Code for America working to release Indianapolis police data
In May the White House launched its Police Data Initiative, a program looking to help departments release more data and also use that information to improve how they do things.
Over at WBEZ.org we spoke with a group in Chicago working with the Charlotte-Mecklenburg Police Department on the second problem, trying to improve their early intervention system to find officers at risk for misconduct.
In Indianapolis, a different organization is working with police there to try and open up the agency’s data on police interactions with citizens.
This year fellows from Code for America partnered with the White House and the Indianapolis Metropolitan Police Department to figure out how to release IMPD’s Internal Affairs data. In the process they created an Open Police Data Census, cataloguing what information is available from departments across the country.
This week we spoke with fellows Chris Reade and Laura Ellena about the project and what they’ve learned so far.
How did this project come together and how did you come to work with the White House?
Reade: The way Code for America works, is that we’re all experienced tech industry, city government folks. We get assigned to city that has a broad goal they’d like to have technology achieve. We got assigned to Indianapolis and their Public safety department.
They wanted to look at a bunch of different ways they’ve collected data in the process of doing their jobs and seeing what they could do with that data to help the department. Police departments collect a lot of information in the course of doing their jobs, and they think there is valuable uses for a lot of that in a law enforcement sense, but also in a community relations sense.
We were lucky and got invited to go with Indianapolis to the White House’s police data convening. They had brought together a whole bunch of different police chiefs from around the country to use data for a few specific ends. One is building trust in the community and the other was internal focused, how do we use the data we’re compiling to do our jobs better.
As a part of that meeting [Indianapolis] committed to releasing a pretty healthy set of data. We were really excited to be a part of that meeting, but also excited to be help them. We’ve been working with them since on a subset of the datasets to automate the release of that data and figure out what’s the best way to get that to the public and present it to the public and enable the public to use that in a way that furthers everyone’s goals.
When you first got started, what were those interactions like?
Reade: Everyone was really excited to have us there. There are a lot of ideas of that people have for all sorts of different uses that police data could be used for to further all kinds of different goals. So not just the community relations aspect. We went on a couple of ride alongs, we talked to a lot of the moving parts in the public safety department in how they use technology, collect data and are trying to make better use of it. There are a lot of moving parts in any public safety department. You have 15 different logins you’re trying to remember to get through your day. There’s a lot of technology in the field, and I think people are looking for a way to tie related datasets together.
What has the day to day been like for you?
Reade: We’ve been doing a lot of reaching out to anyone who would be interested in using this data external to the public safety department. We’ve been talking to neighborhood groups, researching institutions in Indianapolis, more activist focused groups. We’ve been trying to collect what people are looking for from this police accountability data, and then the technical aspects of releasing this. We’re lucky in that most of the information that we’re focused on releasing is in one application, so extracting the information from that system is a privacy-focused way that is then useful to the people outside the police department are interested in seeing that information
What datasets are you looking to release?
Reade: Indianapolis like many cities has a couple of ways of viewing crime data. When you talk to people about police data, often the first thing they come up with, is what the actual crime information is. Indianapolis is already pretty good about that. We’re focused on officer involved shootings, use of
force, and citizen complaints.
What have you been hearing from groups about what they’d like to see released?
Ellena: Location is really important to people; everyone is curious what is happening in their neighborhood to give some context to the data. Both department of public safety and community groups are interested in educating residents on how to have better interactions with the police, so thinking about how to provide that context next to the data.
Having worked through this a little bit, what are your goals for when the fellowship ends (Nov. 15)?
Reade: What we’re looking at now is a way of extracting information from the internal affairs application, scrubbing that information of any identifying information, anonymizing the location data, and then summarizing it in a way that meets the needs of the community, and then provides enough context to understand what the numbers actually mean and what the department is doing to change those numbers on a publicly-available website.
Ellena: We want to satisfy the data nerds who want to download the whole dataset and play with it, but also give your average resident a summary website that lets them to see what’s going on without getting too technical.
Reade: One other initiative we’re working on, before we dove into this we wanted to see what other departments were doing, so we put together a police open data census to see what other departments are doing. Who’s releasing this, in what format and how much context they provide.
Ellena: This is so new, there are no best practices yet for police departments, so we built that partially for ourselves to be able to make recommendations in Indianapolis, but also for anyone else who is interested in police data and other cities who may want to release things on their own.
What sort of things were available when you started looking?
Reade: The big thing we focused on was use of force and officer involved shootings, and there are a couple of departments who have really pushed ahead. Dallas comes to mind, Philadelphia, Denver and the district attorney’s office, they release really detailed narratives on every incident.
A lot of departments do release this information, but not in the open data way we approach things. Many departments will release an internal affairs report every year that has all of their complaint data and often a lot of use of force information. Unfortunately, they’re often in PDFs squirreled away deep in the departments website. I don’t think that’s necessarily on purpose. A lot of departments that have concerns about this don’t realize they’re already publishing something very similar that might be better served on a website than in a PDF document.
Was there anything that surprised when you saw what departments made available?
Reade: The depth that some departments are comfortable going into real incident-level descriptions was something we were a little surprised by given that it is new and pretty sensitive for everyone involved. What we’ve heard, some groups really do have a desire for that level of detail.
Where there any major data needs?
Reade: The incident level information is pretty varied. Most places don’t get into the incident level information. And there isn’t a trend in the size of the department and their willingness to do it. A lot of big cities are in it, a lot of small cities are in it, there are a lot of big cities that don’t seem to release this information online. Which doesn’t mean they aren’t releasing it, I just means that it might not be within our Googling skills.
Ellena: It sounds like a lot of departments are in the process of making the transition from tracking these things on papers. If they don’t have an internal affairs system, it’s likely that this information is in file cabinets rather than a database.
What do you see happening if all this data was out there? How would it improve policing?
Reade: One of the big ones is just demonstrating the police department's willingness to be transparent, to be open to community oversight and recognize that they are servants of the people and people deserve to have the information they need to provide that oversight.
We also think that grounding the conversation in something tangible is beneficial. Having real numbers and trends to point at will help people to have a better conversation about policing in the community.
Ellena: With the community and internal, too, maybe you can identify if there are reasons you respond to a call or times and days that these incidents that go poorly have in common, and can we use that information to make things better.
ellievsbear
prguitarman
dededos
discoholicmusic
baratheonpilled
romaincrisis