I have 2 species of wildflowers. Using scanning electron microscopy. I took images of the petal surface. On the petal surface I measured the height and width of individual cells. I have 1 to 3 individuals per species. I want to compare these measurements between species Are the measurements of cells independent or do I need to use a statistical test to account for the lack of independence? If so, what tests should I consider?

  • $\begingroup$ What is the goal of the analysis? Are you interested in determining if they are different, or is this purely exploratory? $\endgroup$ – Demetri Pananos Feb 10 '19 at 0:01
  • $\begingroup$ I observed that there are sometimes conical cells on the surface of the petals. I wanted to see if there was any significant differences between the height of the cells between two closely related plant species. $\endgroup$ – StatsNPlants Feb 10 '19 at 0:51
  • 1
    $\begingroup$ There is very likely correlation within an individual plant. Try a mixed effects model, where the random effect is the individual and the fixed effect is species. Can you post some a little data to help us cement the problem in our minds? $\endgroup$ – Demetri Pananos Feb 10 '19 at 1:40

Depends a bit on the nature of your question, but I should think that height and width of cells are going to be highly correlated and both are measures of cell size. I would probably just multiply height by width to get a better approximation for cell area/size, Alternatively you could run a PCA analysis.

| cite | improve this answer | |
  • $\begingroup$ I mainly want to compare the heights of the cells between species. But I do not not have many plants. I made multiple measurements on the same plant. Is there any way to compare? $\endgroup$ – StatsNPlants Feb 10 '19 at 0:52

Your Answer

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

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