DaHuuuuuudge said:
Kasz216 said:
DaHuuuuuudge said:
Kasz216 said:
Or honestly... hell
http://www.socialresearchmethods.net/kb/statcorr.php
Null Hypothesis: r = 0
Alternative Hypothesis: r <> 0
The easiest way to test this hypothesis is to find a statistics book that has a table of critical values of r. Most introductory statistics texts would have a table like this. As in all hypothesis testing, you need to first determine the significance level. Here, I'll use the common significance level of alpha = .05. This means that I am conducting a test where the odds that the correlation is a chance occurrence is no more than 5 out of 100. Before I look up the critical value in a table I also have to compute the degrees of freedom or df. The df is simply equal to N-2 or, in this example, is 20-2 = 18. Finally, I have to decide whether I am doing a one-tailed or two-tailed test. In this example, since I have no strong prior theory to suggest whether the relationship between height and self esteem would be positive or negative, I'll opt for the two-tailed test. With these three pieces of information -- the significance level (alpha = .05)), degrees of freedom (df = 18), and type of test (two-tailed) -- I can now test the significance of the correlation I found. When I look up this value in the handy little table at the back of my statistics book I find that the critical value is .4438. This means that if my correlation is greater than .4438 or less than -.4438 (remember, this is a two-tailed test) I can conclude that the odds are less than 5 out of 100 that this is a chance occurrence. Since my correlation 0f .73 is actually quite a bit higher, I conclude that it is not a chance finding and that the correlation is "statistically significant" (given the parameters of the test). I can reject the null hypothesis and accept the alternative.
|
-_-
This is a statistical inference test for a graph. The null hypothesis states that there is no correlation, whereas the alternative hypothesis states that there is. You seem to think that alpha levels are only used in conjunction with r-values, however the point i'm trying to make is that alpha values for determining statistical significance (the original segue for the whole 'disagreement') do not need to have an accompying r-val. An example? Let's say that i have to test data from a sample survey that talks about proportions of married men. You would use an alpha-level in your statistical inference to test the probability of obtaining a sample such as this.
I know I'm right, i'm not sure why i keep responding. -_-
|
I never said Alpha levels were only used in conjuction with R levels. I suggested however that they were used with conjunction and R value. While you suggested alpha levels have nothing to do with correlation or R Values. As an example your talking about not seeing anything about correlation with anything in statistical significance. Essentially I was argueing that squares are rectangles, while you were suggested they aren't.
Considering we are talking about whether the correlation with gun ownership and homicide is significant... I don't see what your perception of the arguement has to do with anything. Since i would still be correct in terms of the original dataset
This can be seen by the fact that you just needed to invent a whole new data set unrelated to the data at hand to give an example of Cronbach's Alpha. Which I do know about, do aknowledge exist, and have even used before. I just don't understand what it has to do with the data at hand.
To me, it seems like you originally thought tha Cronbach's Alpha was the be all end all of statistical significance. Or jut thought it was the only use of "Alpha" Ignoring Alpha in regards to Pearsons R... or just not knowing of Pearson R's existence. (Which is what you'd use in the original dataset in question.) Which i suppose I could see if you spent a lot of time working for say the Census or the BLS since you might not have a lot of use for Pearson's R.
Though perhaps the whole thing is a misconception on both of our parts about what the arguement was about.
Though why if this was your arguement you didn't say "What about Cronbach's Alpha?" or even just link to it... I'm not quite sure.
|
Look, you said statistical significance means that something is proveable, which is false. I corrected you. I seem to be sucked into the Kasz method haha
http://gamrconnect.vgchartz.com/post.php?id=4878899
|
That's exactly what the Pearson's R does do though... it shows is something is significant enough to be used as proof.
If it isn't statistically signficant it can't be used as proof. I mean... that's just the basics. Any correlation that is not statistically significant can not be used as proof of a position.
Though yes, i do have a problem with restating facts and showing proof over and over again even when it's clear the other person isn't interested in such things. Largely because such things still prove informative for third parties.
