Page 91 |
Save page Remove page | Previous | 91 of 160 | Next |
|
small (250x250 max)
medium (500x500 max)
Large (1000x1000 max)
Extra Large
large ( > 500x500)
Full Resolution
All (PDF)
|
This page
All
|
83
stage least squares estimation, an OLS model using the IVs in place of the original
multisectoral program variable is executed. To test that the instruments are valid, a
Wald coefficient test for relevance and a Residual OLS test for exogeneity are
preformed. Finally, Model 7 uses a Probit estimation which is similar to Ordinary
Least Squares except that the dependent variable is a latent or dummy variable taking
on the value either 0 or 1. This quantitative portion will be executed to help discern
those variables that carry the heaviest explanatory weight for the models below.
Models and Results
Model 1: Re-estimation of Over (1998)
Model 1 is a re-estimation of the model presented in Over (1998). The
original data in Over’s (1998) study could not be obtained but similar data was used
when available. One variable from Over (1998) could not be obtained: the percent of
foreign-born people in the population. Furthermore, for the majority of countries,
data on gender inequality and gender parity index are published for only one year.
The first execution of Model 1 (in column I) represents the results from Over (1998)
using available data and the second execution of Model 1 (in column II) represents
the results including a variable for multisectoral programs. The purpose of Model 1
is to re-estimate an early study that looked at the determinants of HIV/AIDS and to
see if the inclusion of the multisectoral program variable in this prior study holds any
weight. Below is an algebraic form of Over’s (1998) model:
Object Description
| Title | Political determinants and economic effects of HIV/AIDS: a push for the multisectoral approach |
| Author | Davis, Dollie |
| Author email | dollieda@usc.edu; dolliesdavis@gmail.com |
| Degree | Doctor of Philosophy |
| Document type | Dissertation |
| Degree program | Political Economy & Public Policy |
| School | College of Letters, Arts and Sciences |
| Date defended/completed | 2008-07-15 |
| Date submitted | 2008 |
| Restricted until | Unrestricted |
| Date published | 2008-10-30 |
| Advisor (committee chair) | Wise, Carol |
| Advisor (committee member) |
Nugent, Jeffrey B. Chi, Iris |
| Abstract | The proposed dissertation offers an explanation for the large differences in HIV/AIDS rates among 89 low and middle-income countries throughout the Sub Saharan African, Asian, and Latin American regions over a ten-year period (1995-2005). The HIV/AIDS rates in these countries vary widely and seemingly independently of economic wealth. One possible determinant of these differences is the presence and degree of development of strong multisectoral programs aimed at both prevention and cure of HIV/AIDS. The main hypothesis for this dissertation is: "A country's success in combating HIV/AIDS lies in the government's ability to implement an effective multisectoral program." This hypothesis is explored through quantitative models using data from the ten-year period (1995-2005). Results show that the presence of a multisectoral program over the ten-year period is associated with a significantly lower HIV/AIDS incidence rate by 2005. This effect is produced by controlling for various political, economic, societal, and institutional factors. Although there is some anecdotal evidence which suggests that multisectoral programs help to improve the HIV/AIDS problem in developing countries, there has been little if any empirical work done on this subject to date. |
| Keyword | multisectoral; HIV/AIDS; economic development |
| Geographic subject (region) | Carribbean |
| Geographic subject (continent) | Africa; Asia; South America |
| Coverage date | 1995/2005 |
| Language | English |
| Part of collection | University of Southern California dissertations and theses |
| Publisher (of the original version) | University of Southern California |
| Place of publication (of the original version) | Los Angeles, California |
| Publisher (of the digital version) | University of Southern California. Libraries |
| Provenance | Electronically uploaded by the author |
| Type | texts |
| Legacy record ID | usctheses-m1724 |
| Contributing entity | University of Southern California |
| Rights | Davis, Dollie |
| Repository name | Libraries, University of Southern California |
| Repository address | Los Angeles, California |
| Repository email | cisadmin@lib.usc.edu |
| Filename | etd-Davis-2422 |
| Archival file | uscthesesreloadpub_Volume44/etd-Davis-2422.pdf |
Description
| Title | Page 91 |
| Contributing entity | University of Southern California |
| Repository email | cisadmin@lib.usc.edu |
| Full text | 83 stage least squares estimation, an OLS model using the IVs in place of the original multisectoral program variable is executed. To test that the instruments are valid, a Wald coefficient test for relevance and a Residual OLS test for exogeneity are preformed. Finally, Model 7 uses a Probit estimation which is similar to Ordinary Least Squares except that the dependent variable is a latent or dummy variable taking on the value either 0 or 1. This quantitative portion will be executed to help discern those variables that carry the heaviest explanatory weight for the models below. Models and Results Model 1: Re-estimation of Over (1998) Model 1 is a re-estimation of the model presented in Over (1998). The original data in Over’s (1998) study could not be obtained but similar data was used when available. One variable from Over (1998) could not be obtained: the percent of foreign-born people in the population. Furthermore, for the majority of countries, data on gender inequality and gender parity index are published for only one year. The first execution of Model 1 (in column I) represents the results from Over (1998) using available data and the second execution of Model 1 (in column II) represents the results including a variable for multisectoral programs. The purpose of Model 1 is to re-estimate an early study that looked at the determinants of HIV/AIDS and to see if the inclusion of the multisectoral program variable in this prior study holds any weight. Below is an algebraic form of Over’s (1998) model: |
