New Study Shows As More People Talk While Driving, Accidents Are Dropping
from the counterintuitive dept
For a long time, we've questioned the conventional wisdom when it comes to laws banning mobile phone use in cars. It's not that we didn't think it was dangerous, but the bans themselves seemed not just too targeted, but likely to have unintended consequences. For example, last year, we noted a study showing that states that banned texting while driving saw an increase in accidents, likely because people kept on texting, they just did it by keeping the phone lower down to "hide it," taking their eyes further off the road.But as for talking while driving, we've all heard the stories about studies saying that drivers who talk while driving are as bad, if not worse, than drunk drivers. However, some new empirical research is calling that into question. johnjac points us to some new research that showed as more people used phones while driving, the number of accidents actually went down. Of course, this is just a (negative) correlation, and there are obviously lots of confounding factors, so I wouldn't (as the link above does) claim that driving while yakking makes the roads safer.
However, another part of the study does dig into this even more. It got access to mobile phone data, which allowed them to separate out those who were driving (based on changing towers often), and compared the number of phone calls just before 9pm and just after 9pm, when many mobile phone operators had call fees drop. They found that the number of calls jumped, but studying the data of car accidents right before and right after 9pm, they found no evidence of an increase.
The economists who did the study suggested a few possible explanations:
People who start talking while driving become more cautious. People who act like jackholes behind the wheel with a cellphone will act the same without one. And although cellphones clearly distract some drivers, they may also help other drivers stay alert.Frankly, there may be some other explanations as well -- starting with the fact that at 9pm, there isn't likely to be that much traffic on the roads. I wonder what a similar study would show closer to rush hour (if there were such a natural experiment where there might be a sudden jump in calls). If the roads are less congested, then the "danger" is probably much lower. Also, just because a phone is moving, it doesn't mean the person holding it is driving -- they could be passengers. So I'm not exactly ready to agree that this report really says what the authors claim, but it certainly is an interesting study to look at.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: correlation, talking while driving
Reader Comments
Subscribe: RSS
View by: Time | Thread
Oh god, what horrible methodology. Last time I looked, the average car has 4 adult sized seats, some up to 7. It isn't only drivers who have phones. Perhaps their passengers start to talk on the phone after 9 as well, when it gets cheaper.
There is no way to connect the number of "calls on the move" to the number of talking drivers. You can guess at it, but there is no way to be certain.
I could say that the number of calls after 9 is a combination of lower rates and younger people finishing working at their retail jobs, many of which close around that time. Are they in a car, or on a city bus? Who knows?
More non-science used by TD to try to prove something. "we have already shown that..." is coming next.
[ link to this | view in thread ]
BS
[ link to this | view in thread ]
Re:
If you read the rest of the article Mike does point this out.
[ link to this | view in thread ]
Re:
Wow, you need trolling classes. It's supposed to be something at least passingly logical....
[ link to this | view in thread ]
"More Cautious"? Not necessarily
[ link to this | view in thread ]
Driver training
We haven't seen an increase in police accidents as they use new technology have we? Perhaps because they are specifically trained to deal with it? (and when to just not do it)
Adding new technology, without TRAINING, is going to cause problems.
Another example, red light cameras. Nobody trained any drivers on how to react to them and so people slammed on their brakes when they 'remembered' there might be a camera. The drivers behind them aren't expecting someone to slam their brakes on and so accidents happen.
In some of those cases, perhaps they were driving too closely, but that too is a training issue isn't it?
[ link to this | view in thread ]
Drunk drivers claim this too
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re:
For instance consider the percentage of cars that have only the driver in them (Around where I live that is a very large percentage most of the time) so if there are a lot of people talking on moving cell phones, it is more than reasonable to assume that some of them must be talking while driving.
