competition complete

DAT102x: Predicting Evictions
Hosted By Microsoft


Time Remaining

End date: May 15, 2019, 11:59 p.m. UTC

About Evictions

Over the past five decades, housing costs have risen faster than incomes, low-cost housing has been disappearing from the market, and racial disparities in homeownership rates have deepened. This has put many in a perilous situation. As the Eviction Lab explains:

Today, most poor renting families spend at least half of their income on housing costs, with one in four of those families spending over 70 percent of their income just on rent and utilities. Only one in four families who qualifies for affordable housing programs gets any kind of help. Under those conditions, it has become harder for low-income families to keep up with rent and utility costs, and a growing number are living one misstep or emergency away from eviction.

Housing issues are intertwined with many other social problems, including poverty, educational disparities, and health care. Data science provides tools for predicting risk of eviction and understanding the factors that increase that risk. Digging into underlying patterns in evictions data can be an important first step toward greater visibility and better policies.

In previous capstone challenges, we looked at how poverty and heart health are distributed across US counties. Now, we are challenging you to consider how poverty, income, health, ethnicity, and other sociodemographic factors are related to evictions, and to use your skills to build a model for predicting eviction levels in counties across the United States.

Competition tile photo by Garry Knight