Statistics Forum - Statistics Help
 

Go Back   Statistics Forum - Statistics Help > Statistics Help > Probability


Reply
 
Thread Tools Display Modes
Old 06-24-2013, 04:01 PM   #1
Steve_in_MA
Junior Member
Points: 672, Level: 13 Points: 672, Level: 13 Points: 672, Level: 13
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Jun 2013
Posts: 1
Thanks: 0
Thanked 0 Times in 0 Posts
P_value for Pearson Correlation



Working on a rather large dataset - many different things that I am attempting to find a correlation for...using Pearson correlation. (BTW - I am a programmer - stats & probability I am learning and attempting to use as I go.) Anyway, I am having some difficulty simply using Pearson correlation to correlate (see attached pdf for a small sample dataset) The problem is Pearson is giving a good correlation value for data that should be somewhat uncorrelated - for example:
> e
[1] -1.061894 -1.061769 1.944929
> f
[1] -2.223669 -2.215122 -2.074776
> cor(e,f)
[1] 0.9986946 <<<< would expect this to be somewhere between .3 and .6


> cor.test(e,f)

Pearson's product-moment correlation

data: e and f
t = 19.5519, df = 1, p-value = 0.03253
alternative hypothesis: true correlation is not equal to 0
sample estimates:
cor
0.9986946


I see that the first 2 pairs are correlated - but the last pair should move things away from a high correlation. What is going on here? The p-value for the correlation leaves this correlation as statistically significant.
Attached Files
File Type: pdf sample Sheet1.pdf (10.2 KB, 80 views)

Last edited by Steve_in_MA; 06-24-2013 at 04:03 PM.
Steve_in_MA is offline   Reply With Quote
Reply

Thread Tools
Display Modes



All times are GMT -5. The time now is 08:17 AM.


Sponsored Links
Talk Stats - Stat123 - Standard Normal Table - Chi Square Table - Statistics Blog

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 2008-2013, StatisticsForum.com