Also, it is pretty bad form to merely criticize something without making a suggestion of how to improve. You say the methodology is bad, but I say it is incredibly unethical to hold a controlled experiment that may put peoples lives in danger. The only methodologies that are ethical are to use data that is currently available and draw conclusions from them that are appropriate for the confidence of the data. It is reasonable to assume that a large jump in the number of calls would correspond to a jump in the number of calls made by people who are driving. I wouldn't dare try to claim that a 100% jump in call volume corresponds to a 100% jump in yakking while driving, but I am very confident claiming that there would be some jump. Throwing out reasonably good data because it isn't perfect is simply childish. In the real world you will never get perfect data.
[ link to this | view in thread ]
As pointed out by Mike, someone else in the car could be making a call, so it's rather on weak legs to start off with. But, if people are using a headset to talk on their phone it would still show a cellphone jumping from tower to tower.
It's rather crippling, for the actual argument that the study is trying to promote, but at the same time it is a good exercise in thinking critically about a study.
[ link to this | view in thread ]
Re: Re: Re:
[ link to this | view in thread ]
Seriously.
While this is one of those situations where its hard to keep people from doing it anyway, the consequences are not a few bucks lost by some major label.
People cant even walk and text + stay out of trouble.
http://www.metro.co.uk/weird/852862-girl-falling-into-fountain-while-texting-is-youtube- hit
[ link to this | view in thread ]
[ link to this | view in thread ]
Re:
How common place was texting 6 years ago, seriously? Let's pull out all the researches for the 80s, you know, back when cell phones were the size of a suitcase. They must have projects lots of accidents back then, considering your car roof wasn't high enough to talk and drive because of the antenna length.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re: BS
I see people veer over the median while talking to passengers. While eating. While smoking. While fiddling with the radio.
I would even argue that driving slower ought to mean they're actually driving safer, even if it inconveniences us.
I also agree with the point that bad drivers are still bad even without a cell phone to talk on.
[ link to this | view in thread ]
Re: "More Cautious"? Not necessarily
I believe they called that the Drunken-Kruger effect.
*rimshot*
[ link to this | view in thread ]
Re: Re: Re: Re:
They put something out there,and then later in the post edge away from it. He refers to it as "some new empirical research ", when in fact it isn't. The methodology sucks. It isn't empirical, it is bogus science. There are any number of things that could be affecting the number of calls, and the study ignores them.
Why cite a study if you don't agree with it? Why not be up front about it: "Seemingly Bogus Study Suggests More Phone Use By Drivers". Instead we get the absolute statement "New Study Shows As More People Talk While Driving".
It's complete BS, and I would figure someone as smart as you would see it. But then again, maybe not.
[ link to this | view in thread ]
Re: Seriously.
I am hard pressed to think of any kind of legislation that should not be "talked about".
[ link to this | view in thread ]
Re: BS
Ah, the scientific method at work.
[ link to this | view in thread ]
Re: Drunk drivers claim this too
Nothing in this study involved asking drivers if they thought they were safer when driving and talking.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re: Re: Re: Re: Re:
I would think anyone with half a brain, especially anyone who reads this site often, would know the difference by now.
Although, if you could read this whole post and still somehow believe Mike to be toting "new evidence that shows talking while driving makes the roads safer," I shouldn't really expect you to have half a brain.
[ link to this | view in thread ]
Re: Re: BS
[ link to this | view in thread ]
Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: BS
I think you miss the point entirely. Driving does not require 100% of your brain power. You can allow idle brain cycles to handle a discussion with someone, so long as driving is the priority.
Me, I think talking to passengers is more dangerous than talking on a cell phone. When talking to passengers, I see people's eyes leave the road so they can observe the non-verbal cues of the person they converse with. At least on a cell phone, the person's eyes stay on the road.
[ link to this | view in thread ]
Re: Re: "More Cautious"? Not necessarily
[ link to this | view in thread ]
Re: Re:
People changing cell towers? If I want from the front of my house to the back of my house, I can change cell towers. My neighbor walking their dog will change cell towers (and walks just after 9 pm, go figure). People walking down the street can change towers. There really isn't any good data here.
