• Defining and selecting indicators

Defining and selecting indicators

This short note highlights three kinds of problems associated with the governance indicators employed by the UNDP's human development report 2002.

The first problem is that of cultural and political bias. There is an implicit assumption that western multi-party political systems are the ideal by which all countries should be measured. Hence, most of the OECD countries get perfect scores in the indicators of Polity, Civil Liberties and Political Rights.

A second kind of problem relates to the UNDP's sources. The subjective indicators of governance rely for the most part on the so-called "In-house expert opinion" of two institutions: the US-based Freedom House and the World Bank's Governance Indicators Dataset. Supplementary data come from a dataset at the University of Maryland, from Transparency International and from the International Country Risk Guide.

The issue here is whether private or quasi-public institutions with limited or no accountability to a generally accepted system of international governance and whose procedures are not easily accessible to the global public, may legitimately be granted such huge powers of judgment over countries, populations and systems. The procedure itself may not meet the test of good governance.

A third kind of problem lies with the construction of the numerical indicators. The maximum and minimum scores vary widely for different indicators. And in some instances a higher number is better while in others a lower number is better.

For example, in Polity (degree of democracy) the range is from a maximum of +10 to a minimum of -10, while in Press Freedom the range is from 0 (full freedom) to 100 (not free). This is confusing and makes the indicators difficult to read and interpret.

• Full text of document
• Send to a friend