Hi,
I use flexible factorial design to model the longitudinal MRI dataset with 3 group having 2 time points (3x2). For this model, I created three factors (subjects, group and time). Now I intend to enter the age as a covariate of no interest in the statistical model. On going through the CAT12 manual, I understand that only the variables changing over time can be entered as a valid/useful covariates. So, I tried entering this covariate (age) in different ways. However, it is throwing off errors listed below. Is there a specific way to enter the covariates (age) for each time points 1 and 2 in the model? Can you please give an example?
1.(Index exceeds the number of array elements (66)
2.(Item 'Vector', field 'val': Size mismatch (required [Inf 1], present [66 2]).)
-In addition, from the CAT12 manual I noticed that incorporating the total brain volume/intracranial volume as a covariate (proportional scaling/ANCOVA) in the flexible factorial design is not necessary. On the contrary, at least some research articles applied the TIV/TBV as a covariate in the study. Does this bias the results in general? Any comments on this please?
http://www.ajnr.org/content/ajnr/36/12/2219.full.pdf
https://www.sciencedirect.com/science/article/pii/S0022395608000782
Could anyone please provide suggestions to resolve this issue?
thanks,
sincerely,
Varun
|