We also don't have how long the calls are for. We don't have any good reference for where this happened. We don't know traffic density, we don't know passenger counts (are there more passengers in cars after 9, example). We just don't know so many things that can change the outcome of this "study".
It is junk science, and rather than trying to disclaim after the fact, the story should start out by calling it junk science.
All this, but no coverage of Assange's claim that wikileaks is losing half a million a week because of the banks and paypal blocking him. I wonder why?
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
Re: Re: BS
Why do you not hold yourself up to the same standards that you hold others up to?
[ link to this | view in thread ]
Re: Re: Re: Re: Re:
In the world of "studies", the verb "shows" is just the word used to say what the outcome of that study was. It doesn't mean that outcome is true or correct - it doesn't mean "shows" as in "shows to be incontrovertibly true"
So for example, I could make the statement "This study shows X, but this other study shows the opposite" and it wouldn't be contradictory. I'm not using "shows" to endorse or confirm the results of either study - just to explain what the study itself purports to show with its data.
Moreover, I think you are confusing the two studies mentioned above. The one that uses the 9pm cutoff and the changing towers is, indeed, pretty questionable - just like Mike says. However the original study simply showed a negative correlation between phone ownership and car accidents, and that is a totally valid observation - it just isn't enough to draw causal conclusions from.
[ link to this | view in thread ]
Re: Re: BS
....amazing. Did you just come up with that? I foresee myself saying that to people on an almost daily basis.
[ link to this | view in thread ]
Re: Re: Re: BS
It seems like most people in the comments, and Mike himself, all agree with you that this study is highly questionable and probably fundamentally flawed. Perhaps if you had simply stated your reasons without turning it into an attack on Techdirt, we could all be having a great discussion analyzing this study and building our proof against it. Maybe we would even uncover some more flaws that you haven't noticed yet.
But no - you'd rather use this as a flimsy reason to bitch about Techdirt, even though most of Techdirt apparently agrees with you. Odd choice.
[ link to this | view in thread ]
Re: Seriously.
Forget security-by-obscurity folks, apparently we're moving on to safety-by-obscurity now.
[ link to this | view in thread ]
Re: Re: Re: Re: BS
Why run a story that starts out making it sound like the study is good and right, and then disclaim it at the end? If the study is bogus, why not start out with pointing out how bogus it is?
Queue the "Studies have already shown that driving and talking on cell phones decreases accidents" link fromanother story coming in about 30-45 days.
You seem smart Marcus. Why do you fall for this sort of crap?
[ link to this | view in thread ]
Drivolution
The only ones left are those who can multi-task.
Yay, we made it!
[ link to this | view in thread ]
Re: Re: Re:
[ link to this | view in thread ]
Re: Re: Re: BS
Wait, you're seriously comparing an actual study -- even with some flawed methodology -- to "well, this is what I see"?
That's beyond stupid.
[ link to this | view in thread ]
Re: Re: Drunk drivers claim this too
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
Re: Re: Re: BS
But yes, it is one of those things that I find myself saying on a daily basis...
[ link to this | view in thread ]
Re: Re: Re:
Do you? Please go and find a definition.
Oh, heck, I know you won't because you don't want to put that much effort into trolling.
"Empirical research is a way of gaining knowledge by means of direct observation or experience. It is used to answer empirical questions, which must be precisely defined and answerable with data."
This study uses pure numbers. No anecdotes, no 'psychological studies', (which, although they may be scientific, don't get me wrong, aren't empirical), nothing qualitative, just numbers.
Now, what part of that claims reasonable sources of error? In the scientific community, we publish and take on preliminary studies all the time. The data we want can be hard to get, (read: expensive, in time/money/both), and the best way to convince someone to give us what we need to do a proper study is to find/perform a preliminary study that suggests the world doesn't work quite the way we think it does.
