0
$\begingroup$

I am analysing results from a study which was a 2 (stimulus type) x 2 (distance) x 2 (speed) x 2 (cardiac cycle) x 2 (grouping variable-between subjects) mixed ANOVA, all of the factors have 2 levels and only one of them is a between subjects measure, the rest are within.

I have a significant interaction between them all. I am unsure how to interpret it or how to find out where the interaction came from.

I have split the data with the between-factor and that again had some significant results. group 1= stimulus * distance * speed * Cardiac group 2 = speed * stimulus , stimuli *distance * speed * cardiac, and distance * speed * cardiac.

I can't further split the file into Cardiac cycle, I think because the data is in a wide format and the cardiac cycle is spreed across a few columns and this is the case for speed, distance and stimulus too.

There seems to be nothing online or in books about 5-way ANOVAS!!!! Any help and advice greatly appreciated.

Thanks,

$\endgroup$

1 Answer 1

1
$\begingroup$

If you understand how to interpret 2-way and 3-way interactions, then the concept applies equally to higher-way interactions. Any interpretation will contain a lot of "if $x=a$ and $y=b$ and $z=c$".

I suspect that it will be very hard to learn anything from a 5-way interaction. Consider analysing only a subset of your data.

Please don't take me wrong, but it is better to consider the analysis before doing the study. If you plan on running an analysis that is extremely hard to interpret, then a less ambitious study design might be called for.

$\endgroup$
2
  • $\begingroup$ Thanks for the reply. I am more interested in the cardiac cycle and grouping variable, would it be acceptable to focus on the interaction with those variables? $\endgroup$ Commented Jul 26, 2018 at 10:03
  • 1
    $\begingroup$ Certainly. Your analysis should be guided in the question you are interested in. $\endgroup$ Commented Jul 26, 2018 at 11:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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