Quality of Government Datasets (QoG Institute)

Research project

The Quality of Government Datasets are compiled by researchers at the Quality of Government (QoG) Institute. The institute was established in 2004 by Sören Holmberg and Bo Rothstein at the Department of Political Science at Göteborg University, and aims at conducting and promoting research on the causes, consequences and nature of good governance or the quality of government. One of the institute’s main initiatives is the compilation of two datasets with a large number of variables that seek to measure various aspects of government. The QoG Cross-Section Dataset contains data for 2002 (or the closest year available), while the QoG Cross-Section Time-Series Dataset spans from 1946 through 2009.

Datasets

  1. QoG Cross-Section Dataset
  2. QoG Cross-Section Time-Series Dataset

Quality of Government Datasets

Website

Quality of Government Datasets (QoG Institute)

Format

Stata, SPSS, CSV

Timespan

2002, 1946-2009

Coverage

See below

Last reviewed

24/11/11

Data types and sources

The datasets contain a wide range of different types of data: expert-coded indicators and classifications, various demographic measures, national accounts data and aggregated individual-level survey data. The data are compiled from numerous freely available and well-known data sources, including datasets produced by independent research projects, international research initiatives, NGOs and IGOs.

Data download

Quality of Government Datasets

Topics

The QoG Cross-Section Dataset and the QoG Cross-Section Time-Series Dataset (version 6 April 2011) are divided into three data types or topics:

  1. WII (What It Is): variables related to the core of the institute’s research area (e.g., corruption, bureaucratic quality, human rights and democracy).
  2. HTG (How To Get It): variables believed to promote the quality of government (e.g., electoral rules, political institutions, legal and colonial origin, religion and social fractionalisation).
  3. WYG (What You Get): variables pertaining to some of the possible consequences of government quality (e.g., economic and human development, international and domestic peace, environmental sustainability, gender equality, and satisfied and trusting citizens).

Geographical coverage

The datasets cover mainly the same countries, but the time-series version contains some additional cases. The coverage of specific variables is often more limited:

  • Cross-Section Dataset: All countries in the world recognised by the United Nations in 2002, plus Taiwan (192 countries in all).
  • Cross-Section Time-Series Dataset: All countries in the world recognised by the United Nations in 2002, plus Taiwan and 13 historical countries that no longer exist.

Time coverage and updates

  • Cross-Section Dataset: Data for 2002 or the closest year available
  • Cross-Section Time-Series Dataset: 1946-2009

The datasets will be updated continuously. Last update: April 2011.

Documentation

The datasets are described in an extensive codebook, which is freely available for download. The codebook is sufficiently detailed for users who wish to determine the datasets’ potential, but in most cases it is necessary to consult the original sources to get detailed knowledge on primary sources and coding practices. See Ledet (2006) (APSA-CP Newsletter) for an independent review of an earlier version of the datasets.

Access conditions and cost

Available free of charge.

Access procedures

Predefined table. The datasets can be downloaded in two separate files.

Data formats

Stata, SPSS and CSV

Comparability and data quality

Most of the data in Quality of Government Datasets are widely used in academic research. This can be seen as an indication that the data quality is generally satisfactory, though extensive usage does not guarantee validity and reliability. A major strength of the datasets is that they often provide several measures of the same concept, allowing users to choose between several indicators and to test whether different ways of measuring a phenomenon has any effect on the analyses. See Chapter 5 in Rydland et al. (2008) for a more detailed discussion of the comparability of political indicators. Also, see Ledet (2006) (APSA-CP Newsletter) for an independent review of an earlier version of the datasets.

Literature

Ledet, Richard. 2006. “Dataset review: the Quality of Government Institute’s Cross-Sectional and Cross-Sectional Time Series Dataset”. APSA-CP Newsletter 17 (Summer).

Rydland, Lars Tore, Sveinung Arnesen and Åse Gilje Østensen. 2008. Contextual data for the European Social Survey. An Overview and assessment of extant resources. NSD Report No.124, Norwegian Social Science Data Services (NSD).