This page contains the policy domain files with the data and metadata. All policies are coded as of December 31 of the year indicated (date of enactment), except statistical variables that cover entire years, such as arrest and incarceration data and fiscal policies. Excel may give you a warning, but these files are all safe to edit.
As of June 2019, we now are going tidyverse-friendly and offering csv versions of the data and metadata in separate files. The Excel files offer some additional features such as comments in cells explaining codings and automatically updating formulas. Sample R code for importing and joining datasets is available here.
Please cite the following article when using the data in research:
- Fiscal policies CSV: data | metadata
- Firearms CSV: data | metadata TXT: explanation of carry indices construction
- Alcohol & drugs CSV: data | metadata
- Mala prohibita & miscellaneous social issues {plus affirmative action ban variable} CSV (contain affirmative action ban variable): data | metadata
- Education CSV: data | metadata
- Land use & environmental CSV: data | metadata
- Labor CSV: data | metadata
- Health insurance CSV: data | metadata
- Tobacco CSV: data | metadata
- Utilities CSV: data | metadata
- Occupational licensing CSV: data | metadata | individual licenses | kloccs2 TXT: metadata for individual licenses
- Asset forfeiture rules CSV: data | metadata
- Miscellaneous regulation CSV: data | metadata
- Courts & tort reform CSV: data | metadata
- Abortion CSV: data | metadata
- Incarceration, arrests, & death penalty CSV: data | metadata
- Marriage CSV: data | metadata
- Campaign finance CSV: data | metadata
To follow the procedures from the original article for deriving policy liberalism and civil libertarianism indices, simply conduct principal component analysis on the available top-level policy variables for each year, available in this spreadsheet (CSV). However, we have also derived our own policy liberalism and civil libertarianism indices for researchers to use. The procedure for deriving these indices is as follows. First, when data are available for year t and year t+a on a policy variable but unavailable for years t+b, where a>b>0, we linearly interpolate the missing values. Second, we conduct multiple imputation on the policy variables using the “amelia” package in R. Finally, we conduct principal component analysis on each imputed dataset and then average across imputed datasets to derive the policy liberalism and civil libertarianism indices for each year.