Predictors of Permanent Housing for Homeless Families

Project Leads: Neil Roche and Anjana Sundaram, The Bill and Melinda Gates Foundation

DSSG Fellows: Joan WangJason PortenoyFabliha IbnatChris Suberlak

ALVA Students: Cameron Holt, Xilalit Sanchez

eScience Data Scientist Mentors: Ariel Rokem, Bryna Hazelton

The Bill and Melinda Gates Foundation, together with Building Changes, have partnered with King, Pierce and Snohomish counties to make homelessness in these counties rare, brief and one-time. The goal of this project was to take part in this multi-stakeholder collaboration, and to analyze data about enrollments of homeless families in these counties in programs serving the homeless population, to identify factors that predicted whether families would succeed in finding permanent housing, and to investigate the ways families transition between different programs and different episodes of homelessness.

We developed algorithms to identify families from the individual enrollments and to identify ‘episodes’ of homelessness including back-to-back, or overlapping enrollments in individual programs. We devised innovative ways to visualize and analyze the ways families move from program to program and through different programs, including interactive Sankey diagrams and trajectory plots (see below). These visualizations were particularly interesting and helpful to the county data leads, who have already used them to identify sources of coding errors and to evaluate some of their programs.

Also view the Project Blog.

Program trajectories are a series of programs that a family encounters during a contiguous episode of homelessness