Seshardi writes: "I want to know which process variable affect my result most". This is not a correlation problem - this is Best Variables problem: searching through a large number of variables to determine a short list of the most influential factors to use in a model. There are several methods for addressing this question. I find a Bootstrap Forest to be very helpful for this purpose. George Hurley presented a paper on Bootstrap Forest in JMP at the 2012 NESUG conference - here is the Lex Jansen link: http://www.lexjansen.com/mwsug/2012/JM/MWSUG-2012-JM04.pdf. I did a presentation on Bootstrap Forest in Base SAS at NESUG but I understand you are working in JMP. My NESUG presentation was just a sub I wrote overnight to fill in for an author who unfortunately had to cancel; I don't think it's in the proceedings but I will be posting a slide show on it in the next few days. If there is specific outcome you want to predict that is amenable to regression analysis, you will want to consider the XCSTAT macro by Raimi and Lund: http://www.mwsug.org/proceedings/2011/stats/MWSUG-2011-SA03.pdf. It's only available in SAS at this time, not JMP.
... View more