I received my Ph.D. in Political Science at UCLA and then joined the faculty at Texas A&M University. From 2008-2009, I was a Fulbright Scholar in Malawi conducting dissertation research on the political economy of HIV/AIDS interventions. I have also collected data in Ghana and Tanzania. I study a variety of topics including African politics, health (particularly HIV/AIDS), ethnicity, and research methods. In particular, I have been intrigued with the importance of village headmen in rural Africa. Methodologically, I am interested in the systematic analysis of qualitative data, field experiments, social network analysis, and anything that can be learned from longitudinal household-level data.
This study examines the local realities of the global intervention against AIDS in Africa. On a continent where 68% of the 33 million people infected with HIV in the world live, and where only 43% of AIDS patients in desperate need of treatment have access to it, what facilitates the gross failures of HIV/AIDS interventions in Africa? When a mobilized international community spends billions of dollars to intervene against a disease, what impedes its efforts to reach intended beneficiaries? With globalization, disease spreads easily across national boundaries, requiring state actors concerned with public health to engage public health problems beyond their own borders, usually delegating responsibilities to far-away agents. Such endeavors require coordination of multiple actors, not just across borders, but also across levels of governance within the target country, creating multiple principal-agent problems in the pursuit of providing for public health. I use a political economy approach that emphasizes the multiplicity of actors involved in a global intervention against disease to bring attention to the incentives and motivations of actors across levels of governance. I focus in particular on the agents implementing interventions. I provide evidence that a major component of the failure of HIV/AIDS programs in Africa is the disconnect in the preferences of providers and receivers. I argue that HIV/AIDS interventions usually fail because local agents actually implementing interventions on the ground act in ways congruent with both their policy preferences and the policy preferences of local citizens. I analyze original data to demonstrate ordinary Africans give relatively low priority to HIV/AIDS interventions, even in the world's highest prevalence countries.
Even in divided societies, people from different ethnic backgrounds engage each other and an overwhelming majority of this cross-ethnic interaction does not yield conflict, but rather cooperation. Whereas the literature focuses on ethnic diversity as an impediment to collective action, this study aims to test whether social networks can overcome the challenges posed by ethnic diversity in collective action. This paper's contribution is to examine the impact of social ties vis-a-vis ethnic difference on cooperative behavior. I analyze data collected from surveys and behavioral economics experimental play of 188 rural Malawians, collected in August 2011. I find ethnic difference (or sameness) does not influence whether subjects choose to trust their game partner but being socially connected to a game partner increases the odds that a subject will cooperate.
The health system in Ghana, like in many developing countries, has undergone significant change during the past decade, including major health policy reforms related to the implementation of national health insurance, user-fee and insurance premium exemptions for pregnant women and children, and universal access to antiretroviral therapies. The purpose of this study is to investigate the health priorities of ordinary Ghanaians using a combination of qualitative and quantitative research methods. A public opinion survey will be fielded in the Greater Accra Region in March 2012.
The Role of Executive Time Horizons in State Response to AIDS in Africa, Comparative Political Studies
Abstract: In this article the author argues that politicians’ time horizons affect the differing levels of state intervention against AIDS. Using data measuring government spending, AIDS policy, and political constraints, the author tests the presumption that the leader of a country can determine a country’s level of AIDS intervention. She looks at countries in eastern and southern Africa to explore the relationship between political institutions that constrain an executive’s time horizon (i.e., competitive elections) and the level of the state’s efforts in the fight against AIDS. Her primary hypothesis is that an executive with a shorter time horizon is less likely to create policy or devote resources to intervene against AIDS. The author finds that lengthening an executive’s time horizon increases the level of government spending on health but that executives with shorter time horizons tended to have more comprehensive AIDS policy than their counterparts with longer time horizons.
Access the re-print (previous ungated version here).
Download zipped folder of replication files and analyses not shown.
An Offer You Can't Refuse? Provider-Initiated HIV Testing in Antenatal Clinics in Rural Malawi (with Nicole Angotti and Lauren Gaydosh), Health Policy and Planning.
International organizations promote provider-initiated, "routine" HIV testing of pregnant women seeking antenatal care as an effort to curb mother-to-child transmission. We offer an account of the perceptions of HIV testing at antenatal clinics in rural Malawi. Although it is both international and Government of Malawi policy that women must be explicitly informed of their right to refuse testing, analysis of in-depth interviews paired with evidence from a collection of observational field journals show that rural Malawians do not perceive HIV testing as a choice, but rather as compulsory to receive antenatal care. This study illustrates dissonance between global expectations and local realities of the delivery of HIV testing interventions.
