Understanding educational polarization in California through an analysis of ballot initiatives in the 2020 election
Shreeharsh Kelkar, Professor
Interdisciplinary Studies Field (ISF)
Closed. This professor is continuing with Spring 2024 apprentices on this project; no new apprentices needed for Fall 2024.
Political scientists have noticed a large contemporary trend in the US where high-income college-educated people tend to vote Democratic rather than Republican, a reversal of earlier patterns; they have labeled this as "education polarization".
Education polarization is best illustrated through the case of Prop 22 in California. In 2019, responding to a California Supreme Court ruling, the California Legislature passed Assembly Bill 5 (AB5) which would have turned app-based gig workers (employees of Uber, Lyft, DoorDash, etc.) from independent contractors into full-time employees. In response, Uber, Lyft, and other gig companies financed a new ballot initiative which came to be called Prop 22. Prop 22 would create an exception to AB5 for app-based drivers who would remain independent contractors but would gain a suit of benefits (lesser than employee benefits but more than what they got currently).
Despite California being a reliably “blue” state, Prop 22 passed resoundingly with around 58% of the voters voting “Yes.” When the results are examined county by county, they illustrate something striking about political/partisan polarization in California, and by implication, the United States itself. The Los Angeles and Fresno counties, with their substantial Hispanic populations, a comparatively low median wage and low educational attainment, voted “Yes” on Prop 22 by margins as large as 55-44 and 65-34 respectively. On the other hand, the San Francisco, San Mateo and Santa Clara counties, home to Silicon Valley (i.e., to the gig companies), with their median income in six-figures, and high educational attainment, voted mostly “No” on Prop 22 (40-59, 49-50, and 53-46 Yes-No votes respectively), i.e., against the gig companies. This suggests that the most important predictor of the Prop 22 was educational attainment: the higher the college educated population in the county, the more likely it was to vote “No.”
This project will analyze educational polarization in California (and by implication, the US) through a quantitative analysis of the fate of four California ballot initiatives in the 2020 election: propositions 22, 15 (a property tax repeal), 16 (affirmative action), and 25 (ending cash bail through algorithms). While there is clearly an orthodox "red" and "blue" position on all these issues, education polarization has scrambled traditional loyalties that voters feel towards different parties, and therefore their stances on issues. What characteristic of the electorate--income, occupation, educational attainment, race, ethnicity--best predicts their position on these issues?
Qualifications: Learning outcomes:
Over the course of this project, the student will be able to:
- articulate the history and dynamics of political polarization in the US.
- develop and hone their skills at quantitative research methods (scatter plots, regressions) and using the US census.
- develop and hone writing skills through writing a report that describes clearly the research findings and justifications for methods.
Tasks
Over Spring 2023, the student will have to carry out the following tasks:
- collect and tabulate demographic data on the different counties in California using the US census
- plot scatterplots and carry out regressions correlating county demographics with their vote on different ballot propositions in 2020.
- writing a report describing the findings.
Student qualifications:
Required:
- some experience with Excel or any other tabulating software
- some experience with basic quantitative methods such as making scatterplots and if possible, regressions.
Desirable but not essential:
- the student should have some background and interest in the history of political polarization in the US.
- ideally, the student should be either a junior or a senior; those majoring in ISF, media studies, american studies, history, anthropology, sociology, rhetoric, political science/economy, computer science, and data science are especially encouraged to apply.
Hours: 3-5 hrs
Related website: https://shreeharshkelkar.net
Digital Humanities and Data Science Arts & Humanities Social Sciences