This is more of a straight-up biostats question (no transcriptomics involved, although I'll be popping the numbers into R), but I didn't get any feedback when I posted to CrossValidated, so if someone statistically-minded in this community could provide some insight, it would be appreciated.

We have a dataset with Time and Treatment as independent variables, and tumor growth volume as a dependent variable. The experiment was repeated on three occasions, with multiple time points and multiple observations per combination of factors for each experiment instance (so 8 observations at time = 12, treatment = 1, experiment1, etc). There's a subset of common factors for time points and treatment between all the experiments that I am focusing on. Note that the samples were not the same physical samples between experiments (so it's not strictly a repeated measures design).

**We'd like to figure out if we can pool the data from the individual experiments, or if the variation between batches is too significant to merge them.** Originally this seemed like a design for MANOVA, but when I put all of the data in the format I figured I need for R, it looks like there's actually only one dependent variable (growth volume) if we treat experiment as an independent variable. I also read in some other CrossValidated answers that there are regression methods that are preferable to MANOVA nowadays anyway, but I wasn't sure where to start with determining an appropriate approach.

Thanks!