Sean C. Thomas @Sean-Thomas-4

I am working with real-world data. I want to test the effect of outside missionaries (specifically their departure) on the number of Catholics in a jurisdiction (either countries or dioceses). My sample is all jurisdictions, or at least all that have been colonized at some point.

Suppose I have the number of outside missionaries in every jurisdiction, measured each year (independent variable). I also have the the number of Catholics in every jurisdiction, measured each year (dependent variable).

I know there are repeated measure designs that can work with repeated measures of the dependent variable (number of Catholics). But, at least according to this 2013 article (Article Analysis of repeated measurement data in the clinical trials

), RMDs don't do a good job accounting for repeated treatments (the progressive numerical decline of outside missionaries). Are there any modifications to RMDs that would enable this? I have in my back pocket the idea to perform a standard repeated measure design starting when each country crossed a specific proportion of maximum outside missionary numbers (like when OM = 0.5 * OM(max)). Wherever I set this would be arbitrary, though, so I would rather use all data available to me.

I am also aware of mixed models analysis. It seems like there are a lot of moving parts to working with mixed models, so it is a bit hard to tell if these would be right for what I am trying to do.

Thank you in advance for your help.