A long term experiment was initiated in kharif 1985 with four fertility levels (0%, 50%, 100% and 150% of RDF) with 6 replications in RBD. However, after 10 cropping cycles there was severe deficiency of Zn in 150% RDF treated plots. Hence out of 6 replications four replications were superimposed with different Zn and OM treatments. Two replications were left as such. So, now the question is how to analyse the data for a valid conclusion.
since you have change across time I would try a longitudinal model. I would see this as somewhat like a simple clinical trial of one experimental drug versus standard treatments.A really good introduction to longitudinal models is Prof Marie Davidian's class notes available at this link.
https://www.stat.ncsu.edu/people/davidian/courses/st790/notes.html
Good luck, D. Booth
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis