Investigating Causal Inference of Policies: Ban the Box on Employment, Wage Transparency on Wage Gap, and Semi-Automatic Indexing & Tagging of CMES Medical Educational Documents

<p>This dissertation is a collection of three chapters that investigate selected policy issues and their broader implications, specifically focusing on the causal inference of policies: Ban the Box on employment and Wage Transparency on wage gaps. The last chapter uses natural language processing to semi-automatically create indexes for medical documents. Chapter 1 - Ban the Box (BTB) policies move the inquiry about the candidate’s criminal record to later in the hiring process. The intention of this policy is to have employers evaluate the candidates based on their performance on the interview and qualifications instead of their criminal history. In this paper, we will study the effect of Ban the Box policy on employment for young men with no college degree in United States. We find that in the metropolitan areas where the policy was implemented, there is a negative effect on employment for unskilled young men; however, we are unable to conclude that these policies negatively impacted employment of unskilled young Black and Hispanic men with strong confidence. Chapter 2 - Gender-based wage disparities persist in the United States, despite significant strides toward workplace equality. This study investigates the wage gap, emphasizing the urgency of addressing this issue. Drawing on research by Blau and Khan (2017), we examine the extent of gender pay inequality and its implications for full-time private sector employees. In response to these disparities, legislative changes have been implemented at both the state and city levels. In particular, Ohio has proactively mandated wage transparency for firms in two of its metropolitan areas. This paper examines the impact of such legislation in two major cities in Ohio: Cincinnati (implemented in March 2020) and Toledo (implemented in June 2020). We find that the wage transparency policy in Ohio has a significant positive effect on female employees, resulting in a 4.8% increase in their income levels, while it has a significant negative effect on male employees, with a 6% decrease in their incomes. Furthermore, among female employees, white women are the only group positively impacted by the policy, while racial minority women experience no statistically significant change in their incomes. Among the male population, the incomes of non-white male employees are reduced by a higher percentage than those of their white counterparts. This suggests that, although the policy has reduced the gender wage gap, it has not addressed the gap between racial minority groups. Chapter 3 - Tagging of CMES Medical Educational Documents: Our project uses NLP (natural language processing) to semi-automatically create indexes for medical documents that are currently in PDF format. We improved the search engine (specifically, the searchable document tags) of the medical education documents in the CMES-Pi database owned by CMES (Continuing Medical Education on a Stick).Our aim is to examine a plethora of medical education documents and their existing tags, and then leverage use of NLP tools to determine an appropriate balance of broadness and specificity for tagging future documents with.</p>

Name thesis:degree_name

Economics, PhD

Level thesis:degree_level

Restricted to Claremont Colleges Dissertation

Discipline thesis:degree_discipline

School of Social Science, Politics, and Evaluation

Year dc:date.available

2025

Author dc:creator

• Avgil, Dana

Contributors dc:contributor

• Scott Cunningham
• Melissa Rogers

• Economics

Repository record dc:identifier

https://scholarship.claremont.edu/cgu_etd/932

OAI identifier oai:identifier

oai:scholarship.claremont.edu:cgu_etd-1954