1) Worms in Kenya. The dataset, ted_miguel_worms.dta is from Ted Miguel's (UC Berkeley) and Michael Kremer's (Harvard) de-worming project in Kenya. Please read the paper to familiarize yourself with the experiment before answering this question. You can find the paper on the course website under Assignment #3. Miguel and Kremer randomize over schools (not individuals) and to introduce de-worming drugs to randomly selected treatment group to estimate the effect of deworming on school attendance. a. First, why randomize at the school level? Think what issues might arise in your evaluation if you randomize at the individual level. b. Suppose you had pre- and post- treatment attendance records for all schools, describe the calculations/comparisons you do in order to estimate the effect of deworming on school attendance (in words and with a DD grid). Why might you prefer this DD estimate to a straightforward OLS estimate using only post-treatment data (i.e. mean attendance in program schools minus mean attendance in non-program schools after the program)? c. Write down the regression you want to run in to estimate the DD you described in part (b)? Explain the meaning of the interaction term