I compared the effect of different concentrations of antibiotics with the anova. As I am only interested in which concentration is significantly different compared to the control I performed a post-hoc Dunnett test. The data of controls and baseline are the same for the different types of antibiotics. However, the results of the post hoc test between baseline and control is significant in 1 of the post hoc test while its not significant in the other. I don't understand how this is possible if they have exactly the same values. What would be a explanation for this? Please see graphs for better understanding.

enter image description here

enter image description here


  • $\begingroup$ The power of different tests is unequal. Dunnett's test may need a more severe type I error for correctness. Although behind a pay wall. It may help to read this. $\endgroup$ – Carl May 11 at 8:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.