Main Article Content
Existing studies on the impacts of amil programs are still rare and potentially contain bias in their impact estimates. This paper explains the selection problem common to simple difference method employed by these studies. Then this paper introduces two evaluation designs, difference-indifferences and randomized experiment, that can more credibly estimate program impacts, and still simple enough to be independently conducted by amil institutions.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.