<p>Is there anyone who would like to reply to my question>?</p>
<p>Man, the free response was cake. Luckily the day before I crammed the formulas for the t test for B compared to a Beta. Thank god I did because we didn't learn it properly during the real class (Thank you princton review). I think I owned question six and was pleasantly suprised when I saw how easy it was. BTW I put Ho: B=1
Ha B>1
Because if the distance is perfectly judged the slope is one but if they overestimate B is greator than one. THis was really the only "hard" part of six, the rest was wimpy-just graphing and some interpretations. Oh and the formula I used was (b-B)/(SEb) where SEb is the STD error of the slope. </p>
<p>Oh yeah for part A im pretty sure the slopes mean this: For each increase in the distance of the objects by 1ft the subjects estimates increase by (whatever the co-ef of x was)</p>
<p>-Sean</p>
<p>Haha, for 6b, I said they prefered model 2 because there was a y-intercept value in model 1, so it would estimate that when the actual distance is 0, people might estimate .2 feet or something, and that wouldn't make any sense. I hope I get at least some partial credit for that though.</p>
<p>THats what I put for part b too, couldn't think of anything else lol</p>
<p>BTW the std is bigger for model two so it is likely to be less acurate...Thats why I didn't use that as a reason</p>
<p>-Sean</p>
<p>So what did you put as a reason to prefer Model 2 over Model 1?</p>
<p>I just put that if the objects are in the same spot people arent gonna be idiots and say that they are .2 ft away or whatever.</p>
<p>-Sean</p>
<p>Shravas: That's what I put, too. I think it was a good answer. It sounds like the most logical one.</p>
<p>On number 4 I totally blanked and did a box and whisker plot. Ugh, I'm so ashamed of myself. And number 6, wow, our teacher never even covered that. Luckily a 40% is equal to a 3. Yay for huge curves.</p>
<p>for #4 I almost made a box plot too. if you made a histogram of the differences of the results of the two methods, you got a perfectly simetrical, approximately normal display.</p>
<p>Oh...I did a box and whisker plot but then again I did 2 sample T...</p>
<p>me too... our teacher always did boxplots.. she's horrible at teaching :(</p>
<p>crap, i blanked on #4 and did the assumptions for a 2-sample t.
didn't even do a graph.</p>
<p>at least i did the test for matched pairs.</p>
<p>if you did a boxplot for the two data sets, they actually both kind of looked like chi-square distributions</p>