That said, I can see why you might trash the study, because it does seem to have plenty of sources of error that you can see. And I agree, it's definitely not conclusive, but, like correlation, it does waggle its eyebrows suggestively while mouthing "look over here".
And besides all that, the point of the study was "There does not exist". When you consider sources of error, you should also consider the influence those errors might have on the data. The types of sources that Mike, (and you), mention, would reduce the impact, not eliminate it, (Less drivers on the road, passengers talking instead of drivers, etc.).
The study shows no significant increase.
Now, if you're capable of thinking, or if you have a strong, unhelpful bias, you might realize that doesn't mean the study is scot-free.
What about un-reported accidents? Maybe people on cellphones get into a lot more of those, but don't report it because there's no one around, or their insurance would freak and they're afraid the insurance might realize they're on the phone while driving. So there COULD be an increase, but it would be invisible.
Maybe, there's nearly no one who drives & uses a cellphone, (As opposed to Mike's point, where the ratio of drivers to passengers using cellphones would reduce the estimated # of cellphone drivers to 1/3 or 1/4).
And THEN, we can consider another thing that's kinda been mentioned. We do, of course, have a lot of studies that people who use cellphones while driving get in a lot more accidents. That people who do, are a lot more distracted, pay less attention to road, don't notice drivers around them, etc.
All those should be a pretty good reason to ignore this study, right?
Wrong. These studies don't necessarily contradict each other.
All those studies above indicate, but do not prove, that cellphones cause traffic accidents. What they prove is what they say they prove and no more: Statistically speaking, the population of drivers who talk on cellphones are more likely to get in an accident.
Correlation.
As DCX2 said:
"I also agree with the point that bad drivers are still bad even without a cell phone to talk on."
So bad drivers might use cellphones more often while driving, but the population of drivers are not significantly affected by using a cellphone.
TL;DR: Yes, the study doesn't mean much. No, it is not wrong, no, you should not throw it out.
What it means, is that closer research should be done into establishing or disproving a causal link between using a cellphone while driving & driving accidents.
Alternatively, we could play it safe and just ban cellphone use . . . nope, wait, we already have well established data that says when we do that, accidents increase because people try to hide their law-breaking rather than stopping potentially unsafe practices.
[ link to this | view in thread ]
Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re:
If you spend any time on this site, you would realize that one of the running themes is to scrutinize and question common assertions, especially those made by big business or policy makers. The reason he ran with this post was that it shows (regardless of the efficacy of the study) shows that the link between driving, distractions, and accidents is complex and therefore not easily legislated without significant unintended consequences.
[ link to this | view in thread ]
Re: Re: Re:
[ link to this | view in thread ]
Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: Re:
Anyone want to do an empirical study and publishing the results?
[ link to this | view in thread ]
Re: Re: Re:
We're going to need some empirical evidence for that.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re:
It is particularly key when you talk about something as involved as traffic accidents. Everything from road conditions, driver skill, car conditions, and about 1000 other factors can be at cause in accidents.
In the end, it's junk science. Drawing any conclusions from it is going to similar be junk, and writing a blog entry based on it, even if you are playing the game a bit, is still junk.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re:
Umm, so? It's still bad methodology, isn't it? You can make a study say almost anything you want by using bad methodology.
[ link to this | view in thread ]
Re: Driver training
Actually, there has been an increase.
Adding new technology, without TRAINING, is going to cause problems.
Like drinking and driving is okay if you just know how to do it. It takes practice.
[ link to this | view in thread ]
See, I Told You So!
[ link to this | view in thread ]
Re: Re: Re: Re:
That is part of the problem, it doesn't use pure numbers. It uses unqualified numbers, which is worse than using interview style studies. All we have is cell phones "in motion" with no way to attribute them to car drivers, car passengers, bus riders, dog walkers, or even segway users. We just don't know. Heck, it could be attributed to atmospheric conditions that may make users switch towers on the fringes of the service areas.
We just don't know.
