hello.
i hardly know anything about stats. i'm writing a paper with some data that i need to compare. i have a control and an experimental series. i graphed the series to compare their regression lines. their slopes are similar, and their intercept has a difference of about 2.3 units. but i need to know if the two data sets are scientifically significantly different. they data set is pretty small, so i'm afraid it wont be. i looked around online, and found the analysis of covariance (ANCOVA) test. is this the correct test for my purposes?
so i ran my results through an online ANCOVA calculator, but i don't understand the results. could someone explain these results to me? i suspect it's telling me that, in fact, my data is not different enough.
thank you for any help you can offer.
i have an HTML results page from the calculator. i'm not sure how to post this into the forum. i'm gonna try just copy and pasting... i might have to work on that before i get the results up.
matthew
Dependent Variable Sample A B Total n 15 17 32 Observed Means 13.7333 13.2353 13.4688 Adjusted Means 13.77 13.2 13.47
Aggregate Correlation within Samples: CV vs DV r = 0.86 r^{2} = 0.74
ANCOVA SUMMARY
Source SS df MS F P adjusted means 2.59 1 2.59 0.78 0.384404 adjusted error 96.24 29 3.32 adjusted total 98.82 30
Test for homogeneity of regressions:
Source SS df MS F P between
regressions1.05 1 1.05 0.31 0.582104 remainder 95.19 28 3.4 adjusted error 96.24 29