This repository provides data, replication code, and a latex manuscript to replicate "The Lessons Private Schools Teach: Using a Field Experiment to Understand the Effects of Private Services on Political Behavior" accepted at Comparative Political Studies.
For any questions regarding the replication code and data, please contact Emmerich Davies
Data can be accessed from CPS's dataverse or the author's personal github page
To replicate the R scripts, the easiest way would likely be to open the cps_downstream.Rproj file in RStudio and the files should be easy to run from there.
For stata, you will have to change the absolute paths at the beginning of each stata do-file.
Please email me if gaining access to non-public data is not straightforward. I should be able to point to locations where this is available with requests now if the information below is out of date.
- For the data to reproduce the tables in
data/scripts/08_analysis_academics.do, please contact Karthik Muralidharan. Data is taken from "The Aggregate Effect of School Choice: Evidence from a Two-Stage Experiment in India" published in The Quarterly Journal of Economics - For the data to reproduce the table in
data/scripts/04_analysis_downstreamexperiment.do/sample_comparison.csv, you will need access to files that are not mine to share:- The census rural directory that matches the 2001 and 2011 Indian Census. This was initially downloaded from Sumit Mishra's website, but is unfortunately no longer available there. Fortunately, this has now been replaced by the Indian Government's Local Government Directory.
- The raw census data. Again, this is publicly available from the Indian Government's Local Government Directory.
- Likewise, to reproduce
data/output/figures/mapdistricts.pdf, you will need access to Indian state and district level shapefiles.- For state-level shapefiles, those are available at Datameet's repository of state-level shapefiles
- For district-level shapefiles, those are available at Datameet's repository of district-level shapefiles
- All other data is available in
data/raw
All R packages should be publicly available through CRAN. A list of packages required to run the code here is provided at the top of every script and in data/scripts/01_r_setup.R. You can also run data/scripts/01_r_setup.R that will check if your computer has the required packages, download, install and load them if not (this is very aggressive), and create the requried file paths. NB: If you clone this repository from github, the file path should automatically be created.
The code requires two stata packages:
center. You should be able to install it either by runningdata/scripts/02_stata_setup.door typingssc install centerin your Stata command line.estout. You should be able to install it either by runningdata/scripts/02_stata_setup.door typingssc install estoutin your Stata command line.make_index. You will need to manually download this.
Analysis was conducted using Stata V16.1 and R V4.0.2 (2020-06-22)