I have a question regarding a means separation test to test interactions. Is it possible to perform a Tukey's Test with interactions? One of my terms in my model is significant and I want to know which ones they are.
For example my model is
ModelB<- lm(RATING.2~Allele%in%QTL11.Source+Family%in%QTL11.Source+REP)
anova(ModelB)
Response: RATING.2
Df Sum Sq Mean Sq F value Pr(>F)
REP 2 1.301 0.6507 0.9266 0.3993
Allele:QTL11.Source 8 105.021 13.1276 18.6941 < 2.2e-16 ***
QTL11.Source:Family 37 68.644 1.8552 2.6419 6.873e-05 ***
Residuals 100 70.223 0.7022
---
TukeysTestB<- HSD.test(ModelB,"Allele%in%QTL11.Source")
TukeysTestB
NULL
Why am I getting a NULL output, Is this not possible to test?