How do I do post hocs for ANCOVAs?
t= (YbarAdj(i) – YbarAdj(j)) / Sqrt [ MSw(adj) * (1/n(i) + 1/n(j)) + (Xbar(i) – Xbar(j))^2 / SSX(w) ]
where YbarAdj(i) is the adjusted mean of group i
MSw(adj) is the adjusted mean squares within from the overall analysis
Xbar(i) is the mean of the covariate for group i
SSX(w) is the sums of squares within associated with the covariate.
use t(df=N - #of groups - 1) as your critical value.
Advertise on Talk Stats