http://blogs.law.harvard.edu/politicshiv/2006/05/08/aids-program-effort-index-api/ Just what it says Sun, 29 Nov 2009 22:25:26 -0500 http://wordpress.org/?v=2.8.4 hourly 1 http://blogs.law.harvard.edu/politicshiv/2006/05/08/aids-program-effort-index-api/comment-page-1/#comment-106 Politics and Policy of HIV/AIDS » Amy Patterson’s new book The Politics of AIDS in Africa Tue, 16 Jan 2007 21:45:48 +0000 http://blogs.law.harvard.edu/politicshiv/2006/05/08/aids-program-effort-index-api/#comment-106 [...] I did want to discuss one thing here though. To compare the countries in her case studies, Patterson uses the Aids Programme Effort Index (API). The API is a survey instrument that is intended to reflect the country’s level of effort on AIDS on a number of criteria such as political support and legal/regulatory environment. A handful of experts knowledgeable about the different criteria were consulted in each country, generating a score on a 100-point scale for each criteria and overall score. South Africa, for example, scored a 75 on the 2003 API, similar to Uganda (76) and less than Botswana (80) and Rwanda (81). Swaziland and Zimbabwe got a 60 and 61. In reading up on the API, which has gone through three iterations since 1999, I found that the average number of experts surveyed for each of the criteria was four people. The result is that you get something sixteen people consulted per country. Somehow, I wonder if the scores that are generated have any cross-country generalizability. I have very strong doubts. I know that data limitations make it very hard to do this kind of work, but when Patterson ultimately concludes that there is no overarching explanation for differences in country behavior on AIDS policies, I wonder if she’d reach the same conclusions if we had a had better measure of country effort on AIDS. [...] [...] I did want to discuss one thing here though. To compare the countries in her case studies, Patterson uses the Aids Programme Effort Index (API). The API is a survey instrument that is intended to reflect the country’s level of effort on AIDS on a number of criteria such as political support and legal/regulatory environment. A handful of experts knowledgeable about the different criteria were consulted in each country, generating a score on a 100-point scale for each criteria and overall score. South Africa, for example, scored a 75 on the 2003 API, similar to Uganda (76) and less than Botswana (80) and Rwanda (81). Swaziland and Zimbabwe got a 60 and 61. In reading up on the API, which has gone through three iterations since 1999, I found that the average number of experts surveyed for each of the criteria was four people. The result is that you get something sixteen people consulted per country. Somehow, I wonder if the scores that are generated have any cross-country generalizability. I have very strong doubts. I know that data limitations make it very hard to do this kind of work, but when Patterson ultimately concludes that there is no overarching explanation for differences in country behavior on AIDS policies, I wonder if she’d reach the same conclusions if we had a had better measure of country effort on AIDS. [...]

]]>