That was my interpretation of the goal of the paper. But without knowing the quantity of driving before 9pm vs. after 9pm, it’s difficult to compare. There aren’t many accidents at 4am, for example, not because 4am is a uniquely safe time to drive, but because not many people are driving at that hour in order to have accidents.

In other words, the time distribution of phone usage and driving may not match. A disproportionately lower number of the post-9pm calls may be occurring from the car; the calls are being made in higher volumes, but not being from the same locations. The data doesn’t tell us, either way, so the story isn’t complete.

If I understood the study correctly, it was an attempt to deal with the causality issue. Someone may have a cell phone to their ear, and get in an accident, yet using the cell phone may not have been the cause. A sharp increase in use after 9pm should result in an increase in accidents after 9pm (other distractions being equal)

You might’ve had a chance at making a valid point if you knew what the point was. BTW, it’s not off by as much as you thought: http://www.avert.org/worldstats.htm

If you’re still confused, I can also use other axis labels like # cars or mile driven, and you can argue they don’t kill people.

-

You can’t possibly be correct and simultaneously see real world data results like this.

Maybe next you can argue why additional data isn’t important because, come on, the numbers are obvious. The more cars in the world, the less total fatalities*!

*note: per quantity population, and not absolute. Don’t want to get misconstrued again…

If AIDS had proliferated through the population at the rate that cell phone usage has, then you’d have a valid point. But it hasn’t, so you don’t.

Anyone who has a knowledge of research methods can see why there is a problem with this huge contradiction between the phone penetration rate and fatality rate. Your arguments aren’t logical, given the data conflicts. You can’t possibly be correct and simultaneously see real world data results like this.

So if I make a similar graph for my disease example, I guess that’s good enough for you?

-
Thanks for posting that article, I hadn’t seen this one before.

Yeah, think tank quality standards. I don’t think their “consistency” on issues is a virtue.

I see people every day driving slower because they are on the phone.

California’s traffic jams may be solved by blocking cell use on the freeways..

Cell phone use way way up. Accidents down. Reality trumps theory.

Thanks for posting that article, I hadn’t seen this one before.

It has gotta stick under Agenthex’s craw that the point that I’ve made continuously throughout this thread, the point that he dismissed as “specious logic”, is specifically cited in this paper as one of the problems with the anti-phone agenda, and is quantified in Figure 1 toward the front of the paper. Funny how that happens.

The paper did a nice job of summarizing a lot of the research on the topic and commenting on the flaws. But the free-minutes-as-benchmark argument is a questionable approach to take for making the point. Traffic studies typically measure fatality and accident rates based upon vehicle miles traveled, and this one doesn’t do that.

I’m just waiting for PCH to comment on your study because I’m sure he cares if the calls are made (or not in this case) in cars.

Explain the mystery and complexity.

Cell phone use way way up. Accidents down. Reality trumps theory.

http://aei-brookings.org/admin/authorpdfs/redirect-safely.php?fname=../pdffiles/WP07-15_topost.pdf

Press release/synopsis of the above study

http://www.berkeley.edu/news/media/releases/2007/08/13_cellphone.shtml

-because the world is not as simple as you’d like to assume it to be.

Cell phone use by drivers is up – astronomically.

Accidents are down, for quite a number of years, including those years in which cell-phone use has increased markedly.

Cell phone use while driving is not the danger it is popularly supposed to be.

When this thread started, I was among the phone-nazis, wanting them banned from all motorways. I am able to admit being wrong.

It would be if the instance rate is high OR correlation is low. It’s very unlikely that such high correlation occurs just in these relatively exceptional events.

This is not trivial to do because it’s hard to catch in the act, but at least these guys try. There also another “problem” in drawing a policy decision from this because they only study accident cases (tho that’s even harder to do broadly), but then I’m not passing this off as authoritative.

I don’t see how you can slam this and try to make conclusive remarks with even worse research.

Once again – their “correlation” is trumped by casting a wide circle around the definition of what constitutes a phone-related accident. Their definition is so expansive that numerous false positives are a given.

The use of assumptions like that is questionable. That makes it appear not to be a study, but a bogus setup that was written long before the “research” was conducted. The definition had to be contorted in order to get the desired result.

The fact that you have to continually defend obviously half-assed research, instead of finding better research, suggests that you either aren’t familiar with the research enough to find a better example, or else there are no better examples to offer.

If you wanted to derive some support from that study, it should be that cell phone use during driving isn’t extremely common in that part of the word. And it’s quite likely because of the nature of distraction vs. disorientation in the intoxication case the accidents are not serious enough to warrant a law.

Those are logical arguments instead of distracting with macro trends (which don’t even support your case since non-alcohol related is on the rise, unless you think they’re drunk AND on the phone, lol) or conflating effectiveness of law vs actual phenomena.

The researchers obviously felt the need to “count” incidents that weren’t incidents. Using a phone ten minutes prior to an accident was reason enough to include it as a contributory event.

