Visualising the aftermath of Covid-19 policies
Were they effective in curbing the virus’s spread?
In the digital ocean of COVID-19 dashboards, finding one that seamlessly intertwines state-specific policies with everyday case trends in the U.S. is like discovering a rare gem. That’s what we set out to create — a dashboard not just charting the pandemic’s course but also aligning it with the policies implemented across states. Why? To unearth the often elusive correlation between policy decisions and COVID-19 case fluctuations.
The Making of a Data Navigator
Our journey began with sourcing data from USA Facts and the COVID-19 US State Policy Database. After meticulous data processing and overcoming the tedious task of merging disparate datasets, we fashioned a dashboard that doesn’t just show numbers but lets the user see the effects of statewise policies for themselves. The dashboard vividly presents daily new cases alongside policy milestones for each U.S. state, from January 2020 to August 2023.
Deepening Our Understanding: Data Analysis and Visualization
Starting off with our analysis, we turned our attention to the subtleties within the data. It was not just about numbers; it was about understanding the narrative behind these numbers. Each data point represented a day in the life of the pandemic, a policy decision, or a change in the public’s behavior. This level of detail was crucial in crafting a dashboard that was not just informative but also insightful.
Technical Challenges and Solutions
One of the most significant technical challenges we faced was ensuring data accuracy and consistency. Working with large datasets meant that even a small margin of error could lead to significant misinterpretations. To mitigate this, we employed advanced data cleaning techniques and cross-referenced our findings with multiple reputable sources. This rigorous approach ensured that our dashboard reflected the most accurate and up-to-date information possible.
Another challenge was presenting this complex data in a user-friendly manner. We wanted our dashboard to be accessible not just to data scientists or public health experts but to anyone interested in understanding the pandemic’s trends. To achieve this, we focused on simplicity and clarity in our design. We used color coding to differentiate between various data points and implemented interactive elements like sliders and filters to make the exploration of data more engaging and intuitive.
From Design to Beta: A Pivot in Approach
Our initial project design envisioned dual choropleth maps to show COVID case numbers pre- and post-policy implementation. However, after insightful discussions with our professor, we realized this approach didn’t fully capture our project’s motivation. We pivoted towards a time-series analysis, aiming to directly correlate policy implementations with changes in daily COVID case numbers.
The Beta Release Dashboard
Our beta dashboard brought this vision to life. It featured a map chart at the top, displaying new COVID case numbers per state for selected dates. This was complemented by a dynamic line graph in the lower half, triggered by selecting a state on the map. This graph illustrated the daily case trends for the chosen state, with big colored circles marking key policy implementation dates.
Notably, the beta version included a date slider, allowing users to observe how case numbers evolved over time. This interactive element was crucial in showing the pandemic’s progression and the immediate impact of policy decisions.
Through the Users’ Lens: A Kaleidoscope of Insights
Building the beta dashboard involved significant data cleaning and transformation. We meticulously converted county-level total case counts into state-specific daily new case numbers, cross-verifying data with reliable sources. The policy data required similar attention, ensuring accurate representation of each state’s policy timeline. However, we decided to test our beta project with potential users of the dashboard before commiting to our design.
Our user sessions were a pivotal component of the project, offering a kaleidoscope of insights that were instrumental in refining our dashboard. These sessions revealed the multifaceted nature of user interaction with our tool. The invaluable feedback from our user evaluation sessions — involving students from economics, family therapy, and information management — illuminated several areas for improvement.
Diverse Perspectives and Challenges
Each user brought a unique perspective, informed by their academic and cultural backgrounds. This diversity was crucial in understanding the different ways users interacted with the dashboard. For instance, the student from economics, with a keen eye for trends and patterns, provided feedback on the dashboard’s ability to illustrate economic impacts of the pandemic. The family therapy student offered insights into how the dashboard could be used in understanding community and family-level impacts of the pandemic.
One common challenge faced by users was geographical familiarity. Users who were not well-versed with U.S. geography found it challenging to navigate the map and identify states. This feedback led us to incorporate more intuitive geographical representations and a state search function, ensuring the dashboard was accessible and user-friendly, regardless of the user’s familiarity with U.S. geography.
Enhancing Data Visualization and Interactivity
Another key insight was the need for clearer data visualization. Users indicated that while the data was rich and informative, it could be overwhelming. This led us to refine our visualization techniques, ensuring that the data was presented in a more digestible and engaging manner. We improved the visibility of policy markers and optimized the interactivity of the graphs, allowing users to easily draw correlations between policy implementations and case numbers.
Feedback-Driven Enhancements: Transforming Insights into Action
The feedback from these sessions was not only about identifying challenges but also about recognizing what worked well. Users appreciated the comprehensive nature of the data and the interactive elements like the date slider, which allowed them to explore the pandemic’s progression over time. This positive feedback encouraged us to maintain these features while working on the areas that needed improvement. We learned about the challenges faced by users unfamiliar with U.S. geography, the need for clearer data visualization, and the importance of an intuitive UI/UX design. These were our solutions to for the same:
Intuitive Geographic Representations: Adding a filter for state names to the map for easier identification.
Clearer Data Visualization: Improving the visibility of policy markers and data points on the graphs.
Enhanced UI/UX Design: Implementing a state search function and optimizing font colors for better readability.
Final Dashboard and Tutorial Integration: Guiding Users Through Discovery
Our final dashboard further refined these functionalities. We focused on enhancing user experience, ensuring the dashboard was not only informative but also intuitive and accessible. We incorporated these feedbacks in our final release, along with an added option to compare two different state trends- a feedback we recieved from our peer and the proffesor during our class presentation.
We also recognize the importance of guiding users in navigating our dashboard. To this end, we’re excited to embed a comprehensive video tutorial within our blog post. This tutorial will walk users through each feature of our dashboard, ensuring they can fully leverage its capabilities to uncover the hidden narratives within the data:
Conclusion: A Data Journey Worth Sharing
Our COVID-19 dashboard project is more than a mere aggregation of data; it’s a tool for understanding the intricate relationship between public policy and health outcomes in a crisis. As we continue to refine our dashboard, we remain dedicated to making data not only accessible but also meaningful. To acess our interactive dashboard, please follow this link to tableau public : https://public.tableau.com/app/profile/sharvil.shastri4139/viz/COVID19NumberofcasesperdayandPolicyImplementationforeachU_S_State/FinalDashboard?publish=yes
We plan to navigate further into the realms of data science, public policy, and user experience design. Our mission remains steadfast: to turn data into dialogue, numbers into narratives, and statistics into stories that resonate with and enlighten our audience.
Project done by me (Sadaf Davre) and Sharvil Shastri for the final project of our course, INST 760 : Data Visualisation under the guidance of our Proffesor Md Naimul Hoque.
References:
USA Facts. (2023). COVID-19 Spread Map. [Online]. Available at: https://usafacts.org/visualizations/coronavirus-covid-19-spread-map/
Boston School of Public Health. (2023). COVID-19 US State Policy Database. [Software]. GitHub. https://github.com/USCOVIDpolicy/COVID-19-US-State-Policy-Database/
