Working Projects
A Field Experiment on Discrimination against Immigrants in the U.S. Health Care Market. With Danbee Lee. [Working Paper].
Government can hardly achieve the fairness of public service provision if street-level bureaucrats use their discretions to treat clients systematically differently. As a part of service receivers, noncitizen immigrants are growingly important in political and economic issues in the United States. However, although previous evidence has well-documented street-level discrimination against certain groups of people, whether and how noncitizen immigrants are treated equally in the health care market remains an unanswered question. This study proposes an audit field experiment to identify the causal relation between noncitizen immigration status and potential discrimination on the frontline of U.S. nursing homes. Evidence from this field experiment will provide the first national wide evidence on discrimination against noncitizen immigrants in the U.S. and how public, nonprofit, and for-profit nursing homes treat noncitizen clients differently. Findings will also reveal the theoretical mechanism of street-level bureaucrats’ discriminating behavior.
How Does Government Decision Makers Understand Evidence? A Conjoint Experiment. With Yuan Cheng, Shuping Wang, Western Merrick, and Patrick Carter. [Working Paper].
Evidence-based policy/decision making is now institutionalized in almost all state governments in the U.S. Such initiative requires/encourages policy/decision makers to incorporate research and analyasis of programs into policy and funding decisions in key policy areas such as behavioral health and criminal justice. Nevertheless, it is still unclear the extent to which government decision makers prioritize programs that maximize the use of evidence-based practices. Given the complexity of scientific evidence and absense of data, it is also unclear which aspect of the evidence drive their decision making process. In this study, we work with government officials to investigate aforementioned questions. The results will provide government agencies handful information to improve EBPs and offer scholars guidance for more effective research communication.
Using Generative Pre-Trained Transformers (GPT) for Supervised Content Encoding: An Application in Corresponding Experiments. With Alexander Churchill and Shamitha Pichika. [Working Paper].
Supervised content encoding applies a given codebook to a larger non-numerical dataset and is central to empirical research in policy studies. Not only is it a key analytical approach for qualitative research, but the method also allows researchers to measure constructs using non-numerical data, which can then be applied to quantitative description and causal inference. Despite its utility, supervised content encoding faces challenges including high cost and low reproducibility. In this report, we test if large language models (LLM), specifically generative pre-trained transformers (GPT), can solve these problems. Using email messages collected from a national corresponding experiment in the U.S. nursing home market as an example, we demonstrate that although we found some disparities between GPT and human coding results, the disagreement is acceptable for certain research design, which makes GPT encoding a potential substitute for human encoders. Practical suggestions for encoding with GPT are provided at the end of the letter.
Resource Publicness and the Influence of Commercialization in Nonprofit Organizations. With ChiaKo Hung and Huafang Li.
Pursuing commercial revenue is widely considered a risk venture for most nonprofit leaders, particularly because of their concerns about its crowding out effect on donations. In this study, we argue that such a crowd-out effect is not universal. We theorize the moderating role of resource publicness and test our hypotheses by both administrative data and online experiments. More recent research suggests that resource publicness (defined by major revenue including government funding, donation, and commercial income) may shape the organizational identity perceived by the public. We expect it may cost organizations more when they are pursuing revenues with lower publicness. Specifically, commercial revenues may crowd out more donations in donative and government-funded nonprofits than in commercial nonprofits because seeking commercial revenue is perceived to follow a profit-seeking logic that violates people’s perceived publicness of the nonprofits. Commercial nonprofits will suffer less negative influence from generating more commercial revenue as it is aligned with its identity and people’s expectations.