high

http://www.stat.psu.edu/~resources/

http://www.stat.psu.edu/~jglenn/

Awesome!

http://www.stat.psu.edu/~jglenn/stat503/

http://www.stat.psu.edu/~jglenn/stat503/04_blocking/04_blocking_latin.html

http://www.edgarweb.org.uk/choosedesign.htm

R as effect size (Andy Field, p357)

R square = SSM/SST

However, (quoting Andy Field) the measure of effect size is slightly biased because it is based purely on sums of squares from the sample and no adjustment is made for the fact that we're trying to estimate the effect size in the population. Therefore, we often use a slightly more complex measure called omega squared --

w squre = SSm - (dfm)MSr / SSt + MSr
dfm = the number of experimental conditions - 1

** Most of the time, it's not that interesting to have effect sizes for the overall ANOVA because it's testing a general hypothesis. Instead, we really want effect sizes for the contrasts. Planned comparisons are tested with the t-statistic and we can use the equation --

r(contrast) = square root (t square/ t square + df)