COW Direct Contiguity, 1816-2006 (v3.1)

Overview

Version 3.1 of the Correlates of War Direct Contiguity data identifies all direct contiguity relationships between states in the international system from 1816 through 2006. The classification system for contiguous dyads is comprised of five categories, one for land contiguity and four for water contiguity. Land contiguity is defined as the intersection of the homeland territory of the two states in the dyad, either through a land boundary or a river (such as the Rio Grande in the case of the US-Mexico border). Water contiguity is divided into four categories, based on a seperation by water of 12, 24, 150, and 400 miles.


Citation

In any papers or publications that utilize this data set, users are asked to give the version number and cite the article of record for the data set, as follows:

Correlates of War Project. Direct Contiguity Data, 1816-2006. Version 3.1. Online: http://correlatesofwar.org.

Users are asked to cite the current article of record for the data set, as follows:

Bibliography Entry: Stinnett, Douglas M., Jaroslav Tir, Philip Schafer, Paul F. Diehl, and Charles Gochman. 2002. "The Correlates of War Project Direct Contiguity Data, Version 3." Conflict Mangagement and Peace Science 19(2):58-66.

Additional details of the basic coding process were elaborated in

Charles S. Gochman, 1991, "Interstate Metrics: Conceptualizing, Operationalizing, and Measuring the Geographic Proximity of States since the Congress of Vienna," International Interactions 17(1): 93-112. 


Data Set

All files pertaining to this data set are to be found in the file " DirectContiguity310.zip." This data set may be downloaded by clicking here. This downloadable data archive includes a text file with full codebook, as well as three data files that are provided in .csv comma-delimited format, which can easily be read into any spreadsheet or statistical software. The three data files are:

contdir.csv: The master data file (with one entry per contiguity relationship)
contdird.csv: A directed dyad-year-level version of the master data
contdirs.csv: A state-year-level data set calculated from the master data

Direct Contiguity Master File: ContDir

This file contains the master records of version 3.0. It is the one from which all other files are ultimately derived and to which all future changes will be made. It has a non-directed format in order to prevent asymmetries from accidentally being introduced into the data set. It is written in comma-separated-variable (CSV) format, and the variable names are given in the first record in the file. The available variables are as follows.

• Dyad:  A variable that combines the two states' COW country codes to facilitate merging with other dyadic data.  This takes the format AAABBB, where AAA is the lower value of the two country codes and BBB is the higher value.
• StateLNo:  COW state number ("country code") of lower numbered member of this dyad
• StateLAb: COW state abbreviation of lower numbered member of this dyad
• StateHNo: COW state number ("country code") of higher numbered member of this dyad
• StateHAb: COW state abbreviation of higher numbered member of this dyad
• ContType:  Type of contiguity relationship
            1: Separated by a land or river border
            2: Separated by 12 miles of water or less
            3: Separated by 24 miles of water or less (but more than 12 miles)
            4: Separated by 150 miles of water or less (but more than 24 miles)
            5: Separated by 400 miles of water or less (but more than 150 miles)
• Begin: Year and month when this contiguity relationship began (YYYYMM)
• End: Year and month when this contiguity relationship ended (YYYYMM)
• Notes:  Additional information about the contiguity relationship such as the body of water that is crossed (for sea contiguity), the portion of the state's territory from which the distance was calculated, and the reasons for the beginning and (where relevant) ending of the relationship.
• Version:  Data set version number (3.1 for all records)

Dyadic Direct Contiguity: ContDirD

This file is derived from the master file and includes annual records for each (directional) pair of states that is directly contiguous. Dyad-years in which direct contiguity is absent are not included in the file to reduce greatly the size of the file. Users may assume that if no record is present for a given dyad-year, no direct contiguity exists between the state pair in that year.  It is written in comma-separated-variable (CSV) format, and the variable names are given in the first record in the file. The format of the remaining records is as follows.

• Dyad:  A variable that combines the two states' COW country codes in a way that facilitates merging with other dyadic data.  This takes the format AAABBB, where AAA is the lower value of the two country codes and BBB is the higher value.
• State1No:  COW state number ("country code") of first member of this dyad
• State1Ab: COW state abbreviation of first member of this dyad
• State2No: COW state number ("country code") of second member of this dyad
• State2Ab: COW state abbreviation of second member of this dyad
• Year:  Year of this observation
• ContType:  Type of contiguity relationship
            1: Separated by a land or river border
            2: Separated by 12 miles of water or less
            3: Separated by 24 miles of water or less (but more than 12 miles)
            4: Separated by 150 miles of water or less (but more than 24 miles)
            5: Separated by 400 miles of water or less (but more than 150 miles)
• Version:  Data set version number (3.1 for all records)

State-Level Direct Contiguity: ContDirS

This file is derived from the master file and includes annual records for all states in the COW interstate system. The variable names are given in the first record in the file. The format of the remaining records is as follows:

• StateNo:  COW state number ("country code") of state
• StateAb: COW state abbreviation of state
• Year:  Year of this observation
• Total:  Total number of direct contiguities for this state this year (land or sea)
• Land:  Total number of direct contiguities by land for this state this year (i.e., ContType = 1)
• Sea:  Total number of direct contiguities by sea for this state this year (i.e., ContType > 1)
• Version:  Data set version number (3.1 for all records)

 


Data Generation Procedures

Excerpts from the article of record:

The classification system for contiguity is comprised of five categories, one for land contiguity and four for water contiguity. Land contiguity is defined as the intersection of the homeland territory of the two states in the dyad, either through a land boundary or a river (such as the Rio Grande along the US-Mexico border). Water contiguity is based on whether a straight line of no more than a certain distance can be drawn between a point on the border of one state, across open water (uninterrupted by the territory of a third state), to the closest point on the homeland territory of another state. Four different levels of water contiguity are recorded, based on the distance between the two states' territories: up to 12 miles (reflecting the widely recognized 12-mile limit for territorial waters), 24 miles (reflecting the maximum distance at which two states' 12-mile territorial limits can intersect), 150 miles (from the original 1816-1965 version of the data set, reflecting what was considered the average distance that a sailing ship could travel in one day), and 400 miles (the maximum distance at which two 200-mile exclusive economic zones can intersect).

The coding for each dyad is always the closest form of contiguity for that year. Thus, if a dyad shares a land border and is also separated by a stretch of water less than 400 miles, that dyad will be coded based on the closest form of contiguity (the land border). If the type of contiguity for a dyad changes during a given year, only the closest classification is recorded for that year, in order to prevent multiple records for one dyad-year from appearing in the data set.

A variety of sources have been used in the process of collecting these data. The sources used for earlier versions of the data are identified in Gochman (1991); more recent updates have relied on various editions of the Times Atlas of the World, National Geographic Atlas of the World, Hammond World Atlas, and Oxford Atlas of the World.

Full details on changes made in the data for v3.1 are included in the codebook that is downloaded with the data files.

 


Questions and Feedback

The contiguity data sets are hosted by Paul Hensel, University of North Texas, under the COW Data Set Hosting Program. In case of questions or concerns concerning the data or coding rules, he may be contacted by email at phensel@unt.edu.