So với lái xe, đi máy bay cực kỳ an toàn – nhưng tại sao lại không có cảm giác đó?

A commercial airplane crash can change how people behave long after the immediate tragedy. News outlets replay the images, people affected share their harrowing stories, and an extraordinarily rare event becomes easy to imagine. Some travelers cancel trips; others drive to their destination instead.

These choices may feel cautious. Statistically, the alternatives people switch to may come with greater risk.

Commercial airline crashes command attention because they are rare, catastrophic and beyond a passenger’s control. But the choice to avoid flying does not eliminate risk. As a new report from our research team at Embry-Riddle Aeronautical University outlines, skipping a flight and opting to drive instead is typically much riskier, even if it doesn’t feel like it.

Airline travel is safer than headlines suggest

People tend to respond strongly to events that are vivid, while paying less attention to familiar hazards – the risks of which manifest more frequently but less visibly. As a result, efforts to avoid one danger can create another.

In our recent peer–reviewed report, prepared with support from Embry-Riddle Aeronautical University’s Boeing Center for Aviation and Aerospace Safety, we compared fatality risk in U.S. commercial aviation with other transportation modes, recreational activities, household hazards and occupational risks. While the Boeing Center receives some funding from The Boeing Company, it operates independently.

We focused on commercial Part 121 airlines operating in the U.S. and analyzed data from 2016 through 2025. Part 121 carriers generally include large U.S. airlines and regional carriers that operate scheduled passenger service. Because fatal accidents are so rare, data from a single year would not have provided a stable risk estimate.

Across several measures of risk exposure, including distance traveled, individual boardings, hours spent traveling, annual risk and lifetime odds, the data consistently shows that U.S. commercial aviation is exceptionally safe.

For the same distance traveled, we estimated the fatality risk as about 626 times higher in a passenger car or truck than on a commercial airline. The risk associated with motorcycle travel was more than 22,000 times higher than commercial airline travel. Even highway bus travel, itself a relatively safe option, carried about nine times the risk of commercial airline travel over the same distance.

When risk is measured per passenger boarding, the numbers remain striking. From 2016 through 2024, the long–run estimate was approximately one passenger fatality per 97.7 million boardings on commercial airlines.

These numbers do not mean flying is risk–free, just that the risk is small enough that researchers need many years and enormous amounts of exposure to estimate it. Across the 10-year study period, 73 people died in accidents involving commercial airline travel in the United States.

These findings complement earlier research. A peer–reviewed 2013 study comparing U.S. transportation modes examined data from 2000 through 2009 and also identified commercial aviation as the safest way to travel. A 2018 study using data from the same period similarly ranked commercial aviation as the safest mode.

More recently, a 2022 study analyzed U.S. news coverage, travel patterns and accident data from 1996 through 2017. It found that increased news coverage of aviation accidents was associated with more road traffic and more fatal road accidents in subsequent months.

The 9/11 aftermath and the risk of avoiding risk

The aftermath of the Sept. 11, 2001, terrorist attacks provided a powerful example of how behavior can shift risk from one setting to another. Air travel declined sharply after the 9/11 attacks. Subsequently, researchers were able to investigate whether fear of flying led to additional deaths on the road.

Psychologist Gerd Gigerenzer analyzed traffic patterns for the final three months of 2001 and estimated 353 excess road deaths nationwide. He characterized many people’s decision to drive rather than fly as an attempt to avoid a “dread risk”: a low–probability event in which many people could die at once.

Transportation researchers Michael Sivak and Michael Flannagan later reanalyzed the period. They used road fatality trends from the first eight months of 2000 and 2001 to estimate how many deaths would normally have occurred during the final three months of 2001. They then compared that estimate with the number actually recorded in the past three months and calculated 1,018 additional road fatalities from October 2001 through December 2001.

But their findings complicated the idea that people who avoided flying merely drove to the same distant destinations. Much of the increase occurred on local roads, not interstate highways, and pedestrians and bicyclists experienced a disproportionate share of fatalities. This pattern suggests that travel behavior may have changed more broadly after the attacks, perhaps because people canceled longer trips, stayed closer to home or changed their daily routines.

It is hard to pin down how much of this increase was related to people changing their travel patterns in response to the 9/11 attacks, since these studies cannot track each individual’s transportation decisions.

But both studies suggest that a collective effort to avoid a highly visible danger can increase exposure to ordinary hazards whose victims are less visible. When more people choose to drive rather than fly, the roads become busier and the number of traffic deaths can rise.

The question shapes the risk estimate

One reason discussions about risk become confusing is that there is no single way to measure it. The answer to the question “Is it safe?” depends on what you measure.

How much risk does a person face for each mile traveled? What is the risk of one trip? How much risk accumulates during one hour of travel? What is the chance of dying from this cause in a year or over a lifetime?

These are legitimate questions, but to make a valid comparison, you need to look at the same measure of risk for both types of transportation. Comparing the risk of one flight with the annual risk of driving, for example, mixes unlike quantities.

A traveler deciding whether to fly or drive between two cities can compare the two modes over the same trip or distance. This comparison provides a clearer perspective than comparing a highly visible plane crash with a familiar daily routine.

Time matters too. In our estimates, an hour in a passenger car or truck carried about 74 times the fatality risk of an hour aboard a commercial aircraft.

Feeling in control is not the same as safety

Driving gives people a steering wheel, a brake pedal and the belief that their skill can protect them. Flying asks passengers to place control in the hands of pilots, airlines, regulators, air traffic controllers and a complex technical system. Psychological research has linked a stronger sense of agency – the feeling that people control their actions and their outcomes – with greater risk–taking. That difference in perceived control helps explain why a rare aviation disaster can feel more threatening than routine road travel.

Media attention also matters. A major airplane crash is treated as news of national or international importance. Road deaths occur across many locations and at different times, so the public does not see the cumulative toll.

The emotional intensity of coverage does not mean journalists are wrong to cover crashes. It just means audiences need context alongside the images.

Crafting better risk communication

Telling anxious travelers that flying is “safe” is rarely enough. Research on risk perception shows that people do not judge danger through statistics alone. Fear and vivid images also shape how risky an event feels. But providing more context about the risks from other forms of transportation can still help.

Effective communication about risk does not minimize tragedy or dismiss the real fear people may feel. But it can help people understand that when millions of travelers shift from one type of transportation to another, the risk moves with them.

For travelers, the practical question is not whether an option has zero risk. No option does. The question is whether a choice reduces their total exposure to risk or merely replaces a frightening, highly visible risk with a greater one that feels routine.

To put the rarity of fatal airline events into perspective, an average of 2.48 million travelers passed through TSA screening each day in 2025. Modern commercial aviation in the United States has achieved a safety record that is difficult to grasp precisely because fatal events are so unusual.

When fear leads people away from that system, the safest–feeling decision may not be the safest decision at all.

Mihhail Berezovski received support from Embry-Riddle Aeronautical University’s Boeing Center for Aviation and Aerospace Safety for the research discussed in this article. The center operates independently of The Boeing Company.

Alex Chaparro receives funding from Embry-Riddle Aeronautical University’s Boeing Center for Aviation and Aerospace Safety for the research discussed in this article. The center operates independently of The Boeing Company.