Another RMark question: Factors as individual covariates

[kquail]

I realized my wording in my question was a bit misleading. I just edited my post above. I should say that I don't want to estimate a parameter for any of the broods individually. I really just want to take into account the overall effect of being from different broods when estimating survival, in the same way I would treat it as a random effect in a GLMM.

[jlaake]

So that is an entirely different question and Evan may want to split this off into a separate thread maybe with a subject of random effects. I have not kept up with development in MARK since I retired 8 years ago, but as far as I know, the only random effect models in MARK are individual random effects and not a group (brood) effect. The only capability I know of is variance components whereby you would group by brood and fit an effect for each brood and then get a mean and variance across broods. I'm sure there is a section on this in Cooch and White. Maybe Gary, Evan or someone else may know more about this than I and can join in here.

I have fiddled with random effects in marked package and you could fit a random brood effect in a CJS model there. However, I'm not sure I ever worked out the proper random effect prediction error. Largely dropped that when I retired.