Access the re-print or a previous, ungated version: CCPR Working Paper #033-08.
AIDS Exceptionalism: Another Constituency Heard From (with Patrick Gerland and Susan Watkins), forthcoming at AIDS and Behavior
Abstract: Amidst current debates over resources for AIDS, we examine the policy preferences of the people who are navigating AIDS in their daily lives. Survey and ethnographic data on the prioritization of HIV/AIDS interventions were collected in a longitudinal cohort study in rural Malawi. Study participants gave higher priority to problems other than AIDS. Confining analysis to HIV-positive survey respondents only weakly raises the prioritization of HIV/AIDS services. The ethnographic data are consistent with these findings: although rural Malawians are fully aware of the risk of dying from AIDS, other problems are perceived as more pressing for their community.
Access the pre-print.
Local Demand for a Global Intervention: Policy Preferences in the Time of AIDS, forthcoming at World Development
Abstract: The success of global health and development interventions ultimately depends on local reception. This paper documents local demand for HIV/AIDS interventions in Africa and seeks to explain patterns of demand using data from a country hard-hit by AIDS. As international agencies and national governments scale up HIV/AIDS interventions in Africa, I find HIV-positive respondents more highly prioritize HIV/AIDS programs, however, cross-national opinion data paired with interviews of villagers and their headmen in rural Malawi show weak prioritization of HIV/AIDS. The data illustrate a misalignment of policy preferences in the global-to-local hierarchy, highlighting the import of studying preferences of intended beneficiaries.
Access the pre-print.
The Politics of Local Research Production: Surveying in a Context of Ethnic Competition
Abstract: Even when a survey is devoid of politics, data collection is not immune from local politics in a context where groups compete for resources at the national level. Survey data relies on enumerators' abilities to elicit truthful responses, and in Africa a substantial portion of research on ethnicity relies on surveys administered by non-co-ethnic enumerators. This paper is a first look at ethnicity-of-interviewer effects on data collection and data quality in an ethnically divided African society. Using observations gathered in rural Malawi in 2010, this paper provides a thick description of a significant challenge to measurement in survey research: surveying in a context of ethnic competition. I provide an empirical account of the local reality of ethnoregional competition as demonstrated in employment (or, more likely, unemployment) of local research assistants. Analysis of quantitative data of research assistant job applicants suggests research assistants' regional background was not a significant predictor for employment. Nonetheless, qualitative data present local perceptions to the contrary and suggests spurned applicants generated study refusals. The subsequent survey data shows nuanced findings about the effects of enumerators' ethnicity: general missingness is not predicted by ethnically matching enumerators to respondents, refusal to respond to questions about sexual behavior was higher for co-ethnic enumerators, and respondents were inconsistent about reporting their ethnic identity, and this inconsistency seems related to the ethnicity of the enumerator. The findings are critical for research conducted in divided societies, especially but not exclusively in situations where scholars are studying the causes and effects of social division.
Email me for a current version.
The May 1 Marchers in Los Angeles: Overcoming Conflicting Frames, Bilingual Women Connectors, English-Language Radio, and Newly-Politicized Spanish Speakers (with Michael Suk-Young Chwe, Darin DeWitt, and Michael Stone)
Abstract: In this paper, we study protest participants in the May 2006 immigration rights marches in Los Angeles. Analysis of original survey data of 876 march participants yields five main results. First, despite substantial dispute among organizers on how to frame the marches, we find protest participants were similar across march locations organized by different coalitions. Second, we find Spanish-English bilingual participants seemed to benefit from being in two media environments, as they reported more information sources about the protest events than monolingual participants. Thirdly, women reported hearing about the protest events from more information sources, and Spanish-English bilingual women reported hearing from more information sources than any other group, suggesting they acted as social connectors behind the massive participation. Fourth, we confirm the importance of Spanish-language radio as an information source, but our data also point to the significance of television and English-language radio. Finally, analyzing data of first-time protesters, we estimate the immigrant rights marches newly politicized 125,000 people in Los Angeles who spoke Spanish and not English.
Email me for a current version.
Phone: +1.979.845.6783
Fax: +1.979.845.2511
Email: kim@polisci.tamu.edu
Updated: May 2012