But you don't even need math to do these :P The problem with statistics is that your data could be the result of random chance, that's why there are tests like the t-test to see if your data supports your hypotheses. You can't really conclude anything from data unless you test it first. It could be that there is no relation between the average topic rating and the join date, that's why I thought it would be cool to see a [url=http://en.wikipedia.org/wiki/Student%27s_t-test#Slope_of_a_regression_line]t-test.[/url] You already have 2 of the 3 components to do the test. ^b is your calculated slope and b0 is 0, all you have to do is use [url=http://upload.wikimedia.org/math/f/8/6/f86ecf5eb9ab362755f02557703fa2af.png]this formula[/url] and compare it with a [url=http://www.sjsu.edu/faculty/gerstman/StatPrimer/t-table.pdf]t-table[/url] to see your your [url=http://en.wikipedia.org/wiki/P-value]P-value[/url]. If your P-value is lower than you 0.025 , you can say that there is a linear relationship with a significance level of 95% (because of the 0.05/2). Or you could just input your data into a statistical program like 'R' and let R do the work. EDIT: If you're using all data available instead of a sample, feel free to ignore this post.