it needs to use stata application

it needs to use stata application


1 ECON 104 Summer 2023 Assignment 1 In this assignment, you will estimate the effect of an effort to increase voter turnout through encouraging phone calls. Voters were randomly assigned to get (or not) an encour- aging phone call. As can be seen by examining the data, not everyone assigned to get a call (treat real = 1) actually answered and listened to the call (contact = 1). Due to the randomization, the analysis in this assignment should generate a “gold standard” estimate of the effect of getting assigned to receive a phone call on voter turn out. The data is from: Arceneaux, Kevin, Alan Gerber, and Donald Green. “Comparing Experimental and Match- ing Methods Using a Large-Scale Voter Mobilization Experiment.” Political Analysis, 2006, 14, 37-62. This study was intended to assess the effectiveness of get-out-the-vote efforts. In the experiment, the authors attempted to deliver the following message to those selected for treatment: “Hello, may I speak with (name of person) please? Hi. This is (caller’s name) calling from Vote 2002, a nonpartisan effort working to encourage citizens to vote. We just wanted to remind you that elections are being held this Tuesday. The success of our democracy depends on whether we exercise our right to vote or not, so we hope you’ll come out and vote this Tuesday. Can I count on you to vote next Tuesday?” For this homework, use the dataset on Canvas that corresponds with the first letter of your last name. The dataset includes the following variables (among others): • vote02: Voted in 2002 – the outcome of interest • treat real: Assignment to the treatment group • contact: Answered the phone call and responded to the question “Can I count on you to vote next Tuesday?”, regardless of their answer 2 • newreg - Newly registered voter • busy - Whether the phone line was busy when the call was made • age - Age of the individual in years • female - Gender of the individual • vote00 - Voted in 2000 (prior to treatment) • vote98: Voted in 98 (prior to treatment) • state: State of residence 1 for Iowa and 0 for Michigan • comp mi: Value of 1 for - competitive district in Michigan (Michigan A in the paper) • comp ia: Value of 1 for - competitive district in Iowa (Iowa A in the paper) 1. How many individuals are in the data set? How many were assigned to the treatment and control groups? How many individuals in the treatment group actually received and listened to the call? What percentage of individuals in the data voted in the 2002 election? 2. Provide evidence that the randomization worked by comparing the means of the sample characteristics in the treatment and control groups. Please create a clean table that includes columns with the means for each group, the difference between the two groups, and the p-value of the difference. The table should be comprehensible on its own. 3. Is the table you produced in question 2 consistent with the randomization being cor- rectly implemented? Why or why not? Explain. 4. Estimate the difference in the voting rates for the treatment and control group. How big an effect did getting assigned to get a phone call have on the probability of voting. Is the difference statistically significant? Do you think the difference is large in a practical sense? Explain your answer. 3 5. Create a carefully labeled table where each column corresponds to a regression. The first column contains the parameters of a the regression vote02i = β0+β1treat reali+ui. In each following column you should add one more covariate (control variable) to the regression. 6. What effect does adding covariates have on your estimate of the treatment effect? What does this tell you about the relationship between the covariates and being assigned to the treatment group? 7. Will comparing the voting rate of the treatment and control groups generate an unbi- ased estimate of the causal effect of getting assigned to get an encouraging call on the probability of voting? Why or why not? Explain. 8. Please attach all of the code you wrote to generate your results.