Eviction Filing Rates & Homelessness

Using data from the US Department of Housing and Urban Development, the Princeton University Eviction Lab, the US Census Bureau and the MIT Elections Lab, I create a panel dataset observing rate of eviction filings and rate of homelessness by state by year, plus political leaning and population density ranking by state. I explore the data and create a couple of regression models to assess whether a state's rate of eviction filings can indicate higher levels of homelessness.

Data Sources

1. Gromis, Ashley, Ian Fellows, James R. Hendrickson, Lavar Edmonds, Lillian Leung, Adam Porton, and Matthew Desmond. Estimating Eviction Prevalence across the United States. Princeton University Eviction Lab. https://data-downloads.evictionlab.org/#data-for-analysis/. Deposited May 13, 2022.
2. MIT Election Data and Science Lab, 2017, "U.S. Senate statewide 1976–2020", https://doi.org/10.7910/DVN/PEJ5QU, Harvard Dataverse, V7, UNF:6:NFZ83YH7C/fCm6x0stmMwA== [fileUNF]
3. United States Census Bureau. 2016. "Intercensal Estimates of the Resident Population for the United States, Regions, States, and Puerto Rico: April 1, 2000 to July 1, 2010". Washington, D.C.: US Census Bureau. URL
4. United States Census Bureau. 2021. "Annual Estimates of the Resident Population for the United States, Regions, States, the District of Columbia, and Puerto Rico: April 1, 2010 to July 1, 2019; April 1, 2020; and July 1, 2020 (NST-EST2020)". Washington, D.C.: US Census Bureau. URL
5. United States Department of Housing and Urban Development. 2023. "2007 - 2023 PIT Estimates by State". Washington, D.C.: HUD. URL

Currently working on:

• Cleaning up code to better adhere to best practices & improve readability
• Looking for the bottom of the rabbit hole of model assumption violations -_-

General thoughts & comments

2/28: So far I don't think there's much evidence of filerate & homelesscap being significantly related in any way. But at the same time I am very out of my depth; both variables of interest seem very multimodal and I have not really worked with multimodal distributions before. Not sure if this rabbit hole will lead anywhere but oh well. And I'm still not 100% sure I understand the difference between the Least Squares Dummy Variable and within-fixed effects models so I'll just keep scouring the internet I guess!

I doubt I will get to the point of confidently drawing any conclusions from what I have here, at my current level of skill and experience (and with my distrust of this dataset (mostly the PIT, and also my methods of aggregation during the cleaning process)). I may pivot slightly and try going back to working with counties instead of states.

Directory

• data/
  • raw/ Datasets directly downloaded from above sources
    • codebooks_etc/
  • half-processed/
  • homelesspanel.dta
• code/
  • 1-combiningpops.py Combines & cleans the Census population estimates
  • 2-datacleaning.do Creates panel dataset
  • 3-explore.r
  • requirements.txt
• notebooks/
  • explore_model.rmd
  • explore_model.nb.html
• README.md