Dev sprint 8: Transition
Sprint 8 was aimed at aligning around goals & expectations for the CDE and setting direction for the next 60 days. Specifically, 18F was asked to assess the value and feasibility of visualizing ORI-level data. Our research focused on assessing the accessibility of our API and exploring approaches for surfacing ORI-level data.
- The team built and tested a prototype for selecting ORI-level data within a state: https://crime-data-prototypes.fr.cloud.gov/demos/ori2/
- Added documentation to make the API easier to work with.
- Developed a “data dictionary” and descriptions for the datasets made available by the CDE to improve the download experience.
- Selected bug fixes and UI/UX polish.
To assess the usability of the API the 18F and CJIS team spoke with 3 participants, which included a member of the media and technical staff from CJIS and 18F. To understand the value and feasibility of visualizing ORI-level data we spoke with 2 types of participants — state UCR staff and a developer and data visualization specialist from the private sector.
- How do you find out about APIs you can leverage?
- What are the holes in the API UI and documentation that prevent people from getting started?
- Does the naming scheme make sense to users? Are there inconsistencies in the documentation that hinder usability?
API Usability
- The current level of documentation is not sufficient for enabling broad access to the data via the API — more work is needed to better explain what the API can do and how it’s structured, especially for those that are unfamiliar with the UCR program.
ORI level navigation
- UCR staff we spoke with receive requests from a broad spectrum of consumers that are similar to our target audience for the CDE. The general public tends to think about crime in terms of “towns and cities”, not ORI.
- When visualized, low amounts of data raise concerns about the integrity of the data.
“The charts and graphs are nice, but just give me the data.”
“We typically don’t use the FBI’s site, we have our own (state-based) data to benchmark against.”
“ORI doesn’t mean anything to me. I tend to think about cities and towns.”
“Low amounts of data make me think something is wrong.”
Sprint 9 will focus on evolving opportunities for improving the granularity of the CDE experience. 18F will also conduct activities in parallel that help prepare the site for launch, such as planning for pilot testing.
