I have an R data frame, set up like so:

Site Year Species Total_Observed
A    2000  Bird_1    4
B    2000  Bird_1    2
C    2000  Bird_1    9
A    2000  Bird_2    0
B    2000  Bird_2    1
C    2000  Bird_2    3
A    2001  Bird_1    7
B    2001  Bird_1    2  etc

And it goes on like this for some time. Lots of data, multiple sites, over many years. I'd like to do some kind of regression to see whether bird_1 is increasing over time in the same areas that bird_2 is decreasing over time, but I'm not sure how since this is partly a geospatial problem and it seems to me that bird_1/bird_2 observations can't be averaged due to their correlation by site. How do I approach this problem?


1 Answer 1


Are the sites connected to each other, or can you neglect spatial interactions ? If you can neglect it, just start with fixed effect models with time and site dummies (plus a variable with the other kind of birds). If not, look at spatial regressions.

  • $\begingroup$ Thank you! These sites aren't connected to each other, so I think I can neglect spatial interactions. I'll take a look at fixed effect models. $\endgroup$
    – aeiche01
    Commented Jul 21, 2019 at 17:48

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.