Engagement, iteration, credibility: What makes a successful data platform?
A new report highlights the US COVID Atlas
The US COVID Atlas was prominently featured in a new report that examines the effectiveness of data platforms designed for social impact. The report, Designing Data Platforms for Action & Influence: Lessons Learned from a Case Study of Five Data Platforms, aimed to understand the most successful ways to communicate complex data about the social determinants of health. Research was led by the USC Price Center for Social Innovation in partnership with the USC Annenberg School for Communication & Journalism, with support from the Robert Wood Johnson Foundation.
In recent years there has been an explosion in the development of data platforms, dashboards, and visualization tools, particularly in the wake of the COVID-19 pandemic. You’re here, so you’re likely already familiar with the US COVID Atlas, but maybe you’ve also checked the CDC or your city or county’s COVID dashboard, or other data platforms hosted by news outlets or community interest groups. There is no shortage of data to dig into.
The new USC/RWJF report highlights the different roles that different data platforms play but also new findings in successful stakeholder engagement, journalist insights, and data communication through media. It focuses on evaluating five data platforms: The US COVID Atlas, John Hopkins Coronavirus Resource Center, City Health Dashboard, Neighborhood Data for Social Change, and the Opportunity Atlas.
For us, the US COVID Atlas aims to fill a specific role in the wide array of available data platforms: We embrace the inherent uncertainty in data, and encourage people to look at the narratives being told (or not being told) by the data in different ways .We multiple data from multiple sources, offer multiple visualization options for comparison, and integrate spatial statistics for hotspot analyses. We are also in the process of adding an oral histories layer to breathe new life into the data with first-hand stories and lived experiences to the map.
In the report’s key findings around Stakeholder Engagement, researchers found that the five data platforms’ stakeholder engagement strategies fell on a spectrum from static engagement to dynamic engagement. The US COVID Atlas was identified as one of the most dynamic platforms:
The US COVID Atlas . . . has incorporated feedback and made significant design changes to the platform to better serve the needs of its users. Since its initial launch, their team developed a protocol to get ongoing feedback from users and even expanded its intended audience beyond healthcare network groups to also include journalists, researchers, and the greater public as a result of informal inquiries they received. They described the Atlas’ approach as “never quite finished…. there is always something that needs to be refined and reconsidered.” Ultimately, the COVID Atlas re-launched a “Version 2” in an effort to refine and add new features that had been requested by users.
Other important findings in the report were around the value of the quality and quantity of data available to users.
Across all user interviews, three features emerged as highly desired in a data platform:
- Data available in real-time
- Data disaggregated by subpopulation
- Data available at a granular levelUsers and platform developers alike stated that they would like platforms to have the ability to disaggregate data by subpopulation. This may include datasets broken down by demographic characteristics such as race, ethnicity, or gender — which can help expose inequity in health outcomes and resource allocation.
The Atlas has always updated in near real-time, adding new COVID data from our four main sources daily, and remains free and open source, with all code and raw data accessible on our GitHub. We’ve recently been focused on adding data disaggregated by sub-population to dig into the vast inequities we’ve seen in COVID impacts, but this has been tricky at the national level with COVID-19 data. While the CDC has not yet released county-level case, mortality, or vaccination data disaggregated by race and ethnicity or age, we have thought about how to better use the data that we do have on where people live, and how we can use that to proximate COVID trends. Check out the dot density mapping feature to visualize where people of different race or ethnicity communities live, overlayed on county-level COVID data. The overlay features, including Native American reservations, southern Black Belt counties, and hypersegregated cities also help identify places and populations disproportionately affected.
Lastly, the report also highlighted that iterative design that meets the needs of users is generally important.
While every platform in this study utilizes interactive maps to display data, some users stated that more simple visualizations, such as infographics, can effectively communicate essential insights as well. Engaging users in the design process can help platforms develop visualizations that effectively tell compelling data stories and share data. Testing and tweaking visual designs in response to user feedback can be a time intensive process, but users and platform leaders agreed that it is worthwhile.
We’re excited to be working on new tutorials based on feedback we’ve gotten from users via existing relationships, email, and social media in recent months, as well as engaging our new Community Advisory Board for their feedback and guidance on our work. We’ll also be hosting virtual feedback sessions in the coming year, so stay tuned for more to come on that front.
Overall, this new report lays out useful lessons for the Atlas as well as other data platforms interested serving community data needs. Check out the other data platforms featured in the report; each is engaging with and presenting data in unique ways. We’re grateful to the USC Price Center for Social Innovation and USC Annenberg School for Communication & Journalism for including us in this report, and look forward to integrating these findings with the Atlas’ work to further equitable health outcomes across communities and regions.
— -
We love hearing from Atlas users, so please feel free to get in touch or shoot us a message (@covid_atlas) with any questions, collaborations, or feedback.
