Will H. Moore's Polity Data Page

The Polity IV page is the project's current home.

The page below the break contains some information about the Polity project, but is out of date (it was relevant circa 2003). I leave it here for whatever value it might have.

The polity project collects data about the institutional characteristics of states throughout the world from 1800 through the 1990s. The data are often used to measure democracy. This page assumes the reader has basic familiarity with the project. If you are not familiar with it, please see the following for overviews of its components:

Kristian Gleditsch's Polity Data Archive is probably the most useful WWW site for the Polity data.

This page has some information (and links) for the following:

Polity II

The Polity II data set is available through the Inter-university Consortium for Political and Social Research. An abridged version of the Polity II Codebook is available on line in the following formats:

Polity III

The Polity III data set collects a limited set of the Polity II variables for all countries in the world from 1945-1993 (including correction of errors during the post World War II era contained in Polity II). The authoritative description of Polity III is:
Keith Jaggers and Ted Robert Gurr. 1995. "Tracking Democracy's Third Wave with the Polity III Data," Journal of Peace Research, 32:469-482.
The Polity III data were deposited with the ICPSR, and are are available on-line at the Polity Data Archive (from the University of Colorado's Globalization and Democracy Program).

Please Note: The Polity III data were updated through 1998 in a new version of the data called Polity98, and then another update was done through 2000, which is called Polity IV.

Mike Ward and Kristian Gledistch have corrected a number of errors in the original release of the Polity II data. Here is a description of the corrections:

The May 1996 Version of the POLITY III Data


Using the criteria procedures as laid out in the POLITY II codebook (Gurr et al. 1989) for computing the institutionalized democracy and autocracy scores from the authority characteristics, it was found that several entries in the current version of the POLITY III data appear to have been miscoded. The new recoded data were generated by an S program implementing the coding procedures. The program and the corrections were reviewed with the principal POLITY III coder, Keith Jaggers, who verified the procedures implemented and the accuracy of the corrections. Here is a list of the major differences.

  1. The most dramatic difference in terms of magnitude is that the democracy and autocracy scores were transposed for Russia in 1989. The democracy score should be 1 and the autocracy score should be 5 for this year, but were reversed in the old version of the data. These scores have been reorganized in the May 1996 version of the POLITY III data.
  2. 1484 cases out of the total of 13907 entries in the POLITY III data have been assigned a democracy score that differed by -1 from what the correct values should be according to the coding rules. Of these, 1170 are in the nineteenth century. All the entries in the twentieth century are prior to 1945. The countries that have been coded with incorrect democracy scores include Afghanistan (1894-1934), Austria (1848-1917), Baden (1841-47), Belgium (1932-51), Brazil (1824-88), Bulgaria (1883-85), China (1800-1910), Denmark (1849-1900), Egypt (1923-45), France (1814-47), Germany/Prussia (1878-1917), Hungary (1867-1917), Iran (1941-45), Iraq (1924-45), Italy (1861-21), Japan (1868-1944), Korea (1880-1910), Luxembourg (1867-89), Morocco (1880-1911), Nepal (1800-45), Norway (1873-97), Netherlands (1815-1916), Portugal (1834-1909), Rumania (1910-39), Russia (1906-16), Spain (1820-72), Sweden (1855-1906), Turkey (1876, 1908-18), United Kingdom (1800-36), Wurttemburg (1819-71), Yugoslavia (1921-28, 193940). These are corrected in the May 1996 version of the POLITY III data.
  3. 35 cases are coded incorrectly for the autocracy scores. Mexico has been assigned an incorrect autocracy of 4 instead of 2 for 1822-33; Guatemala has been misclassified between with 4 instead of 5 in 1963-65; Panama is assigned 7 and should have had 8 for 1988-89; Argentina which should have had 6 instead of 7 1825-28; and Lesotho which should have had 7 rather than eight 1973-85. These are corrected in the May 1996 version of the POLITY III data.
  4. There were no entries for Germany/Prussia for the years 1868-70. These have been added and assigned as "in transition" (i.e., -88). There was no entry for Orange Free State for 1858. The entries for 1857 and 1859 are indentical, and 1858 was subsequently assigned these values. The entry for Egypt in 1922 was missing. A new entry was added, assigned as "interregnum" (i.e., -77), given the coding for the preceeding years. There were two entries for Korea in 1876; one was deleted.

REFERENCES:
Gurr, Ted R.; Keith Jaggers and Will H. Moore. 1989. The POLITY II Codebook. University of Colorado at Boulder, Center for Comparative Politics.

Kristian S. Gleditsch
Michael D. Ward
Keith Jaggers
Ted R. Gurr

Polity 3d

The Polity 3d project was established to provide as precise a date as possible for the transition from one polity type to another. This will facilitate a variety of analyses that could not be done well with the original Polity data, especially the analysis of the duration of regimes. The authoritative account of 3d is:
Sara McLaughlin, Scott Gates, Havard Hegre, Ranveig Gissinger, and Nils Petter Gleditsch. 1999. "Polity 3D; The timing of polity changes" Journal of Conflict Resolution 42(2):231-242.
An electronic copy of the paper is available at Kristian Gleditsch's Polity Data Archive.

Polity IV

The Polity IV project supersedes Polity 98 which extended Polity III through 1998. It goes to the year 2000 and contains corrections to the Polity 98 data (which contained corrections to Polity III).

Please Note: While I was involved in the Polity II data effort, I was not a co-PI on Polity III, Polity 3d, Polity98, or Polity IV. If you would like to contact Keith Jaggers, he can be reached by email at kcjaggers@aol.com. Sara McLaughlin Mitchell can be reached at smclaugh@garnet.acns.fsu.edu. Monty Marshall can be reached at mmarshall@cidcm.umd.edu.
Last Updated on 2 Oct 2009 by Will H. Moore

I can be reached by email at:
will dot moore at fsu dot edu
or by phone at 850.644.6924.