Skip to content
This repository has been archived by the owner on Mar 8, 2024. It is now read-only.
Benjamin Trevino edited this page Jan 5, 2022 · 8 revisions

Welcome to the learning_transform911_hawaiii wiki!

The discovery for this repo was inspired by Code for Americaʻs 2021 National https://www.codeforamerica.org/national-day-of-civic-hacking/ndoch-2021/ From the site: On September 18, Code for America will host our 9th annual National Day of Civic Hacking, including a kick-off event with experts in the field, coordinated action-teams dedicated to data, analysis, research, design and more, a lunch-and-learn, and closing remarks. We welcome people of all skill levels, and new and returning volunteers alike. There will be actions available to leverage different skills, technical and nontechnical.

Ongoing data projects and efforts across Code for America are being collected here: https://data.world/codeforamerica

Our learning objectives are:

  • Learn what makes an effective open records request
  • Learn about the available 911 data (and what gaps there are for effective analysis)
  • Learn about the processes that create the 911 data

UIPA Requests

911 Study Examples

Data Analysis Questions from the Code for America T911 Effort

The big question: how do the 911 dispatch dispositions compare to the final dispositions validated in the PD incident report?

  • Can we identify calls relating to the people most impacted by the 911 system: mental/behavioral illness, substance use, homelessness, people historically mistreated by law enforcement. Once identified, what can be learned about these calls?
  • Is there an indication of call severity? How does the perceived severity of calls by 911 call takers align with the impacted people above? What other severity trends are visible?
  • Is there evidence of agencies other than fire, PD, EMS that are involved (mobile crisis teams, CIT trained PD, etc)?
  • Is Text-to-911 available? How often is it used? Are there any trends for texting vs voice?
  • How closely aligned are the 911 call types to the current APCO standards?
  • Can calls with/without minors or weapons be identified?
  • If the 911 data has location data, can you connect PSAP data to other census demographic data (hint: data.world makes this easier)? What can you learn?
  • If you have many years of call data, can you identify time-series trends (call types changing with changes in median housing prices, etc)?

Data Availability Questions

  • Is it possible to get 911 call transcripts in this district?
  • How are 911 calls associated with PD incident reports? What are the connecting fields, and what degree of confidence is there in the connection?
  • If some data (like 911 transcripts) aren't available to the public, could they be made available to an academic project run by a university? Or are there other occasions when this information has been shared?