Using Data to Solve Multi-Jurisdictional Problems in Western Pennsylvania

An ambitious open data project is unfolding in western Pennsylvania. It’s aimed at building an online clearinghouse where local governments and other organizations in the region can publish datasets about topics ranging from jail populations to property values.

The Western Pennsylvania Regional Data Center is being managed by the University of Pittsburgh Center for Social and Urban Research, in partnership with the city of Pittsburgh and Allegheny County, which encompasses the city. The center's online open data portal launched in mid-October. It currently includes data from the county, the city and the university.

The idea is that over time the portal will grow to include information from more sources, including other municipalities and universities, as well as organizations like nonprofits.

"We're really excited about the model," Laura Meixell, analytics and strategy manager in the office of Pittsburgh Mayor Bill Peduto, said by phone on Monday.

Meixell noted that there are 130 municipalities in Allegheny County “and we understand that we have a lot of problems that go in between municipal borders.”  

The data center is seen as an opportunity to work across jurisdictions and organizations, by getting data from around the region in one place, in a structured format. As an example of why this is important, Meixell pointed to issues involving sewer overflows and water quality and said, “the watershed is an area that doesn’t pay attention to municipal borders.”

“We really wanted to build something that can grow into a central information resource for solving problems,” Meixell added.

As part of this effort, the center has sought to develop a legal framework to make it easier for a wide variety of groups to share data through the its online platform. City, county and university attorneys worked together to come up with these legal guidelines, according to Bob Gradeck, the data center’s project manager.

This legal work yielded a two-page agreement that anyone publishing information through the data center has to sign, which outlines a publisher’s rights and responsibilities. “We just make sure that they commit to documenting their data, and all the privacy implications that come with sharing data,” Gradeck said.

Pittsburgh’s Meixell noted: “We can take on new data providers pretty easily and have a pretty clear understanding of where the different responsibilities lie.”

Some of the center’s main privacy concerns surround issues like ensuring that datasets do not include information such as names, social security numbers, or spreadsheet columns that could contain notes with sensitive information.

As of Monday evening, the data portal featured 118 datasets.

Some of them shed light on complex public policy areas.

For example, there’s a daily Allegheny County Jail census, which provides information about the race, gender and age of inmates. As of Nov. 1, there were 2,419 people jailed.

There’s also data from the Allegheny County Office of the Medical Examiner that documents instances of fatal accidental drug and alcohol overdoses, including the substances involved. Heroin, the painkiller fentanyl, and cocaine were among the leading drugs contributing to the deaths documented so far this year in the county.

Pittsburgh, meanwhile, has posted a dataset with 28,397 service requests submitted through the city’s 311 system between April and November.

The most frequent requests—6,709 of them—pertained to potholes.

There’s also a slew of geographic datasets, like one with the locations of 739 sets of city steps in Pittsburgh, and another with coordinates for 143 city bridges.

So far, the most commonly accessed data was related to property assessments, police incidents and 311 requests, Gradeck said.

Meetings with “data user groups,” he added, will help guide the prioritization of future data releases. The meetings are meant to be open to anyone with an interest in the data center’s work. “It's getting a lot of people together that just want to say: ‘hey, I want to learn more about this,’ or ‘can you share this kind of data,’ or ‘here's what I'm working on,’” Gradeck explained.

Financial support for the data center has come from a number of sources including the University of Pittsburgh and the Richard King Mellon Foundation.

The foundation committed to provide $1.8 million.

Gradeck said the goal for now was to keep the cost of the center in the “hundreds of thousands” of dollars per-year range, but was unable to immediately provide an exact dollar amount for how much the project has cost to date. “We’re trying to be pretty lean,” he added.

The data center’s staff, he said, was shaping up to be the equivalent of about three-and-a-half full-time people, including a project manager position, a research liaison and developers.

The software used for the center’s data portal is a system called CKAN. It was developed by the nonprofit Open Knowledge Foundation, and is now managed by the CKAN Association.

Gradeck has some advice for other places considering projects like the data center.

“Don't just jump out and buy technology,” he said. “Because that's something you do after you figure out exactly what kind of initiative you need.”

Referring to the data center, Pittsburgh’s Meixell said: "We really see it as leadership in open data for our region.”