0
$\begingroup$

I've done two-way ANOVA with interaction (edit) and now I want to do a post-hoc test. I have 2 factors and both have just 2 categories. I've found the function HSD.test from agricolae package and I really like that it gives you the groups and letters regarding how the groups differ. But I'm not sure if this function can be used with two factors or if it works right only if you have one-way ANOVA. Does anybody know?

Edit: I did the ANOVA to test a hypothesis that neither of the 2 factors has an effect on the continuous dependent variable. I rejected the hypothesis and now I want to know how the 4 groups differ between themselves. How should I test this?

Improve this question
$\endgroup$
6
  • $\begingroup$ Welcome to the site. Can you share a bit more detail about your data? Also, if each factor really only has two categories then there’s no need for post hoc tests. To see why, consider the null and alternative hypotheses of ANOVA. $\endgroup$ Commented Dec 13, 2024 at 1:15
  • $\begingroup$ Actually it's a bit more complicated, in total I have about 10 blood parameters (continuous, e.g. concentration of Na) and 2 factors - season (start/end) and locality (A/B). I did two-way ANOVA with interaction for each parameter seperately with Holm correction. To present my results, I would like to do a boxplot for each parameter and I should somehow highlight between which boxes is a difference... So I liked that the HDS.test gives me the letters which I could write above the boxes. But now I realized I probably should do the Holm correction to the HSD test as well, but I don't know how. $\endgroup$ Commented Dec 13, 2024 at 11:58
  • 1
    $\begingroup$ Thanks. Please consider adding the info from your comment to your question as it provides important context. It now sounds like you want to compare the results of different significance tests, which is not what Tukey's HDS is for. $\endgroup$ Commented Dec 13, 2024 at 17:43
  • $\begingroup$ I'm sorry, I'm quite confused. Could you please elaborate? I'm sorry for my ignorance, but I'm new to this.. $\endgroup$ Commented Dec 13, 2024 at 18:23
  • $\begingroup$ Ah, ok, I see what you mean, thanks for the edits. I still think you may have little more to do, but see this answer: stats.stackexchange.com/questions/115818/… $\endgroup$ Commented Dec 13, 2024 at 19:38

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.