Friday, April 23, 2010

No Effect, no Cause II, The Sequel

Last year, I noted, that where there is no effect, cause is also absent:
Using a cell phone while driving is A Very Bad Thing. Studies clearly show the accident risk while texting is eight times greater than when keeping both eyes on the road, and all ten thumbs on the wheel. People talking on cell phones are four times more likely to cause a crash.

No doubt.

That undoubtfulness leads to a hypothesis: Since cell phone use while driving is risky, and cell phone usage has become increasingly common since the mid-1990s, then there must be an increase in accident rates over the period.

Turns out that isn't the case.

At the time, my main beef was that The Sudies had assumed their conclusion before so much as getting their analytical key into the ignition. The fact of a declining accident rate in the face of skyrocketing cell phone usage would seem worth something more than deep silence: by assessing risk without regard to actual consequences, they were trafficking in crimes against statistics.

Secondarily, and to some very well mannered derision, I surmised that cell phone usage simply does not cause enough accidents to be statistically noticeable.

As it happens, my original surmise might have been close to the mark:
A new study suggests laws banning the use of hand-held devices while driving have not reduced the rate of accidents in three states and the District of Columbia.

In addition to the nation's capital, the report by the Highway Loss Data Institute reviews insurance claims in New York, Connecticut and California. It also compares the data to other areas that do not have cell phone bans.

"The laws aren't reducing crashes, even though we know that such laws have reduced hand-held phone use, and several studies have established that phoning while driving increases crash risk," said Adrian Lund, president of the Highway Loss Data Institute.

There were no fluctuations in collision rates before and after the laws were put in place, the report said.

"So the new findings don't match what we already know about the risk of phoning and texting while driving," Lund added.

The reporter didn't say whether Lund's head exploded immediately afterwards.


Blogger erp said...

The next step in highway safety will be to ban drivers from chatting up passengers in the car.

Whatever statistics show, common sense determines that things like texting and dialing, not to mention putting on nail polish, that takes the driver's eyes off the road should be disallowed. ;-}

April 24, 2010 6:30 AM  
Blogger Hey Skipper said...

Yes, common sense says any distractions should be disallowed.

The larger point here, though, is either the idiocy, or mendacity, of the initial studies, and assessment of the follow-up.

In his series on statistics, David mentioned what should be the statisticians prime directive: where there is no correlation, there is no causation.

What has been demonstrated is that the impact of cell phone usage upon the car accident rate is indistinguishable from the null hypothesis.

The even larger point is how adherence to a pre-determined notion can be unshakable, despite clear, unequivocal, evidence to the contrary.

As with Warmenism. According to the theory, increased GHGs must increase the risk of severe weather.

But that correlation does not exist.

Which means GHGs increase the risk of severe weather.

April 25, 2010 1:23 PM  
Blogger erp said...

Of course there's a larger point. It's a given that learned "studies" are undertaken to support predetermined conclusions. That satisfies David's rule that first comes the theory, then come the data.

Not intuitive -- you betcha.

April 25, 2010 4:15 PM  
Blogger Hey Skipper said...

That satisfies David's rule that first comes the theory, then come the data.

Does it?

Auto mishap and cell phone data both existed before there was any theory linking the two.

Sometimes theory leads to data collection.

Sometimes existing data is used to confirm (or not) a theory.

This case was the latter.

Consequently, that is why I disagree theory must precede data.

April 25, 2010 5:48 PM  
Blogger erp said...

Sorry, I forgot s/off at the end of my last comment.

April 25, 2010 8:12 PM  
Blogger Ali said...


Have you got any comments on the ash cloud that hit European airspace a couple of weeks ago?

April 27, 2010 12:06 PM  

Post a Comment

<< Home