Reference no: EM131876247

Question: Google Foundation has hired you as part of their impact evaluation team. Your first assignment is to evaluate a Randomized Control Trial that they have implemented before you arrived. Two years before you arrive, they have installed free broadband internet access to 100 Indian villages, which they randomly selected from a list of 200 villages that all had expressed their interest in participating in the project. All 200 villages are very close to one another in the same small and densely populated region within one Indian subdistrict.

The board asks you to estimate the Average Treatment Effect of their intervention on village-level GDP per capita.

a) What data (which variables) do you require from their project team to answer this question?

b) Which regression equation would you want to use these data for? Describe each variable and subscripts and explain the interpretation of the regression coefficients.

c) What potential concerns would you raise to the board about interpreting your findings from b) as the Average Treatment Effect of the program? Discuss at least 3 and be specific about each concern.

d) The board is also interested to know whether the program affected village-level GDP per capita differently as a function of the initial average years of schooling in the village. They would also like to know whether the program had different effects as a function of whether or not the village had road access at the time that broadband got installed. Which additional data would you need to request from their project team, and which regression equation would you use these data for to answer these questions? Explain each variable and subscripts and explain the interpretation of the regression coefficients.

e) After you have started working on this, the former leader of their project implementation team tells you that she is concerned that the program may not actually have randomly allocated treatments across the 200 villages, and that some cheating may have gone on (she heard reports that the richest villages were more likely to be put into the treatment group). What concern would this give rise to in your estimation? What type of village-level data would be the best-suited to verify whether the project leader's concern is true, and what regression specification would you use these data for to investigate whether her concern is true?