Dear all
Just a quick query
Say factor X1 has 'a' levels and factor X2 had 'b' levels. Say in a 2
way ANOVA, each of the axb combinations are repeated n times. I have
seen numerous examples of this where the significance of the X1*X2
interaction is looked at and then illustrated on a typical interaction
plot.
However, my question is, say in a multiple regression we have
independent variables Z1, with c levels, and Z2, with d levels...and we
have looked at the significance of the Z1*Z2 interaction in the
model...now, can we still illustrate the Z1*Z2 interaction on a typical
interaction plot if the cxd combinations are not each repeated a
constant number of times, i.e. we could have one combination occurring
100 times, another combination occurring 80 times, another 25 times etc.
Many thanks for your help on this
Kind Regards,
Kim
|