Open ehumph opened 4 hours ago
Note - # of visits as the outcome should probably be modeled with a poisson regression. Rate is a better choice for linear regression, since it's technically bounded between 0 and 1. Mea culpa.
Yeah might revise not to have year treated as numeric either. Might have to think more deeply about these!
could maybe add this resource - it's pretty good: https://tilburgsciencehub.com/topics/analyze/regression/linear-regression/regression-summary-output-r/
Slide #35. There are two beta-sub-2 on the slide. One 2 should be a 3