The Dangers of Commuting

Beckett Dumo
Fall 2023 — Information Expositions
7 min readDec 20, 2023
Photo by CHUTTERSNAP on Unsplash

Driving is a normal part of life for many around the world. It is also one of the most dangerous activities we do. Car-related injuries are the leading cause of death for people ages five through twenty-nine. Most people take for granted how dangerous driving is when they embark on a trip to work or school. Car crashes can happen to anyone at any time. No matter how skilled a driver they believe themselves to be, all it takes is one distracted driver.

In 2021, there were 4.4 million licensed drivers in Colorado. The population of Colorado in 2021 was 5.8 million, which means about 75% of Colorado was licensed to drive. Furthermore, 91% of 2021 Colorado drivers had reported driving distracted at least once a week. Half of the drivers surveyed also reported they drove distracted the majority of the time. Moreover, Darrell Lingk, Highway Safety Office director at CDOT, claims, “Distracted driving is one of the leading causes of crashes on Colorado roads.” It is a terrifying prospect that about half the drivers we share the road with in Colorado are distracted.

After reading about this statistic, I wished to explore Colorado driving deeper by looking at Boulder County’s open data. As a commuter in Colorado, I wished to learn more about the causes of car crashes and ways to avoid them. The dataset I used for this investigation into Colorado driving was Boulder County crashes mapped from CDOT data from 1999-2020. In addition to CDOT data, I supplemented it with 2021 census data on Colorado and 2022 Colorado Traffic station data.

I first approached this investigation by looking into which type of accident was most common.

The blank accident type is a head-on collision.

As shown in the bar graph above, rear ends are disproportionally more common than the other accidents. Distracted driving accounts for 87% of rear-ends. Rear-ends make up the most significant portion of accidents in Colorado. This makes sense after looking at the percentage of distracted drivers in Colorado. The second and third most prevalent accident types are broadside and approach turn. A broadside or t-bone happens most commonly at an intersection where one driver fails to yield to the other at a stop and drives into the side of the car. Broadsides are one of the most fatal car accidents because of how much weaker a car door is compared to the front or back of a car. Approach turn or left turn crashes happen when one driver waits to take a left and fails to yield to oncoming traffic. If you have driven in Boulder before, you know how daunting left turns can be. Both of these crashes are caused by distracted or confused driving.

Following accident types, I looked at the time column of the data and broke it into years, months, days, hours, and minutes. I then looked for patterns between hours of the day and crashes. This data is plotted below.

On the left is a time-series line graph of the sum of all crashes per hour of the day. The right is a bar graph of the same data but with more accuracy at each hour.

These graphs tell a straightforward story: the most dangerous time to drive is 5 pm. This makes sense with rush-hour times of 6–9 am and 4–7 pm. As shown in the charts, 8 am is the most common part of the morning commute for crashes, and 3–5 pm is the most dangerous time for the afternoon commute. I explored this deeper with a look at months and days.

While days show an indiscriminate story of crashes, with the apparent fall at 30 being because only seven months have 31 days, the months tell a chilling story. April is the safest month to drive, and October is the most dangerous. This could be due to many reasons, but it may be because of holidays. Holidays are a time that brings many people together from far away places, which means more drivers on the road. Moreover, holidays give many a reason to drink copiously. The other reason could be seasons. Fall and Winter months have the highest recorded crashes. Fall brings the rainy season, and Winter brings snow and ice, all dangerous driving conditions.

Approaching crash statistics from the holiday perspective, I looked at crashes involving impairment.

Time-series line graph of the sum of impairment involved in crashes with the month of the year.

The data flips the narrative on its head. May, August, and October are the most common months for crashes involving impairment. While this shuts down the argument that the holiday season is responsible for the crash increase, it still brings up intriguing data. Times when you would expect to see the most impaired crashes would be holidays like Christmas and Thanksgiving, but their respective month’s trend lowest on impaired crashes. In contrast, May reigns king in impaired crashes, albeit only averaging about 70 more than the low end of the data. Impairment crash data got me interested in other statistics involving drinking and cars, so I looked at hours of the day impaired crashes.

Time-series line graph of the sum of impairment involved in crashes with time of day.

The data tells a similar story to what you would expect: around 1 am or bar closing time is the most common time for impaired crashes. People trying to drive home from a late night out without considering the effects of alcohol on driving are endangering not only themselves but everyone else out on the road as well. The data also shares another story: drinking after work. There is a high spike in impaired crashes at 5 pm. This would suggest that many stop for a drink after work and continue their drive home. This dangerous habit is one that millions of Americans partake in.

Approaching crash statistics from the weather perspective, I looked at the weather conditions during crashes.

Boxen Catplots of the sum of injuries against the weather at the time of the crash

When looking at the crash and weather data, no apparent trends appeared, but when I changed crashes to injuries sustained in the crash, trends appeared. As seen above, around 50% of crashes in each weather condition ended with one injury. Moreover, snow and rain disproportionally have more crashes with injuries than the other weather types. Unknown means the weather type wasn’t reported at the crash. More than 80,000 entries in the data don’t have weather reported for it.

Delving deeper into weather data led me to the road condition column, which reported the road condition at the time of the crash and whether any treatment, like salt, had been applied to the road.

Boxen Catplots of the sum of injuries against the road conditions at the time of the crash

As seen, dry still reigns king for injuries during accidents due to the sample size of crashes. Furthermore, wet and icy reoccur as contributors to injury and severity of accidents. This is all to be expected, but the data also shows the effectiveness of road treatment. Note the last set of boxplots; roads with treatment trend far lower in injuries sustained than their untreated counterparts. This is a testament to the effectiveness of CDOT procedures during adverse weather.

Finally, for fun, I looked at county data and used the census data to find the county with the most injury-inducing crashes per capita.

Despite Weld and Jefferson counties being larger in population than Boulder, Boulder has the most severe accidents out of all the counties. This is most likely due to the increased population density in Boulder compared to the other counties.

The exploration of data uncovered many new ideas for me. The three most common types of accidents are related to being distracted on the road. Fall and Winter are the most dangerous times to drive. There are drivers on the road impaired at 5 pm. The dangers of weather and road conditions. All these discoveries have given me insight into the dangers of driving. While I was aware of the dangers of driving, I didn’t think the numbers would reveal such a straightforward story about the adages my parents have told me.

Distracted driving remains the biggest plague to safety on the road. The previously mentioned statistic of 91% of Colorado drivers driving distracted once a week should resonate with you. As the data proves, the dangers of driving distracted can’t be understated. Try to limit taking your eyes off the road by setting the phone down, changing stations at a stop, and eating when you get home. Furthermore, the weather conditions will only worsen as we delve deeper into the winter months. Try to avoid driving at times of inclement weather. If you must go out, lay off the gas and feather your breaks.

At the end of the day, every time you drive, you risk your life. Every driver has accepted the possibility and consequences that come with driving to enjoy the convenience. The problem lies in humans. No matter how hard you focus on preventing accidents, all it takes is one negligent person. Despite the harrowing statistics around driving, not everyone on the road with you will care. Taking steps to make sure you are driving cautiously and properly can help. You can lead by example, give a safe distance, signal, and drive attentively.

--

--