The California Health Data Project team is exploring the feasibility of crowdsourcing human services data statewide so that application developers can query data from a single database and allow those apps to be scaled across the state of California. In order to get there, we first need to understand local efforts to collect, standardize and make this data available.
We're using GitHub Issues to organize feedback and work out in the open, so that we can better understand who's doing what and where the greatest synergies lie. But we need your help.
Background
The California HealthCare Foundation (CHCF) launched the California Health Data Project in March 2015 with three pilot projects running in Sacramento, Fresno, and Los Angeles.
The project is bringing together the California Health and Human Services Agency (CHHS), local governments and most importantly, the community — service providers, civic technologists and advocacy groups — to ensure this valuable information is finding its way into the hands of people and organizations that can put it to good use.
The Sacramento pilot project is led by Code for Sacramento core team member Joel Riphagen and myself. We held a Sacramento Health Data Roundtable meeting this past May to discuss current uses of health data in the region and explore opportunities to create additional value by leveraging the CHHS data portal. We also collected additional feedback from the community through surveys and met with some stakeholders individually.
Significant Interest in Social Services
A consistent theme that emerged from conversations in Sacramento (as well as Fresno and Los Angeles) pointed to the lack of openly available information regarding social services. Specifically, our communities have expressed strong interest in a Web application that would allow users to query information about providers based on their location and specific need (SMC Connect is an example of one possible implementation). The California Department of Social Services (CDSS) does not collect information about social service providers and therefore would not be able to publish that data to the state portal. So we needed to conceptualize a more practical solution.
While the creation of a comprehensive statewide database that incorporates the needs of all stakeholders far exceeds the scope of this project, a potential opportunity exists to deploy a lightweight solution to crowdsource this data and publish it upstream to the state portal that could serve as the first iteration to validate user needs. Best of all, CDSS is willing to jointly explore the feasibility of this project, as it’s consistent with the department’s mission:
"The mission of the California Department of Social Services is to serve, aid, and protect needy and vulnerable children and adults in ways that strengthen and preserve families, encourage personal responsibility, and foster independence."
Our Conceptual Solution
We envision a solution that consists of four components:
1. A Web application that allows users to submit new or updated information about social service providers;
2. Publication of approved submissions to the CHHS Open Data Portal;
3. A Web application for querying data about social service providers; and,
4. A feature for flagging and reviewing erroneous information.
We believe this is an elegant solution because it leverages existing technology, could be deployed within a few months, is completely voluntary, would require few additional resources, and empowers communities to contribute their on-the-ground knowledge.
Web Application for Data Submission
To efficiently crowdsource the data, anyone should be able to submit new or updated information about social service providers without the need to create a user account. Form fields should be minimal, representing only the most critical information a user needs to discover relevant human services.
Publication to the Data Portal
Data entered into the application would automatically sync with a social services data set in the CHHS portal. That data set should include a disclaimer that the state is not responsible for the accuracy of the data, i.e., the state’s role in this effort is merely to provide a single database where information about local social service providers throughout the state may be stored.
Web Application for Querying Data
To enable a consumer facing experience with the crowdsourced data, we would develop and deploy a client-side Web application from which users may query data from the state portal via the portal’s application programming interface (API). The Web application would help users statewide find the services they need based on their location.
Feature for Flagging Erroneous Information
An approval process on the front-end would help filter erroneous information and ensure data quality, but requiring such an approval process could pose too a barrier for implementation. So we instead recommend that users have the ability to flag erroneous information from the web application, which would trigger a review process whereby the information could be openly discussed and resolved by the local community. These volunteer data stewards would play a similar role to that of Wikipedia editors.
How You Can Help
Before moving forward with this ambitious effort, we want to learn more about existing efforts at the local level to modernize the collection and publication of human services data. Perhaps we can replicate or scale existing efforts; for example, by federating human services data already published on local open data portals. So we’re asking stakeholders to help us answer some of the following questions:
• Who operates 211 in your county?
• Are there other entities in your county that currently collect and publish human services data?
• How is human services data collected and published in your county? Are there efforts underway to modernize how this data is collected and published?
• Does the data conform to an open standard, such as the Human Services Data Specification? Can the data be accessed via API?
• Are local stakeholders interested in publishing this data to a statewide database?
• What are your general thoughts, questions, or concerns about crowdsourcing human services data?
We’ve made it easy to add this information to a publicly accessible forum using GitHub Issues. If your county already is listed, simply click on the discussion thread and add your comments. If your county is not listed, create a new issue with the name of your county as the title. Please note that you will need to first create a GitHub account, which is a quick and painless process. If you prefer not to use GitHub, you can also drop us a line at health@codeforsacramento.org.
This commentary originally was published on Medium.
