The latest round of the Worldwide Governance Indicators (WGI) were released today under the auspices of… we’re not entirely sure. Ownership issues aside, the WGI remain the most widely used governance data in the world.
Previously an effort of the World Bank Institute under Dani Kaufman’s leadership, “The WGI are [now] produced by: Daniel Kaufmann, Brookings Institution, Aart Kraay, World Bank Development Economics Research Group, [and] Massimo Mastruzzi, World Bank Institute.” Except that, “The WGI do not reflect the official views of the World Bank, its Executive Directors, or the countries they represent. The WGI are not used by the World Bank Group to allocate resources.”
A primary feature of the WGI is their near global coverage, thanks to the use of dozens of disparate surveys and expert assessments as component data, including Global Integrity data. We’ve discussed the use and potential misuse of the WGI at some length in our book, A Users’ Guide to Measuring Corruption.
And once again, the Worldwide Governance Indicators can be found here.
I wanted to take the opportunity to thank the WGI authors (especially Aart and Massimo) for making two adjustments to the use of Global Integrity data in the WGI at our request. To quote their documentation:
“At the request of Global Integrity, we have dropped the 2003 round [note: we call this our “2004 data”] of the Global Integrity Index from our indicators for 2003, 2004, and 2005, as there were changes in Global Integrity’s methodology between the first and subsequent rounds of this exercise, making the first round not fully comparable with subsequent rounds.”
This dovetails with the warnings we have made explicit for the past two years about compatibility issues between our 2004 data and newer data.
“Note that in 2007 & 2008 we have carried forward [Global Integrity] scores for those countries that were covered in previous years but not in the current year.”
We had asked that the authors document their practice of “flatlining” data (my term, not theirs) for years in which data sources (such as Global Integrity) did not carry out fieldwork but for which the WGI report the existence of data. Without disclosure of this, we felt the practice was easily misinterpreted to mean the authors had new source material when none existed.
Not sure whether you should care about these changes? Check out our A Users’ Guide to Measuring Corruption to beef up your methodological know-how.
— Nathaniel Heller