CTA
Ridership
Changes

due to COVID-19

Vittorio Iocco

TK Kwon

Sheng Long

Connor Mayberry

Mara Ulloa






With the start of COVID-19, CTA "L" ridership plummeted

To provide some context:
On December 12, 2019, a cluster of patients in Wuhan, Hubei Providence, China begin to experience shortness of breath and fever.
On March 15, 2020, U.S. states begin to shut down to prevent the spread of COVID-19.

The drop was substantial, with a nearly 70% drop in average ridership from the previous years

But, this drop was not felt evenly...

Using data from the CTA and the American Community Survey, we mapped each station to the census tract that contains them, and obtained the scatterplot on the left.

The general trend here, highlighted by the regression line, is that we see this negative correlation between median income and drop in aggregate yearly ridership between 2019 and 2020.
As shown in the graph on the left, neighborhoods with higher income had a higher percentage drop in ridership. This could be caused by a number of factors, including:
1) Wealthier individuals opting to take private transportation to avoid COVID-19 exposure
2) Wealthier individuals have jobs that offer flexibility in remote work compared to lower income jobs.

Neighborhoods with higher education levels are also positively correlated with income, and could represent the same result as shown on the left.

Neighborhoods with a higher percentage of white individuals are also likely to be wealthier neighborhoods, and would similarly experience a drop in ridership as seen in the earlier graphs.

Recovery from 2020 to 2021 has been sluggish so far...


The data in the graph on the right shows the percentage increase in ridership between 2020 and 2021.
As you can see, there is no clear slope to the regression, and it seems that the effects of COVID-19 are still being felt in 2021.
Only 74 out of 143 stations maintained or increased aggregate yearly ridership.

Compared to 2020, 2021 has no discernable increase in ridership across groups with different education attainment.

Similarly, we see no correlation between rate of recovery in aggregate ridership and percentage of white population in the neighborhood.

Although there are no general trends, we can see some patterns already -- for example, there is cluster of green line stations that continued to have decreasing ridership.






Explore the data further

Choose X variable: