AQMeN - Applied Quantitative Methods Network

Dear all,

I am trying to decide which post-hoc contrasts to use, and would like some advice.  I have a large data set (n = 570), and am looking to run a priori contrasts following a one-way ANOVA.  I have unequal numbers in my groups however (ranging from 17 to 218), and therefore intended to use Tukey's HSD.  The paper which we are replicating used Scheffe's contrasts however, so I am now a little confused as to which test to run.  I have tried researching this topic, but there doesn't seem to be much consensus over when to use the various tests.  Please could someone shed some light, or refer me to a relevant paper?

Thanks,