This represents the results of an experimental partnership between the Police Data Accessibility Project (PDAP) and the Center for Analytical Approaches to Social Innovation at the University of Pittsburgh (CAASI). A multidisciplinary group of students works to answer questions from elected officials and advocates in Pittsburgh.
Each folder represents a community question. Start with any folder's README and explore from there.
Here are some things to check out / understand before you get started:
- To learn more about PDAP, start here. Try out the search function! See what it's like to make a request!
- Familiarize yourself with the terms we use and look at some examples of different record types.
- Sign up as a volunteer at PDAP with our docs.
- Look at some open requests using this project.
- Poke around this repository to see some past student work.
To get familiar with the project, try some of our basic volunteer tasks!
- Using this doc as a guide, find an agency in the area. Try to find as many sources of data about the agency as you can.
- Spend an hour labeling potential data sources.
You may be working on a request alone or in a group.
- Initialize the project
- Add a new folder with a README based on the template.
- Make a commit with the README and folder.
- Open a draft pull request for the work in progress. Use the phrase
"fixes #\<issue number>"in the body of the PR.
- Make comments on the request issue asking questions or making updates as needed.
- Make commits to the request's branch as you work; they will show up on the draft PR as you push them to GitHub.
- When the response is ready for review, request a review from others on the project.
GitHub is for storing code-related files and projects. It's great for open-source collaboration and development, because it helps us with version control over time. You can clone this repository so you have a version on your computer for tinkering. If you make improvements, you can submit them for approval. We recommend using GitHub desktop if you're new to GitHub and want to contribute.
PDAP is assisting with the execution of this research project by providing data, tools, and access, but does not endorse or guarantee the accuracy of its findings.
We ran a pilot of the data help desk from about November 2023 to April 2024. We examined the effectiveness of the experiment to make this program more effective over time.
- Visibility on work in progress
- Participants did not publish anything until it was "final", so there was no opportunity for feedback or iteration.
- Work is not graded, so transparency can be the default. We'll start response by creating a README and a draft pull request for each response.
- Collaboration
- Participants exclusively worked alone, so we did not benefit from the group being multidisciplinary.
- We will treat questions as our collective responsibility by default, with room for individual contributions on tasks.
- Use of time
- Our meetings were exclusively used for Python training; this was important, but we missed the opportunity to hear and learn from each other.
- The first agenda item for working meetings is stand-up style progress updates where we show each other our work.
- Project management
- We used Airtable, but participants had read-only access; they did not have enough agency to make changes.
- We're using GitHub for PM; participants will become familiar with it through issues, and gradually gain submission skills.
- Word count and hierarchy
- Students are used to hitting word counts; 5-minute feasibility exercises turned into multi-page written reports, and request responses sometimes missed the forest for the trees.
- In the real world, being concise is its own important skill. We'll edit each other's responses and set max word counts so that details are available only when needed.