Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have 2 columns with numers in them, x and y. I'd like to find out what y could be if x = n and how strong the correlation is, if it's statistically significant or not.

How do I do that?

Pretty basic question, but I've forgotten all my math - feel free to give me something to read. Any help is much appreciated

share|improve this question
    
Look up linear regression and the Pearson correlation coefficient. –  J. M. May 9 '12 at 12:44
    
You would learn this in a statistics course. There is a StackExchange for that stats.stackexchange.com but I don't know whether they welcome beginner questions or not...take a look to find out. –  GEdgar May 9 '12 at 13:00
    
@GEdgar thank you –  Jakob May 9 '12 at 13:09
    
@jacob looking up the Pearson correlation coefficient will define it for you and tell you how to compute it but it may not tell you how to test using it. There is a test of a null hypothesis of no correlation which means the population correlation is 0. Statistics books have tables for the sample correlation that tell you what value for the estimate is statistically significantly different from 0. Tables were dervived by F. N> David in the early 1900s. –  Michael Chernick May 17 '12 at 15:29
    
There is also Fisher's transform that transforms it to a normal distribution wher the tables of the standard normal can be used to test significance. –  Michael Chernick May 17 '12 at 15:29
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.