<p>i did the same thing as you hipster!</p>
<p>I did Median/Q1= Median/Q3. If they were equal it was a symmetric distribution, but it would be right skewed if the left term is smaller then the right term.</p>
<p>Can someone PM me question 9 and 34? I honestly don’t remember them and now I’m curious to what it might have been.</p>
<p>What type of error (type I or type II) did you guys say could have happened for the question where the p-value was given, number five I think, about the resuscitation with chest compressions and mouth to mouth versus mouth to mouth alone?</p>
<p>My made-up statistic was (Min + Q1 + Q3 + Max)/4, so I basically took the average of these points. If it was greater than the median, it was skewed right because it meant that the points extended further right than left.</p>
<p>Is this okay? I don’t know if I was creative enough, but my solution did answer the question (examining skewness based on the 5-number summary)</p>
<p>NYEM, I think it hypothetically would have been a Type II error - failing to reject the null hypothesis when there actually was a significant difference between the two treatments.</p>
<p>i said type II i think…because you initially failed to reject the null hypothesis from the first part…type II is when you fail to reject a false null.</p>
<p>I think all these work as long as they measure skewness most of the time. Even the Mean/Median rule doesn’t work all the time.</p>
<p>i obviously had a different version because i had no free reeponse regarding type I/type II errors</p>
<p>
I did the same thing as you.</p>
<p>My friends did outliers. Q1 - 1.5IQR and Q3 + 1.5IQR.</p>
<p>This is what a ton of people did:
(Max - Q3) > (Q1 - Min)
If the above statement is true, the graph is skewed right.
Is this even correct? I don’t know.
Basically, the right tail has to have a greater range than the left tail for it to be skew right.</p>
<p>I did something weird:
(Q3 + 1.5IQR) > Max,
then the graph is skewed right because of the presence of outliers towards the right.
Don’t know which one’s right…</p>
<p>As for the what type of error, I put Type I.
Edit: I’m pretty sure its not Type II because it asked for what type of error COULD have possibly been committed. Its not II because you’re already failing to reject the hypothesis from your test (part b of that question, I believe).</p>
<p>I said type II error, being that we rejected the the alternative hypothesis because the p-value was just over .05.</p>
<p>^ Failing to support the alternative does not mean that its a Type II error. A Type II is the probability of failing to reject it when it is actually false.</p>
<p>So a Type I could have been committed, right?
i’m confusing myself here…</p>
<p>I put Type II -____-" but I forgot to change my answer for this part…after I realized it said 0.07 not, 0.007. So I basically contradicted myself in part a and part b…eff.</p>
<p>But yes, Type I error (reject null, accept alt when null is true) since we fail to reject the null.</p>
<p>Type II is failing to reject H0 when Ha is true.
Type I cannot happen since H0 was not rejected.</p>
<p>It has to be Type I because according to the test, null is true, so the mistake that can be made is rejecting the null and accepting the alt when null is true.</p>
<p>The p-value was equal to .076 and alpha was set at .05. We rejected our alternative hypothesis according to our alpha level. It was possible that a Type II error occured because we rejected the alternative hypothesis, although it still could have been true.</p>
<p>You cannot reject an alternative.</p>
<p>A type II error is an occurence. This occurence is explained conveniently by a probability represented by beta. Type I could not of been committed because the initial part of that problem had data supporting a failure to reject the null (P>.05)…type I is when you reject a true null.</p>
<p>Rejecting the alternative hypothesis is the same as accepting the null hypothesis.</p>
<p>^
I’m pretty sure it’s not the same…
Correct me if I’m wrong, but you can never prove a null hypothesis to be true. You can only reject or fail to reject.</p>
<p>I hate stats. I’m just getting more and more confused…</p>
<p>Technically, since alternative and null are pretty much opposites, they would be the same thing. But back to the question, I don’t remember what it was asking or what the conditions were but I think I put Type I error.</p>