bacteria levels between each piece of beef</p>
<p>They are two different methods. They are also two different specimen. (10 each)</p>
<p>Great. I actually misread the question (I thought there's only 1 sample of 10 specimen)....oh joy.</p>
<p>there were only 10 specimens of beef, each method A and B was used on each specimen.</p>
<p>IT says two different methods...10 each.</p>
<p>In that case, I stand on my original post :)</p>
<p>AHAHA. My Ho and Ha for 6c were:
Ho: x/y=1
Ha: x/y=1.123 (or whatever the coefficient was)</p>
<p>How wrong I was, how wrong I was.</p>
<p>I also misread 6 (d) and only plotted the regressions line for the contact wearers.</p>
<p>it says 2 methods on "each of 10 different specimens." you're reading too much into it. the each means each of the 10 specimens, not 10 specimens for each method. </p>
<p>if you still dont believe me how do you explain the fact that for each specimen listed the bacteria levels measured by both methods are similar?</p>
<p>for 5b did people do 2 proportion z-test?
c kind of confused me because we dont know the standard deviation or the mean. I wrote the normal conditiosn that I learned but didnt know if i had to mention mean and population standard deviation?</p>