I am looking at differences in bird mass change over the course of a month and whether behaviour effects mass gain.
API is a scale from 1-7 given to the birds to rate how fat they are. These same birds are resampled across the span of a month until they leave to migrate. During this month mass gain is super important for breeding success.
As API is ordinal, I am looking at using a clmm to see whether the amount of vigilance (average_duration) and the mean amount of aggression the birds give (mean_aggressiveness) or receive (mean_receive) effects the amount of mass gain a bird experiences throughout the month. I am also counting Ring ID as a random effect.
I am currently using the "probit" link and "equidistant" threshold but I am not sure if they are correct?
clmm1<- clmm(API.change~average_duration + mean_receive + mean_aggressiveness + (1|Ring_ID), data = API.av.dat,link = "probit", threshold = "equidistant", Hess = TRUE)
Any thoughts on whether this is correct? What link functions should I be using instead?
I have ran the clmm without a link function, with the link function above and without the threshold "equidistant", this changes my values in my table so I am not sure whether what I have at the moment is correct.