This R package contains convenience functions to extend the Unified Democracy Scores of Pemstein, Meserve, and Melton. These are basically wrappers for a couple of functions in the package mirt, which contains powerful facilities for factor analysis, and which you will probably want to use directly to do anything beyond replicating and extending the UD scores with the data provided by this package.

This package also includes an "extended UDS" dataset with latent democracy scores going back to the 19th century and a replication dataset with 61 different democracy measurements from 29 different projects to measure democracy. Both datasets are fully documented. The extended scores (and their generation) are described in more detail in Marquez, Xavier, "A Quick Method for Extending the Unified Democracy Scores" (March 23, 2016). Available at SSRN: http://ssrn.com/abstract=2753830. A more informal description of the extended UD scores is also found in this blogpost.

To install, use:

devtools::install_github(repo = "xmarquez/QuickUDS")

To view the documentation for the datasets, use:

?democracy
?extended_uds

For non-R users, there are csv versions of these datasets here. This folder also contains a pdf version of the documentation.

The extended_uds dataset can be generated with the following code:

# Generate Extended UDS
library(dplyr)
library(QuickUDS)

indexes <- c("arat_pmm","blm","bmr_democracy","bnr","bollen_pmm","doorenspleet","eiu","freedomhouse",
             "freedomhouse_electoral","gwf","hadenius_pmm","kailitz_tri","lied","mainwaring",
             "magaloni_regime_tri","munck_pmm","pacl","PEPS1v","pitf", "polity2",
             "polyarchy_contestation","prc_notrans","svolik","ulfelder","utip_dichotomous_strict",
             "v2x_polyarchy","vanhanen_democratization","wahman_teorell_hadenius")

data <- QuickUDS::democracy[ ,c("country_name","GWn","cown","year","region",
                                "continent","microstate","lat","lon","in_system",indexes)]
data <- reshape2::melt(data, measure.vars = indexes, na.rm = TRUE)
data <- data %>% group_by(country_name,year) %>% mutate(measures_per_cy = n()) %>% ungroup()
data <- reshape2::dcast(data, ... ~ variable)
data <- data %>% arrange(country_name,year)

data2 <- prepare_democracy(indexes)

# The model converges after 1831 iterations, about 2.5 mins in my not-special desktop machine
extended_model <- democracy_model(data2,indexes, verbose=FALSE, technical = list(NCYCLES = 2500)) 
extended_scores <- democracy_scores(extended_model)
extended_uds <- bind_cols(data,extended_scores)
rm(data2,data,indexes,extended_scores)

(For the latest code used, always see https://github.com/xmarquez/QuickUDS/blob/master/data-raw/Generate%20extended%20UDS%20data.R)

To generate a 0-1 index from these scores where 0.5 represents the average latent variable cutpoint for dichotomous democracy measures, the following code can be used:

cutpoints_extended <- cutpoints(extended_model)

cutpoints_extended <- cutpoints_extended %>% filter(type != "a1")

cutpoints_extended <- left_join(cutpoints_extended,
                                democracy_long %>% select(variable,index_type) %>% distinct())
#> Joining by: "variable"

dichotomous_cutpoints <- cutpoints_extended %>% filter(index_type == "Dichotomous")

dichotomous_cutpoints <- mean(dichotomous_cutpoints$estimate)

extended_uds <- extended_uds %>% mutate(adj.z1 = z1 - dichotomous_cutpoints, 
                                        adj.pct025 = pct025 - dichotomous_cutpoints, 
                                        adj.pct975 = pct975 - dichotomous_cutpoints,
                                        index = pnorm(adj.z1),
                                        index.pct025 = pnorm(adj.pct025),
                                        index.pct975 = pnorm(adj.pct975))

For a more extended introduction to the available functions in the package, see the package vignette.

The extended UD scores are available for 24111 country-years (224 unique countries and non-sovereign territories):

(Grey shaded areas represent 95% confidence intervals; blue shaded areas are periods where the country is either deemed to be a member of the system of states in the Gleditsch and Ward list of statte system membership since 1816, i.e., independent, or is a microstate in Gleditsch's tentative list).

If you find this package useful in your work, please cite:

Pemstein, Daniel, Stephen Meserve, and James Melton. 2010. Democratic Compromise: A Latent Variable Analysis of Ten Measures of Regime Type. Political Analysis 18 (4): 426-449.

As well as:

Marquez, Xavier, "A Quick Method for Extending the Unified Democracy Scores" (March 23, 2016). Available at SSRN: http://ssrn.com/abstract=2753830