0
$\begingroup$

My single factor MANOVA showed that there is a statistical difference between the red and black (which represent the effect of treatment given) : the P value was <0.05. But in PCA there seems to not be any difference between the groups. Why is the PCA not aligning with the results from the MANOVA?

For context, I am studying geometric morphometrics and each point is one specimen.

enter image description here

Improve this question
$\endgroup$
3
  • $\begingroup$ Even if there are two groups, that's not guaranteed to be evident on a plot of PC2 and PC1. It might be clearer on plots of other PCs. Note that a significant difference does not mean that groups never overlap. As it is, in your plot I see more red in SW (-, -) quadrant and more black in NE (+, +). . $\endgroup$
    – Nick Cox
    Commented Apr 5, 2019 at 23:11
  • $\begingroup$ Thank you so much for your reply. Please can you tell me why it wouldn't be guaranteed to be evident on a PC plot? $\endgroup$
    – Mariam
    Commented Apr 7, 2019 at 21:39
  • $\begingroup$ Why should it be guaranteed? But rather note that some structure is evident on the plot you show. Add to the plot the bivariate means and medians for your twoll groups. $\endgroup$
    – Nick Cox
    Commented Apr 7, 2019 at 21:51

0

Reset to default

Your Answer

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

Browse other questions tagged or ask your own question.