The Worldwide Governance Indicators (WGI) are a long‐standing research project to develop cross‐country indicators of governance. The WGI consist of six composite indicators of broad dimensions of governance covering over 200 countries since 1996: Voice and Accountability, Political Stability and Absence of Violence/Terrorism, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption. These indicators are based on several hundred variables obtained from 31 different data sources, capturing governance perceptions as reported by survey respondents, non‐governmental organizations, commercial business information providers, and public sector organizations worldwide.
This paper summarizes the methodology and key analytical issues relevant to the overall WGI project. The updated data for the six indicators, together with the underlying source data and the details of the 2010 update of the WGI, are not discussed in this paper but are available online at www.govindicators.org. We also plan to release and document subsequent updates of the WGI purely online, with this paper serving as a guide to the overall methodological issues relevant to the WGI project and future updates.
In the WGI we draw together data on perceptions of governance from a wide variety of sources, and organize them into six clusters corresponding to the six broad dimensions of governance listed above. For each of these clusters we then use a statistical methodology known as an Unobserved Components Model to (i) standardize the data from these very diverse sources into comparable units, (ii) construct an aggregate indicator of governance as a weighted average of the underlying source variables, and (iii) construct margins of error that reflect the unavoidable imprecision in measuring governance.
We believe this to be a useful way of organizing and summarizing the very large and disparate set of individual perceptions‐based indicators of governance that have become available since the late 1990s when we began this project. Moreover, by constructing and reporting explicit margins of error for the aggregate indicators, we enable users to avoid over‐interpreting small differences between countries and over time in the indicators that are unlikely to be statistically—or practically—significant.
This emphasis on explicit reporting of uncertainty about estimates of governance has been notably lacking in most other governance datasets.
While the six aggregate WGI measures are a useful summary of the underlying source data, we recognize that for many purposes, the individual underlying data sources are also of interest for users of the WGI data. Many of these indicators provide highly specific and disaggregated information about particular dimensions of governance that are of great independent interest. For this reason we make the underlying source data available together with the six aggregate indicators through the WGI website.
The rest of this paper is organized as follows. In the next section we discuss the definition of governance that motivates the six broad indicators that we construct. Section 3 describes the source data on governance perceptions on which the WGI project is based. Section 4 provides details on the statistical methodology used to construct the aggregate indicators, and Section 5 offers a guide to interpreting the data. Section 6 contains a review of some of the main analytic issues in the construction and use of the WGI, and Section 7 concludes.
The whole spirit of multilateralism is on life support. Normally you’d want to heap praise on some other country for taking on a larger share of this global burden, but Trump doesn’t think about global problems needing to be globally shared.