I am trying to analyze distance data that were collected via line-transect survey method. I was wondering, when you include a categorical covariate in the formula for your key function, do you need to make sure that a value for that variable is specified for those transects that did NOT result in any observations (in the flatfile)?.
In my mind, it should seem to make a difference in the hypothetical case of one category resulting in many transects with >= 1 detection, and another category resulting in many transects without any detection.