From c75744a0d6efda1d25bba21e890ccaf8012fbbd6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Emil=20Savalanl=C4=B1?= Date: Thu, 20 Oct 2022 01:47:37 +0200 Subject: [PATCH] Update 03_RCT.Rmd typo --- 03_RCT.Rmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/03_RCT.Rmd b/03_RCT.Rmd index 7b0aaaf..544bf39 100644 --- a/03_RCT.Rmd +++ b/03_RCT.Rmd @@ -985,7 +985,7 @@ True 99\% sampling noise (from the simulations) is `r round(sn.RASS.simuls.yB,3) In an Eligibility design, we randomly select two groups among the eligibles. Members of the treated group are informed that they are eligible to the program and are free to self-select into it. -Members of the control group are not enformed that they are eligible and cannot enroll into the program. +Members of the control group are not informed that they are eligible and cannot enroll into the program. In an Eligibility design, we can still recover the TT despite the fact that we have not randomized access to the programs among the applicants. This is the magic of instrumental variables. Let us detail the mechanics of this beautiful result.