There is a great amount of data available today to help citizens understand how their community is changing, provided that there’s a thoughtful way to make meaning of that data. This summer, our Civic Technology fellow, Aaron Myran, is working how public data and APIs can be unlocked, visualized and shared to facilitate civic engagement and policy decisions. His first project looks at the Boston housing market and rent rates. Check out his blog post (below), as well as his code that’s posted on Github. Thank you Aaron for this great work—we look forward to the next segment of this series!
— Cathy Wissink
A lack of real-time data on how urban neighborhood prices are changing makes gentrification a difficult public policy challenge. In some cases, the most accessible longitudinal data for cities is the census. As a result, soaring housing prices in working class neighborhoods can quickly displace residents.
Unlike the census, or even annual tax returns, housing rental websites provide up to date data on the cost of leasing a home or apartment. To display how gentrification is occurring in the greater Boston area and across the country I scraped rental market listings, including zip codes, and price per square foot from rental market websites and used Power Bi to produce data visualizations to understand changes in housing prices.
The dashboard below, created with Microsoft’s Power Bi, visualizes the change in price per square foot for the rental market in Boston over time, starting in mid-June, 2016. The heat map of Boston shows zip codes with a lower average price per square foot for listings in light red and zip codes with higher price per square foot listings in darker shades of red. The visualizations to the right illustrate how these prices are changing over time. To focus on a specific zip code, click on the zip code on the map to filter it in all of the visualizations.
While there haven’t been major changes over the last few weeks, you’ll be able to revisit this dashboard in the future to see how these trends are changing. In addition, this dataset provides information to policy makers on the average affordability of a given neighborhood in real-time. This data, in conjunction with income data can aid government in determining if residents will be displaced by increasing rental prices.*
In addition to collecting data for the city of Boston, I was also able to analyze data for the all of the cities in the United States with more than 50,000 residents. This Power Bi dashboard can also be seen below.**
Next Steps? Housing market data isn’t the only piece of the gentrification puzzle. Overlaying data such as public works projects, distance from public transit lines, changes in employment demographics, percentage of adults with bachelor’s degrees or higher and other data can help forecast where gentrification is occurring and help policy makers plan accordingly.
If you’re interested in displaying how housing prices are changing in your city and identifying gentrification trends and key indicators, you can access my GitHub account with the Python scripts that were used for the data collection and use Microsoft’s interactive data visualization and dashboard builder, Power Bi, to display the data.
*The scraped data represents a sample of the Boston housing market as not all rental listings are put online or are listed on websites in a structured manner.
** A few cities do not use the sites scraped to display data and are not summarized in the dashboard.
Bio: Aaron Myran is a civic technology fellow with Microsoft and a graduate student at the Harvard Kennedy School.