gnevar database

The gnevar database consists of several related datasets relating to the global governance of different issue areas. The focus is on high-quality relational event data for tackling important subjects such as military alliances, trade, fisheries, and freshwater.

gnevar brings several prominent datasets on international agreements together with the goals of:

  • developing as comprehensive datasets as possible by filling in missing data through new manual and automatic coding, including:
    • agreement texts, translated to English, wherever possible
  • establishing as consistent data as possible through continuous, computational internal validity checks:
    • exported functions for further data wrangling
  • facilitating as accurate data as possible through additional archival research with an eye toward temporal precision, including:
    • identifying terminated agreements, and the grounds for that termination
  • enabling comparison across issue areas through variable matching and standardisation, including:
    • document types (LJD codings)

Learn more about the package here.

Package Info

The R package gnevar was created and is maintained by James Hollway (Graduate Institute of International and Development Studies) and a number of contributing students and research assistants.

The data collection effort began in 2012 with a focus on international fisheries agreements, but has broadened considerably since to international water, trade, and military alliance agreements, as well as additional temporal and spatial information on state actors.

While the main effort here has been to catalogue formal, international actors and institutions, gnevar also serves research aiming to go beyond this. For example, transnational actors and institutions should still be studied in relation to and in comparison to these formal networks.

Installing the package

The gnevar R package can be installed from the Bitbucket repository using the devtools package:1

devtools::install_bitbucket("jhollway/gnevar", auth_user = "name", password = "password")

Please email me if you are interested in user access to the repository.

  1. You will currently need to build the package yourself.↩