This post has been getting a huge amount of search-engine traffic recently. Thank you for the visits! If you’re looking for more data on the dangers (or lack thereof) of cell phones in cars, check out this post as well.
Regular readers of this blog will know that the New York Times is waging a sensationalistic campaign to ban the use of cell phones while driving.
To this end, the Times has run at least seven stories, editorialized, released a movie and even produced a (not very realistic, from my experiences) texting-while-driving computer game as part of its ongoing Driven to Distraction series.
A casual reading of the series would doubtless leave most people with the impression that our roads are getting more dangerous and that cell phones are likely responsible. But this is not the case.
The truth is that our roads have never been safer.
Yes, you read that correctly. Despite the dramatic proliferation of cell phones over the last decade—more than 80 percent of Americans now use them—driving in America is safer than it has ever been.
The graph below shows the decline of fatal car crashes and the increase in cell phone use over the last fifteen years. I created it with accident data from the National Highway Traffic Safety Administration and the Pew Center for People and the Press. Traffic fatalities are measured in deaths per 100 million miles driven.
Does this mean that cell phones are actually making us better drivers?
Probably not. Vehicle fatality rates have steadily declined since cars were first widely available in the 1920s. But the graph does show that cell phones are not making are roads dramatically more dangerous. And they are certainly not the kind of threat that warrants a three month crusade from the New York Times.
This does not mean that cell phones are not distracting and potentially dangerous. They are. But as a 2006 study from the University of Michigan’s Transportation Research Institute suggests, they are no more dangerous than many other distractions, like passengers, grooming, and eating. Perhaps we should ban carpooling instead. After all, it could save more lives than banning cell phones.
80% percent of all rear end collisions (the most frequent vehicle accident) are caused by driver inattention, following too closely, external distraction (talking on cell phones, shaving, applying makeup, fiddling with the radio or CD player, kids, texting, etc.) and poor judgement.
It’s almost impossible to avoid the rear end collision so I went out and got one of these sparebumper.com
Nice graph, hadn’t seen that yet! I want to see text messages sent per year vs traffic collisions… I bet there’s a similar trend.
Interesting. How do the overall (not just fatal) accident rates compare to cell phone usage? Could it be that cell phone usage is causing an increase in the number of accidents, and the steadily improving safety technologies (seat belts required, air bags required, roll bars in many vehicles, anti-lock brakes, etc.) are reducing the number of fatalities in those accidents at a higher rate?
If you look at the graph going back to 1920, you’ll see that cars are steadily getting safer. This has been more or less constant, except in the 1960s, when they briefly got more dangerous. What cell phones do not seem to be doing is slowing the trend at all. If they were really such a threat, you would expect the rate of fatalities to level off or increase. But that is not the case.
I’m not sure you would expect the level of fatalities to level off. The last 15 years have seen a VERY large increase in the ability of vehicles to keep their passengers from dying when they do get into accidents, as compared to previous periods (from a pure physics standpoint), as well as accident avoidance features like ABS and traction control. So not only are you less likely to die if you do something stupid, it is actually physically harder TO do something stupid. Also, we have regulated other things that contribute to accident and fatality rates, like the number of hours that a trucker must spend not driving on a given day. I suppose the main point here is that there is just too much noise in the raw data to draw meaningful conclusions. Hence the simulator studies that places like NHTSA commission. Are these done purely scientifically? I doubt it, but in a system this open to other influences, it is extremely difficult to make claims about the dangers of one particular behavior, based solely on the raw data.
I agree with you.
A lot of the studies I’ve seen measure eyeball data and reaction time, but do so in closed courses. Is someone working for Car and Driver trying to type out a specific message while driving around a closed course emblematic of how most Americans use cell phones while driving? I doubt it.
As you said, driving is a really complex activity and it is very hard to to measure what potential distractions are the most dangerous.
It is also true that cars have gotten a lot safer as older models have left the fleet. If the fatality rate levels off, then maybe cell phones are a big hazard. But another possibility is that people talking on cell phones are simply substituting one distraction for another. I.e. they are talking on their cell phone instead of flipping through their CD catalog or fumbling around with mix tapes.
Do you think that the decrease in fatal car crashes has to coincide with the increase in car safety and production. Cars made in 1994 are not as equip or safe as those created now. Get your facts right.
Thanks for your comment and for reading!
As I said, cars have gotten safer and safer every year since about 1920. What you’re suggesting, I believe, is that the reason fewer people are dying from car crashes despite the fact that more and more of us are distracted by cell phones is that the cars we are in are safer. That is, we have more accidents, but fewer people die due to modern safety features.
That certainly sounds plausible, and seatbelts and airbags and the like have surely saved tens of thousands of lives. But the problem with this argument is that since 1996–about the time cell phones became widespread–we’ve not only had fewer fatalities per million miles driven, but fewer car accidents as well. Not just slightly fewer crashes, dramatically fewer crashes. According to data from the NHTSA, there were 6.7 million car crashes in 1996, and only about 5.9 million in 2006.
First, take a look at this interesting graph:
I don’t see any statistical argument here. Yes, crashes and fatalities have gone down as the number of cell phone users has gone up, but there’s no causation, let alone strong correlation shown here. In order to show strong correlation, you would have to construct a model with multiple factors that could affect accident rates, and moreover base that model on multiple countries. You could then determine the importance of each factor – you would probably actually find that cell phone usage is strongly correlated with higher crash rates; this would be the case if other, similar countries with lower cell phone usage had lower crash rates.
Here’s the graph I meant:
http://en.wikipedia.org/wiki/File:64uqp.jpg
That is a great graph. Thanks for the link and I hope you’re enjoying the blog!
My point in all of my posts on texting / cell phone bans is that there is no real science behind it. Sure the NYT is pushing it and it seems reasonable to assume that cell phones are a cause of great distraction to drivers and thus should be banned. One might assume they lead to more deaths or crashes, but that is just an assumption.
If you read through the articles supporting cell phone bans, you’ll find they are either based on anecdote, total BS–like claims that cell phones account for tens of thousands of fatalities–or “research” like eyeball tracking studies done in labs that are so far removed from the real driving experience as to be worthless. Also missing from the debate on banning cell phones is any discussion of the tremendous economic value that talking in the car contributes to the economy.
My point is that there is no real science or statistics behind banning cell phones–only emotion. On a side note, I have a great graph of the decline of carpooling corresponding to a drop in fatalities. I’m still writing up the post on that one.
Well, didn’t you think this data reflects number of drivers who used the cell while driving? Yes, # of the fatal car crashes dropped, but the fault of cellphone owners raised. Not true?