All questions based on the paper "Does Price Matter in Charitable Giving? Evidence from a Large-Scale Natural Field Experiment," by Karlan and List, The American Economic Review (2007). Code for regression of all the questions is to be done in Python in Jupyter Notebooks environment. 1. Table 1 Q1. Load the "charitable giving.csv" dataset and run a regression to assess whether the average "Number of months since last donation" is significantly different between treatment and control. Interpret the relevant regression coefficients and compare the regression-based comparison to the group-specific means reported in Table 1 of the paper. Q2. Is the difference in "Number of month since last donation" between treatment and control statistically significant (at the usual 95% confidence level)? Is this the result you expected? Q3. More generally, describe the take-away from Table 1 in the paper. 2. Response rate regressions Q1. Run a linear regression of response rate (the donation dummy) on the treatment dummy (and an intercept). Interpret both coefficients and compare them to the results presented in the 1st row of Table 2a. Q2. Run a regression on 3 dummies for match ratio treatment (1:1, 2:1, and 3:1 and an intercept). Interpret all 4 regression coefficients. Q3. Calculate the response rate difference between the 1:1 and 2:1 match ratios. Q4. Based on the regressions you just ran and more generally the results in Table 2a, what do you conclude regarding the effectiveness of using matched donations? 3. Response rates in red/blue states Q1. Repeat the regression of response rate on treatment and an intercept (do not include separate match ratio dummies). But this time, base the regression only on respondents in blue states or red states i.e. run 2 regressions, 1 on each of the 2 sub-samples of data. Interpret the coefficients in both regressions. Is the treatment more effective in red or blue states?