I am building a GLMM model with a binomial response variable, and I am having trouble determining which combination or nested or crossed effects to use. My situation is the following:

1.) I have citizen science data of species observations which include IDs for the citizen scientist observers.

2.) I have nearby weather monitoring stations with important predictor variables that I am using. These stations have unique IDs.

3.) Data gathered from weather monitoring stations is time-matched to when citizen scientists submit their observations.

To summarize, I have unique weather stations, unique citizen scientists who make observations at multiple unique weather stations, and these observations occur at different times.


1 Answer 1


Is each scientist only associated with 1 weather station? or could there be some scientists that have data for more than one station?

In the first case the scientists would be nested in weather station (station can have multiple scientists, but each scientist only measured relative to one station).

If every scientist measured with every station (and every station with every scientist) then the 2 are fully crossed. If some scientists have multiple weather stations, but not all, then you still have partial crossing.

I expect that time is crossed with both scientist and weather station, basically you have all the scientists and weather stations at all of the time points. Technically you could have some nesting here (only stations 1 and 2 were measured in 2020, and stations 3 and 4 were measured in 2021, in which case station would be nested in time point).

Hope this helps, but if it is still no clear then maybe posting a link to your data, or a more detailed description of how the data was collected would help.

  • $\begingroup$ As you put it, the scenario is partially crossed; some scientists go to multiple weather stations and observe multiple times at any number of weather stations. Other scientists may only submit one observation. Each weather station has multiple observations with multiple scientists at different times, and each scientist MAY visit multiple weather stations and make multiple observations at different times. $\endgroup$ Commented May 16 at 11:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.