Handling plant response variables that are leaf position /rank based in repeated measures mixed model

[#1]

Dear All,

Please, I am working with plant response variables that are leaf rank dependent. A good example is the leaf length of all leaves on a plant. I chose the mixed model's analysis platform in JMP PRO 13. Measurements were done four times during the growing period. The treatments were four species and three levels of nitrogen. I followed "the fitting repeated measures with JMP PRO 13 ppt.." below.

I chose the time-dependent covariance structures (AR1, Spatial, Toeplitz and antedependent). However, I could only get results when the leaf ranks for the treatments were the same. My data has different leaf ranks per plant (treatment) because of the differences in the development of individual plants. How can I treat the issue of unequal leaf ranks for plants which develop faster than the others? Finding an approach to handle the unequal leaf ranks is important because there are four different species with different developmental rates (leaf ranks) that are being compared.

My fixed model was: Full factorial of species, N level and leaf rank.

The repeated was either time as a continuous or nominal variable.

The subject was developed as unique numbers for treatments at each measurement date.

I appreciate any help you can provide.

[#2]

Please no community help?

[#3 · Accepted solution]

Here is another resource that might help you.