A Guide to a Basic Policy Evaluation Project with R and Python

This tutorial serves as guide for conducting a basic policy evaluation project in Python and R. This project asks you to us the Correlates of State Policy data available from the Institute for Public Policy and Social Research College of Social Science.

You will develop a project using this data set and write a 10 page (double-spaced, 12 point font, printed and stapled) memo that shows the knowledge you gained from this class. To do this project, you will need to first download the codebook describing the data, and then develop a research question that you will be able to answer using the data. You will do the analysis using both regression and difference-in-differences and compare the results. The memo will have four sections: 1) the problem and the policy intervention that the study substantively focuses on; 2) the methods you use to evaluate the impact of the intervention; 3) the findings of your analysis; and 4) a discussion of the importance of the evaluation. The memo must include a minimum of five cites to academic research to justify your hypothesis (you can cite blogs, websites and newspapers but these will not count toward the academic cites).

This tutorial will walk you though the steps required to conduct a policy evaluation project. In general, every project involves the following steps:

  1. Obtain policy data and familiarize yourself with the variables, how they are measured, and their availability
  2. Read the data into your statistical software and prepare it for analysis
  3. Explore the data but plotting variables of interest and calculate summary statistics
  4. Develop your statistical model and analyse your data
  5. Explain your findings

All files, including the data, can be downloaded from my GitHub site here: Policy Eval Files.

Language Options:

  • Policy Evaluation with R

    Workflow As a data analyst, it is absolutely essential that you keep detailed documentation of your analysis process. This means saving all of the code that you use for a project somewhere convenient so that you or anyone else can quickly and easily exactly reproduce your original results.

  • Policy Evaluation with Python

    Workflow As a data analyst, it is absolutely essential that you keep detailed documentation of your analysis process. This means saving all of the code that you use for a project somewhere convenient so that you or anyone else can quickly and easily exactly reproduce your original results.

Last updated on April 21, 2022