griffithdan
commented
2 years ago You have P_lt values that are significant. These pairs co-occur less than expected by chance. The summary table returns the significant pairs so you can inspect them, but you can also get the probability and effects sizes for every pair. I can't comment on the biology or the specific application without knowing more and looking at the raw results. Feel free to email me if you desire that and I will do the best I can to take a look.
I was hoping to run visNetwork on some benthic data. My first step was to use the cooccur package in RStudio.
As I understand it, "if p_lt < 0.05, then the species pair co-occurs at a frequency lower than we would expect to find by chance. If p_gt < 0.05, the pair co-occurs at a rate higher than we would expect to find by chance. Since we’ve stored only significant interactions, either p_lt or p_gt will be less than 0.05 for each row."
However, all of my P_lt values are <0.05 and all of my P_gt values are .99+. Wouldn't this indicate that all of the associations in my cooccurrence network are essentially random (or that the association would be less likely to occur than even by chance)? Does this mean this data is useless for a true network analysis? Why would it say that it only keeps the significant interactions when the stored associations have P values that suggest otherwise?
Furthermore, as I understood it, the cooccur paper indicates that P_lt+P_gt+P_et should equal 1; however, in all of my results, P_gt=~1 while the associated P_lt also holds some value between ~.01-.04, which obviously totals to >1. Not sure how to interpret this or if I've misunderstood something.
Any advice would be greatly appreciated! Thanks.