Irrelevant data clouds and obscures whatever they are trying to prove, to the point where they prove nothing, except that they likely started with the answer and obtained data to backfill to meet it.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: BS
[ link to this | view in thread ]
Re: Re: Re: Re: Re: BS
In any case, I'm not "falling" for anything, and certainly not these questionable studies. I'm just questioning your strange need to turn everything into a rant about how Techdirt sucks instead of, y'know, discussing the topic at hand.
And I can pretty much guarantee the links you foresee won't be coming. Yes I know the type of backlink you mean, but your the wording of your example subtly pushes it over the line into something way more misleading than anything on TD, which is usually not misleading at all unless you are purposely playing dumb in order to find fault.
I really have to ask - is the only reason you are visit Techdirt that, in your view, it's a big delusional bubble that you just have to burst? Because no matter how vehemently you believe that, I just can't see what motivates you to keep coming back here... the web is a big place and there must be some community out there where you can actually contribute.
[ link to this | view in thread ]
Re: Re: Re: Drunk drivers claim this too
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re:
*swish*
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re:
Like Dark Helmet, I thought you were smarter than that.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: BS
I'm just questioning your strange need to turn everything into a rant about how Techdirt sucks instead of, y'know, discussing the topic at hand.
I am actually debating the topic at hand. Sadly, when you start with an incredibly flawed report, and try to present it in a manner that gets good search results and encourages people to read it because it says something outrageous, the discussion sort of ends. Even TD admits (somewhat grudgingly) that the report has some flaws. So the problem is that you cannot debate something when the material is flawed, and thus the conclusion is equally flawed.
it over the line into something way more misleading than anything on TD, which is usually not misleading at all unless you are purposely playing dumb in order to find fault.
Again, you miss the point: That is exactly what TD does in their posts. For that matter, it's the basis for Nina Paley's horrible little cartoons, but we won't go there. Intentionally leaving out data, intentionally ignoring facts, or presenting only narrowly selected facts are methods of turning a little lie into a big truth. It's the way this works, and one of the reason why I enjoy commenting on it. The logic is often so flawed and self-serving, that it is amusing to see who gets tripped up by it.
If I get one less person to be deluded, than I have done well. I look at someone like Average Joe, who obviously "gets it" and realizes that much of what passes for truth here is rather shaky. You guys did do a good job running Daryll out of town, he was right more often than anyone where would want to admit.
[ link to this | view in thread ]
Re: Re: BS
Driving slower does not equal driving safer. Someone moving slowly on a freeway, for example, is actually creating a hazard by moving too slowly.
[ link to this | view in thread ]
not accidents but traffic
The problem is, they don't cause accidents. They cause TRAFFIC.
Everytime I'm driving to or from work, and there's a cluster of slow traffic, it's always a driver driving in the left lane while talking on the phone.
Seems when people start talking on the phone, they "zone out" of their driving habits and start mimicing the driving of the person next to them.
So if you have someone driving at 40mph in a 60mph zone in the right lane, the cellphone-talking driver in the left lane is almost guaranteed to slow down to 40mph while talking.
Crab effect, anyone?
I'm betting this directly leads to a high increase in road rage by frustrated drivers left in their slow-wake.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
Despite what you insist, not DH or myself or any of us are complete idiots. We find this stuff interesting, we see no problem with the way TD reports it, and we don't at all agree with your bizarre and insistent condemnation of every single post here.
You keep saying you think we are smart - then how about you give our judgement a little more credit and stop for two fucking seconds to consider that YOU might not be 100% correct.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: BS
So let me lay this out very clearly for you: I am not missing your point. I get your point completely, and I think it is wrong. If you are going to insist that people must either agree with you or else they just don't understand, then you aren't going to get very far with these conversations.
If I get one less person to be deluded, than I have done well
I would consider starting with yourself.
You guys did do a good job running Daryll out of town, he was right more often than anyone where would want to admit.