By that logic, I could use my mobile phone in my home, terminate the call, then get in the car, back into something in the driveway, and have the researcher include that as a phone-related incident. That obviously doesn’t make any sense.

That erroneous methodology should tell you that the data is shaky. If the phone isn’t being used, then there is no reason to include it…unless, of course, the legitimate data doesn’t support your argument.

I know that you don’t want to see it, because it sucks for the advocates that they can’t make the real world gel with their theory. Using teenagers as a proxy for society as a whole, counting incidents that aren’t incidents, etc., is all very dodgy “science.” You need to do better.

Otherwise, the alternative you seem to prefer is to just trust the driver.

Copy and paste from above since you’re still dodging the main point:

[but] the data here is far from complete, and that study at least tries for correct methodology.

Again, you (as you often do) ignore the statements being made.

I read that study that you cited ages ago. It was flawed then, and it remains flawed now.

The specific problem with it is obvious — it seeks to blame accidents on phones even during instances when the phones were not in use.

The problem with that methodoloy is really, really obvious. Using that “logic”, we would attribute accidents caused by sober drivers to drunk driving, on the basis that they partied it up the week prior.

There needs to be a demonstration of correlation, and then some connection to causation in the real world. There isn’t any. If that study is the best you can do, then go back to the drawing board, because I am not going to continually rehash what is so obviously wrong with it.

Also, in case you haven’t notice, the auto world haven’t exactly stood still with the introduction of the cellular, and that accident rate you keep harping on hasn’t been much in decline lately.

Abstracted, the goal is to reduce serious distractions while driving. One major problem with the phone laws is they don’t fulfill that goal since they still allow hands-free operation.

This seems bit of an unenforceable dilemma, so it may not matter either way.

Once again, you’ve missed the point and gotten your facts wrong in the process.

There is a long-run trend of statistical data for vehicle fatalities and accidents. The overall trend is one of a gradual decline in fatalities when measured on a per mile basis, the standard used for developing these statistics.

The anti-phone brigade has alleged that phone usage is akin to drunk driving. Drunk driving is the main killer of people on our highways.

If adding tens of millions of phone users to our highways over the last several years was truly the equivalent of adding tens of millions of drunk drivers to the road, then this falling trend in fatalities should have been interrupted. After all, this would be the statistical equivalent of turning the United States into a nation of raving car-happy alcoholics over a short period of time, so the impact on the data should be obvious and hard to miss.

But the data that should have been disrupted hasn’t been interrupted at all. The trend of declining fatality rates continues. Not a logical outcome, given the alleged severity of the input.

Nor do we see any impact when anti-phone laws are introduced. Again, if anti-phone laws were so important, we would expect the introduction of these laws to have some sort of statistical effect, so that there is a noticeable benefit compared to not having such laws. But this hasn’t happened, either — anti-phone laws are not accompanied by improvements in these trends.

So here we have a situation in which the real world results don’t change when laws are introduced or avoided. We also have studies that have to be blatantly manipulated in order to produce the desired result.

This is exactly the same sort of situation that we had with the elimination of the 55 mph speed limit. We can all see now how the anti-speed Chicken Littles were wrong on every point. They also used poorly prepared studies based upon faulty assumptions, which failed to produce the real-world outcomes that they should have created had they been correct.

If simulator data conflicts with the real world, as it does here, then the problem is with the simulator, not the real world.

That’s rather the point of the study, evidence and whatnot.

On the other hand, the typical nay-sayer study trusts that the driver would report this accurately.

The effect of call can very well effect subsequent behavior. But whatever, I don’t have access to it now, but if the data is segregable into during and not, you’ll likely get a slight lower figure.

-
There should be blood in the streets from all of these additional accidents caused by cellular intoxication, yet we can’t find them.

Again, maybe because these accidents aren’t fatal, and afterward people tend to forget about actions that fault themselves.

If drunk driving stats depended on volunteered info, I’m pretty sure they’d be lower, too.

If you want to argue that there are real world problems, then these problems should show up in real world data. There should be blood in the streets from all of these additional accidents caused by cellular intoxication, yet we can’t find them.

Just as the speed limit increases caused no measurable catastrophes, no disaster is evident here. Likewise, passing laws against phones shows no blip in the figures that suggest that the laws do any good. Instead, we have studies like yours, which attribute accidents to phones even though the phones weren’t in use.

–
To be clear, I’m not necessarily for the law because I don’t generally care for the nanny state, but the data here is far from complete, and that study at least tries for correct methodology.

Yeah, I liked that one. That was the study that attributed accidents to mobile phone usage even though the phone wasn’t being used at the time of the accident.

Yeah that’s funny because there are reputable studies that contradict what you claim:

“Role of cellular phones in motor vehicle crashes resulting in hospital attendance” by S. McEvoy et al. published in the British Medial Journal.

And ironically the link above which does EXACTLY the analysis you claim would show you to be correct.

Deniers must prefer the kinds of studies that depend on surveys where drivers have to admit fault. I guess they’ll believe in anything as long as it supports them.

Do you think the studies really catch this?