New Member
Dear all,

I have been looking everywhere for answers (forums, books, websites etc.) but I am still struggling with a few statistical questions. I hope this forum will help me get some answers!

My research question deals with diseases affecting benthic organisms.

We conducted 6 transects at 6 different sites (6 transects per site) during two different seasons. Each site subdivided into two zones. During our transects, we counted the number of lesions and we determined the percent coverage. We used these to calculate the occurrence of disease by dividing the number of lesions by the % cover (this is our continuous response variable). Therefore in our model we have: Season (fixed categorical factor), Site (fixed categorical factor) and Zone (random categorical factor nested within Site.

1- We were interested in testing the influence of the factors on disease occurrence. I ran a PERMANOVA and got understandable results. I was also advised to run a GAMM instead.
Question: Which one would you recommend between PERMANOVA and GAMM? I tried to run a GAMM but can’t figure out how to add a random nested factor (Zone). Should I get the same results using a PERMANOVA or a GAMM?

2- We would like to test the potential relationship between disease occurrence and benthic coverage.
Question: Does it make sense to add % cover as a covariate in my PERMANOVA design knowing it is already involved in the calculation of the response variable (disease occurrence)?

I have more questions but I thought I should go step by step to make this post clearer. Any help would be very much appreciated.

Many thanks in advance!