for example, let's say there is a policy that randomly allocated immigrants to every municipality of a country, and I am interested in the effect of increased immigrant population on voting outcomes. I can observe every voting outcome in the municipality, so in a sense, I have the entire population. why do I have standard errors? why is this a 'draw' from the population? is the idea that if I were to go back and do it again, it could be the case that something different could have happened if we went back in time and did it again, so the 'true' effect is still out there?

  • $\begingroup$ Take a look at this question. $\endgroup$ – Dimitriy V. Masterov Jul 4 '19 at 0:09
  • $\begingroup$ For the same group of study subjects/people, it can be a population and also can be a sample depending on the purpose/objective of your study. $\endgroup$ – user158565 Jul 4 '19 at 1:07
  • $\begingroup$ I think this answers my question. so with this interpretation of the population, a repeated sampling would be the conceptual idea of going back, and letting the conditions of the policy change differentially? so like in the marijuana example, if we went back and another state legalized first, that will give us another beta. or in mine, if we went back and let it play out again, different places would receive different number of immigrants at different times-which may very well give a different beta etc.? $\endgroup$ – Steve Jul 4 '19 at 14:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.