Skip to content

Replication materials for "Democracy and the Class Struggle"


Notifications You must be signed in to change notification settings


Folders and files

Last commit message
Last commit date

Latest commit



26 Commits

Repository files navigation

Democracy and the Class Struggle

This repository contains replication materials for "Democracy and the Class Struggle," American Journal of Sociology 124, no. 3 (November 2018): 664-704 (

What do you need?

To replicate my work, you will need:

  1. R, a popular and free statistical software. I wrote these files in R 3.4.1, but it should be forwards-compatible (i.e., you should be able to run these if you download the current version). John Fox offers a useful intro, if you don't have prior experience.
  2. Several additional packages, which don't come installed with R. If you run my scripts without installing these, R will throw an error noting that you haven't yet installed a required package. To make life easier, I have included a script 'installpacks.R'. This generates a list of all the packages I run and, if you remove the comments before the last line, installs them.
  3. Ghostscript, which I use for fancy fonts in graphs. For detailed instructions on how to install Ghostscript, see this blogpost. Once you do this, open the file 'dirs.R' and make sure that 'gsdir_full' points to the directory that contains the Ghostscript executable. If you install GS in your 'Program Files' directory, you shouldn't need to change anything. (If you are happy without fancy fonts, edit the last lines in the '..output.R' files so that embed_fonts() does not run).

What is in here?

  • appendix: Contains the supplementary appendix
  • code: Contains all the R scripts that read data, run the models and generate output
  • data: Contains the starting dataset and country code information
  • meta: Contains .csv's that some of the scripts consult, when running

How to use it?

To run this code successfully, please follow the instructions below:

  1. Before you run the scripts, load the demclass.Rproj file. This ensures that the default working directory is the "code" subfolder.
  2. Run the code sequentially (in order of filename). If you run a later file without having run an earlier file, the script will probably throw an error.
  3. Figures are output as .pdf's and tables as .csv's, but if you'd like standalone .tex files (or .tex) fragments with the explanatory foototes, run 99_figures.R and 99_tables.R. 99_figures.R will also create a folder called 'upload' in the 'output' folder, to which it will copy all figures that I use in the paper and supplementary appendix

Below I describe briefly describe what each file does, and which of the main figures and tables it generates. Note that I don't explain which files must be run to generate the material in the supplementary appendix. This can be done by running all scripts sequentially or by running '00_runall.R', which will run everything from start to finish (in 1-2 hours, depending on the speed of your machine).

  • 00_replicate.R: Runs the main democratization models (annual dataset, preferred specification) and extracts long-run estimates. If all you want is to reproduce the main regression results and/or play around with the dataset, focus on this file.
  • 10_mainmods.R: Runs the democratization models reported in the main paper.
  • 11_mainmods_graphs.R: Generates Figure 2
  • 12_mainmods_tables.R: Generates Tables 3 and 4
  • 14_robmods_main.R: Runs some robustness checks
  • 15EX_robmods_lagdv.R: Runs simulation showing importance of including a lagged dependent variable
  • 16_robmods_bin.R: Runs further robustness checks (modeling transitions/consolidation rather than democratization)
  • 18_robmods_output.R: Outputs results from robustness checks
  • 20_dcapmods.R: Runs the unionization and strike models
  • 21_dcapmods_output.R: Generates Figure 1
  • 30_predictions.R: Runs a variety of counterfactuals
  • 31_predictions_stats.R: Computes summary statistics across counterfatuals
  • 32_preditions_output.R: Generates Figures 3 and 4
  • 40_modelfit.R: Re-runs models on a consistent sample and generates sequential predictions
  • 41_modelfit_stats.R: Computes fit statistics
  • 42_modelfit_output.R: Outputs results from model fit exercise
  • 50_descriptive_output.R: Generates Tables 1 and 2
  • 51_descriptive_graphs.R: Generates descriptive figures
  • 60_unitroots.R: Examines key series for unit root
  • 61_unitroots_output.R: Outputs results from unit root tests
  • 99_figures.R: Creates .tex files and fragments for figures, with footnotes.
  • 99_tables.R: Creates .tex files and fragments for tables, with footnotes

NB: Scripts that are not numbered contain functions or other code that is called by one or more of these scripts.

Questions and Comments?

If you encounter any issues with this code or with reproducing my results, please let me know at adaner.usmani[at]


Replication materials for "Democracy and the Class Struggle"







No releases published