If I'm looking at existing statistical data, how do I try to find a correlation between two variables over time? Scatterplots have confused me, because they don't seem to reflect the passage of time, but just where x and y meet. Pearson's r doesn't seem to do it either w/r/t time.
posted to Ask Metafilter by trondant at 12:40 AM on December 11, 2007 (13 comments)

What's the most you can cost someone who sent you a prepaid business reply mail envelope? I just got one of these, fishing for a tax-deductible donation to a 501(c)(3) group, one I would pay money to see stomped to death. I want to hear from people who've done this and suffered no repercussions, as well as any former or current USPS employees who might know where that fine line is. I'm also open to anything that keeps me on the mailing list, but still costs them more in postage than it's worth.
posted to Ask Metafilter by trondant at 9:27 PM on February 22, 2005 (15 comments)