Challenge 2 - Examining Reasons for Eviction Filing #42
KathrynMercer
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
TLDR: Leading causes of eviction have changed over time. Evictions could potentially be reduced further by providing/subsidizing legal counsel to evictees to mitigate "Nuisance" and "Breach [of contract]" causes for eviction.
What is it: Visualizations of most frequent causes of eviction in the San Francisco area from 1997-2024
What data was used:
San Francisco Eviction Notices from 1997-2024
What tools were used:
Jupyter notebook w/ Python
Pandas, Numpy, and MatPlotLib/Seaborn libraries
Limitations of the data:
This data reflects San Francisco eviction filings only. It may or may not be generalizable to the broader populations of LA/CA (or FL!)
This data is eviction FILINGS, which may or may not reflect actual evictions.
What does it show:
The number of evictions decreased substantially during the COVID-19 pandemic and has not (fully) rebounded to previous levels. I don't know what eviction moratorium/other allowances were made during the pandemic in San Francisco, but there was a big decrease in evictions, so probably they worked.
The causes of eviction have changed substantially over time.
Highlights:
Next Steps:
Interactable visualization - I want to throw this in Plotly/similar to make a the visualization interactable - allowing the viewer to look at a pie chart of most common causes per year or other customizations.
Generalizing to other areas - I'd love to do something similar with other areas (LA, Florida, etc.), but I haven't found data to do that yet. EvictionLab seems to only have aggregated data & only up to 2018. If anyone feels like contacting the LA Housing Department to get access to more data, that'd be cool. They seem to require landlords to upload "Notice to Terminate" online, but the data can only be searched by address online, so not really helpful to us as-is.
Demographic data - This data (understandably) doesn't publish who is getting evicted, so I couldn't make any generalizations about demographics, income brackets, etc.
Mapping - This data can also be organized by zipcode or block to make a geospatial visualization (though I'm less convinced that's helpful). Unless maybe you were going to try to match reason for filing/number of filings with demographics based on zipcode census data.
Beta Was this translation helpful? Give feedback.
All reactions