Revelation
Data Visualization Final Project | Spring 2022
My Role:
Data Analyst, Designer
Duration:
3 Weeks
Platform:
Tableau, Excel, Figma
Area:
Data Visualization
Background
In April, the omicron variant of COVID turned Shanghai, my hometown, into a deadly ghost town. People were forced to lock themselves indoors, medical resources became scarce, and those with life-threatening diseases were left to fend for themselves. The hopelessness I felt during that time was unbearable, but through my work in data visualization, I found a way to channel that frustration into something positive. I decided to use visualizations to examine the relationship between Zero-COVID policies and their impact on daily new cases, hoping to shed light on whether lockdowns were truly the best decision.
Fig. 2 Da Bai (people dressed in white coats) taking the road to give people mandatory COVID testing.
Fig. 1 Compound being enclosed by wooden
walls to restrict residents from leaving.
Topic Brainstorming
Initially, my intention was to focus solely on studying the COVID-19 outbreak trend and corresponding policies in Shanghai. However, after conducting further research, I realized that it would be insightful to compare Shanghai's situation to that of other cities in order to gain a more objective understanding of the current policies targeted towards the city.
I chose to compare Shanghai with Hong Kong as they have undergone a similar outbreak of Omicron virus, share similarities in government control (both being influenced by China's Zero-COVID policy), and have a similar population density (6,659 people per square kilometer in Hong Kong, and 3,925 people per square kilometer in Shanghai). Note that the population density plays a crucial role in the speed of virus transmission and both cities having a high population density it made them good candidates to be studied comparatively.
Data Collection
This is one of the most challenging part of the project. While data for Hong Kong can be readily accessed through Our World In Data, comprehensive data for Shanghai is scarce, due in part to the recent nature of the events. The only source available was from the official WeChat newsletter of Shanghai Municipal Center for Disease Control & Protection, which required manual input of data points into Excel. Additionally, the limited data available only includes the number of new daily confirmed cases and deaths. Information about other attributes such as the age group of confirmed cases, the percentage of vaccinated individuals among positive cases, and the daily testing population is not readily available. Furthermore, since I decided to include policies (text data) into consideration, I had to find a way group, categorize, and effectively represent it visually.
Regarding the policy, I decided to first read through the policy timeline of both cities, extract major/influential ones, and then combine with the trend of newly confirmed cases to categorize the dates into different stages.
Visualization Ideation
My goal is to present a comprehensive understanding towards the differences and similarities of the COVID policies in Shanghai and Hong Kong as well as their respective consequences. The following information will be depicted in the graph: 1) a comparison between the new daily cases in Shanghai and Hong Kong, 2) the policies implemented in each region, and 3) the progression of the virus (such as incubation, outbreak, and decreasing phases).
Here are two initial drafts I created:
Both drafts utilize a similar visual approach, which includes creating line graphs to depict daily new cases in Shanghai and Hong Kong, displaying relevant policy announcements with extended data points, and annotating the timeline with boxes indicating different stages of the virus.
However, upon further consideration, several issues with this design became apparent:
​
01. The inclusion of too much information and key points in a single graph may be visually overwhelming for audiences.
02. The differing timelines of the virus outbreak in Shanghai and Hong Kong make it challenging to accurately categorize the progress in each city using the same date axis, resulting in empty spaces on the graph for Shanghai in the first two months.
​
After additional iterations and brainstorming, I opted to keep the direct comparison between the two cities in a single graph, while displaying the remaining policy and stage-related information in separate graphs for each city.
The Result ↓
Reflection
1. Data can be limited, but your creativity cannot.
I recognize that the data aspect of my project may not be as robust as some others due to a lack of readily available datasets, limited information within the data sources, and the challenge of visually representing text-based data. Nevertheless, I am proud of the work I was able to accomplish. Even though many data sources are text-based, important insights can still be gleaned and effectively communicated through creative organizational techniques and the use of layering on existing graphs to create more informative and dynamic visuals.
2. Data visualization is about clarity and intuitivity. Use more graphs if needed!
Initially, I was disheartened when I discovered that a single graph could not effectively convey all the information I wanted to present. However, I came to understand that by breaking down the information into multiple graphs, the message could be more clearly communicated and information better organized.