A father-daughter exchange, from Lisa the Skeptic: Lisa: Excuse me, I took a piece of this skeleton for scientific analysis. Soon you will have all the facts. Homer: Facts are meaningless. You can use facts to prove anything that's even remotely true!
Jerry Weinberger: So convenient a thing it is to be a reasonable creature, since it enables one to find a reason for every thing one has a mind to do.
Rebecca Brock: People say to me, "Whatever it takes." I tell them, It's going to take everything.
Craig Friebolin: Fatal Crashes vs. Cell Phone Subscribers from 1994 to 2008 Total Crashes: 0.9% Decrease
Fatal Crashes: 6.2% Decrease
Cellphone use: 1,262.4% Increase
Decius: See also - why cellphones don't cause brain cancer.
Friebolin's two data points are misleading in the way they underplay the story on the use of mobile devices. 1. The US population has increased from 1994 to 2008, from 263M to 304M. 2. There are more vehicles per capita on the road in 2008 than in 1994. 3. Total miles driven increased substantially from 1994 to 2008. At a glance, it looks like roughly a 25% increase. 4. Average commute times are rising even faster than total miles driven. In Atlanta, the average commute time is 127 minutes per round trip. 5. The automobile accident rate has declined steadily since the 1920s. (Friebolin's "0.9% decrease" is actually quite misleading because it compares total counts rather than rates. See below ...) 6. Although the absolute number of subscribers has increased (as shown above), per-subscriber activity levels have risen at a vastly greater rate. You'll find growth from 44B MOU in 1996 to 1.68T in 2008 and 2.25T in 2011, according to CTIA. Likewise, texting has gone from negligible levels in 1994, to 33M in 2001, to nearly 200B in 2011. Read those numbers again. That's a 51x increase in voice MOUs from 1994-2011, and a 6000x increase in texts just in the last ten years. Of course, none of these data points are particularly meaningful for the purpose of answering whether a cellphone ban would accelerate the long-term decline in the accident rate. There are way too many variables in this equation to draw such conclusions from the available data. Decius: They aren't talking to the data any more because it's not clear that the data justifies these reactionary bans.
That's true, but I don't think the above-cited data serves to refute the (untested) hypothesis that safety could be improved by a broader ban on mobile device use. It's an entirely separate question as to whether those safety benefits, if in fact they exist, would outweigh the costs. Advocates of a ban are clearly cherry-picking examples, but deflating those bad examples (alone) doesn't disprove the hypothesis that a ban would improve safety. Advocates and critics are talking past each other. Critics say "it's not worth it" while advocates say, "it couldn't hurt." They're asking and answering different questions. The best argument for maintaining the status quo is that according to the available data (see above), despite having talked 50 times more, texted 6,000 times more, and driven 25% more, accident rates have continued their steady decline over the past 15+ years. If Friebolin's "0.9% decrease" were actually a valid measure of the recent trend in vehicular safety (hint: it's not), then that figure alone might be sufficient to consider the ban, because it would mean that the decades-long trend had flattened out. And if true, it might be reasonable to suspect that mobile devices might play a role. As the NHTSA's data shows, fatalities per 100M vehicle miles traveled have declined from 1.73 in 1994 to 1.14 in 2009. Which is to say that safety continues to improve despite the now-pervasive use of mobile devices. But it's hard to know whether a total ban in 2008 would have led to a rate of 1.13 in 2009. Would that have been worth it? If it had dropped to 1.00 in 2009, meaning that the ban had doubled the year-over-year safety improvement, would that have been worth it? RE: Cellphone use versus crash statistics. | 
