Sapat, Alka

Policy formation involves an interplay of decision-making processes that shape all policy process stages. A critical aspect of the design process is selecting policy tools to align with policy goals. The First Step Act (FSA) (2018) aims to reduce recidivism and reform sentencing through a robust policy tool portfolio. However, questions persist about how policy tools are selected. This dissertation evaluates the policy design, tool choices, and collaborative networks associated with these choices. Grounded in the social construction framework and using content analysis and discourse network analysis (DNA), this dissertation examines how policymakers select policy tools to achieve functionality.
Research question one explores the complementarity of the FSA’s policy tool portfolio, and results demonstrate alignment between policy goals and corresponding policy tools. Research question two examines how social constructions of target populations and political affiliations influence policy tool choices; hypothesis one reveals that liberal policymakers align with those with similar political affiliations. Results from hypothesis two indicate liberal policymakers adopt a rehabilitation orientation tool approach, viewing target populations as investments and deserving of support structures. The findings also highlight limited interaction with government agency officials, reflecting a need to incorporate more administrative voices into legislative discourse.

The Refugee Act of 1980 established the first comprehensive U.S. refugee policy. It codified a refugee definition and created the annual consultation process, which requires the president to consult with Congress before determining annual refugee ceilings and resettlement plans. While the Refugee Act of 1980 remains intact, the annual refugee admissions and resettlement plans have changed considerably. The purpose of this dissertation is to analyze this policy to explore its changes from 1980-2018 through the lens of social construction theory. According to this theory, the social constructions of target populations affect policy designs that are adopted with respect to these populations. Policy designs can create and legitimize divisions among different target populations causing some to be perceived and treated as more deserving than others.
This dissertation uses a qualitative research design to analyze narratives within presidential proposal documents and congressional hearings that are held as part of the annual consultation process. These documents serve as the data for this dissertation. I undertake a detailed analysis of the documents of one annual consultation process and related congressional hearings for each president in the period between 1980-2018. In these documents and hearings, different policy actors (congressional members, representatives of the executive branch and state and local governments, and other experts) provide testimony and expert opinions on refugee admissions and resettlement. It is in this context that refugees as a target population are constructed and policies to deal with refugees are debated and discussed by various policy actors. To understand these constructions and the context in which they are created, the narrative analysis elements offered by the narrative policy framework are used as a method.

Public higher education increasingly relies on performance-based funding (PBF) policies to enhance accountability. These policies attempt to steer institutions towards successful outcomes via performance indicators, such as graduation rates. Nationally, PBF policies continue to grow in popularity despite limited evidence that they are effective (Hillman, Tandberg, and Gross, 2014).
Motivated by the apparent conflict between the widespread adoption of PBF policies and the lack of evidence that they actually improve outcomes in higher education, this dissertation investigates the perceived impacts of PBF policies. Florida’s public university system serves as the setting for the study due to its uniquely punitive PBF policy design and the model’s non-standardized performance indicators.