StreetLight Data, which provides transportation analytics to both public-sector agencies and some private companies, has developed a new tool to measure cycling and walking activity. Much like its other data analytics tools, StreetLight Data examines publicly available location and movement data from mobile phones to get an understanding of the different modes of travel on streets, bikeways and other routes.
This information is key to not only put cycling infrastructure where it's needed, but to measure the soundness and use of that infrastructure, said Laura Schewel, CEO of StreetLight Data.
“One of the most important things we need to do for climate change and for urban vibrancy, as well as just for human health, is have more biking and walking. And if we’re not measuring it, we will not be able to manage to that goal,” she said, speaking last week from the annual meeting of the Transportation Research Board in Washington, D.C.
Some of the obvious use cases for the data include “infrastructure prioritization,” which allows transportation officials to make better decisions related to where biking and pedestrian infrastructure should be placed.
“You have to figure out, ‘where do I prioritize investment, and which types of things do I put where?’” said Schewel. “And if you are in the dark in terms of how many bikers and pedestrians are already in a place, you are not going to do a great job prioritizing.”
Information related to the pedestrian and bicycle traffic may be increasingly desired by cities as they look to become friendlier to these modes of mobility — often clearing the way for bike-share operations such as Jump or Lime — in efforts to reduce the number of vehicles on streets.
“It’s a sort of civic pride to have more bikers and multi-modality. There’s a lot of just political power behind the idea,” said Schewel.
Until the emergence of real-time data harvesting operations like StreetLight, knowing how many bikers were using city streets was often a vague and sometimes inaccurate snapshot.
The American Community Survey (ACS), conducted by the U.S. Census Bureau, has generally been the source officials turned to for cycling or pedestrian data. Incidentally, the ACS has pointed to reduced bike use in cities nationwide. The number of workers who commute via bike fell 3.2 percent from 2016 to 2017, USA Today reported. Looking at the nation’s 70 largest cities, 37 had a year-over-year decrease in biking from 2016 to 2017, the League of American Bicyclists reported in September 2018, relying on Census data.
However, the ACS survey may be too narrow a view into cycling, said Ken McLeod, policy director at the League of American Bicyclists.
“It's hard to say how good census data is at being representative,” said McLeod. “It only looks at trips to work and only counts someone as a bicycle commuter if they used a bicycle for most of their trips to work in the week prior to receiving the survey.”
Other data, said McLeod, suggests work commutes make up only 20 percent of total bicycle trips.
“So it's very possible that the census data does not do a great job of representing all trips,” he added. “If a city has a vibrant recreational cycling culture or a retired cycling community that would be entirely absent from the census data.”
Schewel stressed StreetLight Data collects all trips, regardless of the reason for the trip or how long it is.
“Both the survey data and our data are samples. But, I will say, there are some clear things that are different between our data and the survey data,” said Schewel.
“One difference, that is expected, is we see a lot more short, short trips,” she explained. “And that’s because people tend to forget short trips — that’s a known problem with surveys, people don’t recall their short trips very well.
“And the second thing is, the trips that are actually mixed mode. Like if I drove my car to work, and the parking lot was half a block away, and then I walked from the parking lot to the office, most surveys wouldn’t call that walk from the car to the office a trip. But we call it a trip,” she continued.
Cities seem to be interested in having a clear picture of bicycle use, and are generally requiring bike-share operators to share location and use data with the city as a condition for receiving an operational permit. For example, new regulation to take effect March 1 in Seattle will require bike-share companies to report trip data every three minutes, down from every five minutes today.
“We’re switching to a new data sharing protocol in which the vendors make certain data available via Application Program Interface (API) instead of sending reports to a third-party data warehouse,” explained Dawn Schellenberg, a spokeswoman for the Seattle Department of Transportation.
For now, StreetLight Data will not distinguish between personal bikes and those operated by app-based rental companies.
“Right now, a bike is a bike, whether it’s shared or not. So we count them all,” said Schewel
