I've just carried out an ANOVA on my data

My data is comparing juvenile abundance, based upon distance from the nearest adult A and another factor B

Based upon the output A and B are non-significant, however A:B is significant

Just wondering what the best way is to interpret this.

  • 2
    $\begingroup$ You should plot the data. That should help with interpretation. $\endgroup$
    – Roland
    Feb 14 at 6:07

1 Answer 1


It is fairly common to get this observation when conducting interaction effects test in ANOVA.

Since you've found the interaction to be significant, you can interpret that the variables A and B jointly are able to explain the variation in your response variable.

One important thing to note is that when you've found the interaction to be significant, you have to keep the individual variables in the function.

In this case, A:B is significant so obviously interaction term should be present. In addition to this, you need to keep A and B in the equation as well even if they're not significant.

  • $\begingroup$ This interpretation of a significant interaction is either too vague (what does "explain the variation" mean?) or a bit of an over-reach, because there remains the possibility that other variables are needed for an adequate model. You can find many threads here on CV that discuss this issue> $\endgroup$
    – whuber
    Feb 14 at 15:40

Not the answer you're looking for? Browse other questions tagged or ask your own question.