Wow, really? That's your example? Darryl? With his weird stream-of-consciousness rants and his inability to move beyond extremely simple concepts, and his debate tactic of repeating the same things over and over again with more and more spelling mistakes and RANDOMLY capitalized WORDS each TIME? Suit yourself. I prefer a slightly higher standard of discourse.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
For the record I don't condemn every post. But damn, it's hard not to, considering how many of them are built on shaky premises, data tricks, and misleading headlines.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: BS
Darryl actually gets it much of the time, just really fails in communications. If you stopped laughing at his stuff for a minute and paid attention to it, you might glean something out of it.
You can disagree with me, but I don't feel I am wrong. TD has amazing skills for writing, amazing skills for being able to frame issues in such a way that relevant data isn't relevant. It's a true skill, and one that deserves attention. Snopes could run a whole section just debunking the junk that gets posted here from time to time.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: BS
But it's nice that you've finally accepted that I can disagree with you, instead of calling me stupid under the guise of calling me smart.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Hehe... it all comes back to this doesn't it? Get off the lawn kids! I don't like your music!
I have spent the better part of my life surrounded by talented musicians in multiple genres, all of them making tonnes and tonnes of music out of their homes. And a lot of it is excellent - both in terms of my subjective opinion and also in terms of the quality of songwriting, performance and recording on any "objective" scale you wish to measure.
But you have shown your true colours: you don't support art at all, you just want to be its gatekeeper so it won't scare and confuse you any more. All this stuff you go on and on about on this site is clearly all stemming from that simple issue: you can't handle change. What a pity.
[ link to this | view in thread ]
phone use has changed since '05 - so has technology...
My concern is that we shouldn't dilute the message with texting and tweeting while driver. This is killing people on our highways every week. Recent studies show that for every 6 seconds a driver spends texting, 4.6 of those seconds are with their eyes off the road, which makes texting the most dangerous cell phone activity anyone can engage in while operating a 5,000 pound piece of steel and glass. I also think that if technology is going to help, it should get the driver's eyes back on the road where they belong.
I decided to do something about distracted driving after my three year old daughter was nearly run down right in front of me by a texting driver last fall. Instead of an expensive shackle with recurring fees that locks down phones and alienates the user (especially teens), I built a tool called OTTER that is a simple GPS based, texting auto reply app for smartphones. It also silences call ringtones while driving unless you have a bluetooth enabled. Its an easy way to manage that text and drive temptation.
Erik Wood, owner
OTTER LLC
OTTER app
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Since you missed on the basics, the rest is the rest. Too bad you can't see past your nose.
[ link to this | view in thread ]
Re: phone use has changed since '05 - so has technology...
Oh no, tell me it isn't true! Next thing you know, you'll be trying to tell me that all those cigarette company sponsored studies showing the safety (or even health benefits) of smoking were biased. Well, I'm not believing a word of it!
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Considering our fundamental difference in outlook on the state of modern media, I doubt we are going to see eye to eye.
[ link to this | view in thread ]
Study...
[ link to this | view in thread ]
Re: Re: Re: Re:
It fails because they aren't observing the subjects in question, they are unable to say if their data set is "drivers", only that it is "phone users in motion". That isn't strict enough to create valid data. It would be like looking at 1000 people walk buy, taking only pictures of their feet, and stating that many women working downtown wear work boots. Because you didn't first subset the data to "women", and instead the majority of the people who walked by were workers headed for the construction project next door, you have invalid data. Technically, it would be empirical, but because it fails to assemble valid data, it really isn't.
It is truly a basic, clear, and simple failure. It renders the data meaningless. They aren't observing drivers, and as a result their information is faulty.
They don't even attempt (and cannot) in any way seperate out hands free users from "phone in hand" users. So there is no way to tell from this study if we are talking about drivers, and in those drivers, how many of them use hands free units (more and more common).
Junk science. I am not sure what TD's motivation is to run such a piece.
[ link to this | view in thread ]
[ link to this | view in thread ]
[ link to this | view in thread ]