"Fuzzy" RDD and Swiss Religion: Checking the Numbers: Causal Inference Bootcamp

Here we look at the ‘Protestant Work Ethic’ results in table form, and also use instrumental variables to deal with the “fuzzy” nature of the RDD.
Part of Duke University's Causal Inference Bootcamp: watch modules on this topic and much more at modu.ssri.duke.edu

[@leopan895]

it's really awesome to have such great lesson. Thx

[@TheMarketing17]

Great job of simplifying complex method! It really helps my study!

[@DukeEHDi]

Thanks!

[@sangheehwang9679]

It was really amazing video and an interesting example. Thank you so much! :)

[@reajulchowdhury8534]

awesome... you are good in explaining .. It helped me a lot

[@anhthuduong7411]

Thank for your explanation. It helps my study alot

[@anqishan4000]

Awesome video, really interesting paper too

[@panda0229girl]

Very good video. Thx so much. May I have a small question: I agree the fuzzy underestimate the treatment effect, but if the regressor in the OLS (treatment dummy) and 2SLS (share of protestant) are different, why the magnitude can be compared?

[@yiranzhang1858]

good video. very useful for me

[@DukeEHDi]

Thanks, and good luck with your work!

[@SulmanSaleem]

I just wish that that you had the “stat table” on the board. It would have been easier to follow.