In 1969, Stanford psychologist Philip Zimbardo conducted an experiment. He parked a car without license plates and its front hood up on a street in the Bronx, New York City. Within ten minutes, the radiator and battery were stolen. Twenty-four hours later, everything of value had been removed. Soon afterwards, the car's windows were shattered, its upholstery ripped up. That’s when Zimbardo returned to the car… and took a sledgehammer to it. Soon, the car had been flipped upside down and completely vandalized by passersby.
This experiment, and countless others like it, form the basis of the Broken Windows theory. The theory holds that the physical condition of an urban area can incite or prevent vandalism and the development of more serious criminal behavior. In the words of George Kelling, the father of the Broken Windows theory:
“Consider a building with a few broken windows. If the windows are not repaired, the tendency is for vandals to break a few more windows. Eventually, they may even break into the building, and if it's unoccupied, perhaps become squatters or light fires inside. Or consider a pavement. Some litter accumulates. Soon, more litter accumulates. Eventually, people even start leaving bags of refuse from take-out restaurants there or even break into cars.”
Kelling’s theory has been vigorously debated since its inception in the 1980s. One side argues that changing perceptions of safety does not necessarily reduce crime rates. But how can we study the relationship between perceptions of safety and actual safety? We can find statistics related to crime rates, but how can we measure perception?
Which place looks more beautiful? Probably the one on the right. Source: Place Pulse.
This is where the folks from the Place Pulse lab at MIT come in. These researchers have compiled thousands and thousands of Google Maps images of neighborhoods from various cities across the world. Their aim is to quantitatively recognize which parts of cities, and which cities as a whole, are perceived as safe, beautiful, fun, boring, and so on. To determine perceptions, the Place Pulse lab has crowdsourced its experiment. (You can participate in this experiment here.) Online users are presented with two unlabeled images and are told to click the place that looks more beautiful or safe, depending on which category of perception is selected.
Cities ranked on a spectrum from “Less Safe” to “More Safe.” Source: Place Pulse.
From these selections, researchers have ranked cities across different perception categories. For instance, users have perceived Washington DC to be the safest and the wealthiest of all the cities included in the experiment. Users rated Belo Horizonte, Brazil, as the least safe and least wealthy of the cities tested.
Having captured perceptions of safety into quantitative measures, the next step will be to study the relationship between perceptions and reality. Does perceived safety correlate to lower crime rates? Does perceived fun or liveliness correlate to economic growth? These are the questions that remain to be answered.
A possible objection to the project is that we do not know the demographics of the online users participating in the experiments. If primarily middle-class, suburbs-dwelling Americans are logging on, then perceptions of safety and beauty may be skewed towards cookie-cutter housing and Starbucks-filled strip malls.
To get occasional notifications when we write blog posts, sign up for our email list.