Diving in Chicago big data: who are the remaining passengers in the COVID times?
November 6th is a sunny Friday.
It is more than 9 months since the first COVID-19 case was officially reported in the U.S.
With the daily number of new cases sliding below the line of control, the time of the greatest panic and helplessness seems to fade gradually into a thing of the past.
Armed with masks and sanitizers, commuters returned to work, students started to attend hybrid classes, bars and restaurants reopened for business.
Nevertheless, this is not the whole story.
According to data provided by City of Chicago, the number of trips taken by residents decreased violently by 80.0% during the first two months of the pandemic outbreak. Though kept recovering in the following months, it looks like its peak would never reach even 40% of the previous level.
In facing COVID, people are more alert and precautious than reported, at least in Chicago. They do take care of themselves by cutting down the number of trips.
They know that the crisis is not completely over.
The same trend can be also observed from multiple perspectives.
A sharp decline in average trip seconds coexist with a steep increase in the average trip distance, indicating that drivers are driving faster than before. The only reason accounting for this should be a decreased number of cars on the roads.
Also, the average dollars of payment increased slightly, with an average pool count rapidly dropped to a point very close to 1. These two changes together depict the fact that travelers would rather spend more money, and even offer more tips, than take a cheap-but-hazardous ride by car-pooling.
Also, to see from the spatial distribution of the attributes of trips, we could gain similar deduction that people were indeed getting more careful, and took fewer trips during the pandemic times.
However, despite the fact that a majority of Chicago residents behave conservatively, there are still a large number of people who have to travel regardless of the COVID severity.
Put it another way, even though there are less than 40% number of trips taken in a recent month, an accumulative number of nearly 40% could already be daunting. Still there are so many people contributing to this percentage.
Well, who are they? Why did they make such choice to travel so frequently?
By the first impression, we might firstly associate this situation with some socio-economic indicators, such as average household income, poverty rate and unemployment rate. Therefore, I imported the public health statistic dataset provided by City of Chicago, and cross-analyzed these variables with April 2020’s and September 2020’s aggregated number of trips. The unit of anlaysis was community, which was in accordance with the unit of the public health data.
The reason why I selected April and September’s data to study is because these two months reflect two different phase of the pandemic. April was the first breakout time of COVID-19, when the nation started to panic at the pandemic. The most evident changes and turning points all took place in that month. September, however, was the latest month that have thorough data to refer to in Transportation Network Providers’(TNP) dataset. Comparing these two months’ subset can provide an overview on the changes of residents’ attitude and choices before and after COVID-19 getting controlled.
To see from the correlation testing table, however, we fail to recognize a bunch of relationships between most socio-economic factors with the number of trips taken in Chicago, especially when using April’s trip number as dependent variable. The only independent variable that showed statistical significance was income per capita, while it displayed a positive correlation with the number of trips taken. That is to say, even in the pandemic period, the higher income people could gain, the higher probability he/she would choose to take a trip. This correlation, however, was kind of being weakened at the starting point of the pandemic breakout, but it slowly regained its significance during the following months of recovery.
One interesting finding in comparing the Chicago trip data of April 2020 and September 2020, surprisingly, resides in a huge difference in time-travel pattern — both in terms of a week and in regard to a day.
To study this difference, I split the data into a few time-ranging groups, distinguished weekdays from weekends, morning from afternoon and evening, and also, pick-up locations from drop-off locations (in case they appear to have different distribution pattern). The result shows that in April, Chicago residents tended to travel during weekday nights, while in the weekends, they would not travel at any time. Conversely, in September, trips in weekdays were taken almost evenly in terms of morning, afternoon and night, while in weekends, the number of trip suddenly become very large in most Chicago communities. Maybe the time-travel pattern is becoming less abnormal as COVID panic fades, but to judge it from a broad picture, the total number of trips still remain lower than past years.
Thanks for reading.