finding correlations

gryffin13

New Member
It has been a while since I have taken statistics. I used to know it pretty well but now sadly it has been lost. I'm looking for help in finding correlations between two things. The tricky part is that I don't necessarily have numbers to go by. I'll give you an example. I want to know if a particular region is doing better with completing projects. So I have data that shows all of the projects, which ones were complete and which region they belonged to. I am aware that I could just calculate a percent for each region and compare, but that might just be random. I know that I need to start with a null hypothesis that each region has a 50% chance of completing and then find what their actual results are to determine if their difference is significant. Problem is, I can't remember how. I am aware that this will only show correlation and not causation, but that is ok. I just want to be able to take data that I have and say blank is the factor with the highest correlation of a a project being completed. I don't know if that is clear or not, but any help is appreciated. Thanks.

el_rondinho

New Member
You can not calculate a correlation between the region (nominal scale) and the percentage of completion (metric variable). I think what you want to do is to test whether the means between the percentage of completion differs significantly between regions. For this purpose you could run t-tests for mean difference. There are pretty good explanations on how to do this in the internet. I would recommend to google it...