Data Dive to Study Homelessness
We are partnering with DataKind to describe LA's homeless population over time
What is the DataDive?
This July, the Economic Roundtable is partnering with the non-profit data-for-good organization DataKind to build knowledge around long-term dynamics of the homeless population in Los Angeles. Together, we are holding a simultaneous DataDive in Los Angeles and the Bay Area on the weekend of July 21-22, 2018. (The Los Angeles event will either be held at USC or UCLA, pending final confirmation.)
At the DataDive, volunteer data scientists, DataKind Data Ambassadors, and we at the Economic Roundtable will be collaborating to answer the question: “Who experiences homelessness in Los Angeles over the course of a year?”
This is an unanswered question. The answer is important for understanding the total scope of homelessness and for identifying possibilities for helping homeless individuals earlier rather than later.
We are seeking volunteers with statistical expertise who work in R or Python to help us tackle this problem. Experience with missing data, survival analysis, and interactive visualization in R/shiny would be especially useful. If you are interested or have questions about volunteering please fill out this short interest form.
Why are we interested in this question?
In April, the Economic Roundtable released the report Escape Routes, a meta-analysis combining 26 point-in-time data sets to provide a single panoramic description of people without homes in Los Angeles. Point-in-time estimates, though extremely valuable, do not immediately describe the dynamics of who is homeless over periods longer than a day, nor do they reflect the high turnover in the population.
In Escape Routes, we presented a simple model to estimate the entire population that experiences homelessness over the course of the year (the annualized population) based on point-in-time data. On a given night, about half of the people experiencing homelessness have been homeless for over a year. However, our model suggests that group only accounts for a fifth of the annualized population. On the other hand, it is likely that half of the annualized population was homeless for only two months or less.
We frequently read that roughly 50,000 people are homeless in Los Angeles, yet the annualized population is probably two to three times as large. The difference between the point-in-time and annualized populations is under-acknowledged, but extremely important to policymakers and service providers.
At the DataDive we will improve upon the simple annualized population model in Escape Routes, and expand the scope of our modeling targets. We will also analyze population demographics and express measures of uncertainty in our estimates. Finally, we will create dynamic visualizations to make our findings broadly accessible.
Together, we will highlight and advance our understanding of a very important issue at the intersection of data and policy.