From the National Safety Council Defensive Driving CBT I’m in the middle of taking as part of our corporate safety training:
Which is the first mental ability impaired by alcohol?
- Judgment
- Coordination
- Knowledge
- Alertness
The answer is “A,” Judgment.
I thought that was kind of interesting.
31 view(s)
In a later screen, you’re asked to enter the names of five people who are most important to you. Then the stats come up that two of five people in the US will be involved in an alcohol/drug-related traffic collision at some time in their lives, and two of the names end up being scratched out.
Sorry about that, Mom & Margie.
It’s a rather effective scare tactic.
What kind of frelling stats are they using?
Not one person in my family (extended even) been involved in a drug/alcohol related collision or or died.
I guess this training is proving your WIST of “43.7% of all statistics are meaningless”.
I think I would be in fussy/mocking mode in this training.
Using Stan Rule of Rocky Flats Stupid Class/Training #1:
“If the person conducting the training isn’t considering a career change by the end of the class, you have not done your job correctly.”
Alas, it’s CBT, and thus yelling at the computer is not all that useful.
Taking X incidents and dividing them by Y population is a fairly gross statistical method (since it assumes an even distribution, which, as you point out, is unlikely in the small subsets of the population that we know). That aside, I don’t have much else to criticize about that factoid.
Most of the course so far has been okay — a bit repetitive at times, a it overly cute in coming up with rules/guidelines/mantras. But it’s been a lot less objectionable than some training I’ve gone through.
They have probably not taken into account that some people (esp. heavy drinkers) will have multiple crashes.
“Statistics are no substitute for judgment.” – Henry Clay
Ah. Good spot, dear.
Looking at the NSC’s latest bulletin, in 2004 the death rate from traffic fatalities was 15/100,000. Assuming a constant rate, over the course of 70 years, that would work out to 1,050/100,000, or about 1% of the people you know being killed in a traffic fatality.
Looking at the injury rates in 2002, the stat given in the course is a little lower: “About 3 in every 10 Americans will be involved in an alcohol-related traffic accident at some time in their lives.”
Looking at their web page on drinking and accidents:
In 2002, 17,419 people died in alcohol-related crashes. Alcohol was a contributing factor in 40% of all motor vehicle fatalities last year. It slows reaction time, decreases awareness, and impairs judgment.
* Two out of every five people will be involved in an alcohol-related vehicle crash in their lifetime.
* Young people are particularly at risk. Every day, 13 people between the ages of 16 and 24 die in an alcohol-related crash.
* The National Safety Council estimates that alcohol-related motor vehicle crashes cost America $34.1 billion in 2002.
The source of the stats is not clear. Though if alcohol is a contributing factor of 40% of traffic fatalities, that works out to (if I did the math above correctly) less than 1 out of every 200 people you know across your lifetime. Non-fatal crashes would obviously be higher.
Okay, looking here, there’s the statement (emphasis mine):
More than 43,000 Americans are killed in traffic accidents every year. However, some of these accidents might have been prevented if people were aware of the dangers of impaired driving caused by many common prescription and over-the-counter drugs. […] While there is currently no database in place to track the incidence of accidents caused by drug impaired driving, NSC statistics show that traffic accidents lead to more than 43,000 fatalities in the U.S. each year. The NSC hopes to work toward decreasing the number of people killed in car accidents each year by raising awareness of the dangers of drug impaired driving.
But based on that, where the heck does the 2/5 or 3/10 number come from?
Okay, more copious stats over at the MADD site:
In 2002, an estimated 17,419 people died in alcohol–related traffic crashes—an average of one every 30 minutes. These deaths constitute 41 percent of the 42,815 total traffic fatalities. (NHTSA, 2003)
Incidence of intoxication (BAC of 0.08 g/dl or greater) for drivers in fatal crashes in 2003 was highest for motorcycle operators (29 percent) and lowest for drivers of large trucks (1 percent). The incidence of intoxication for drivers of light trucks and passenger car drivers was the same (22 percent). (NHTSA, 2004)
[…] About three in every ten Americans will be involved in an alcohol-related crash at some time in their lives. (NHTSA, 2001)
In 2001, more than half a million people were injured in crashes where police reported that alcohol was present — an average of one person injured approximately every minute. (Blincoe, Seay et al., 2002)
Unfortunately, the one stat we want is cited but unlinked.
Going over the the NHTSA leads me in turn to the National Center for Statistics and Analysis. This is the best report I could find quickly, and even it doesn’t state things quite that boldly (nor go into much detail on accidents, vs. fatal accidents). In 2002 or 2003, “An estimated 275,000 persons were injured in crashes where police reported that alcohol
was present — an average of one person injured approximately every 2 minutes.”