Recent Crash Test Video Footage on TV -- Rebuttal?

[BobDiaz]

UPDATE:

I came across something I missed that raises some question as to the accuracy of the information given in the report by the IIHS. Here's the link to the report:

http://www.iihs.org/externaldata/srdata/docs/sr4404.pdf

On page 4 of this report, you will see the following chart:



It looks reasonable, until you look up the numbers:



Now 15.5 deaths per 100,000 people would be multiplied by 10 to match the scale of the chart. So, it's 155 per 1,000,000 people. Now look closely at the chart, each car appears to be lower than 155 per million people. A reasonable chart would always put this number of deaths per 1,000,000 of a car type. That way the scale would always be the same for each car, but the numbers do NOT appear to match the data.

It appears that they took a total of 1,000,000 cars and looked at the deaths in the 1,000,000. Now if we had exactly 200,000 Mini cars, 200,000 small cars, 200,000 mid size cars, 200,000 large cars, and 200,000 very large cars, the scale is the same for all cars. HOWEVER, if I add the numbers for mini, small, mid, large and very large, I get a number close to the 155 per million. This tells me that the scale changes for each car type and the chart is 100% worthless. It tells us NOTHING!!!

An argument can be made that it only covers 1 to 3 years old cars in 2007, which are safer than older cars. However a look at the death rates over the years shows that the change is not that great.


Source:
FARS Encyclopedia

In 2007 the Fatalities per 100,000 vehicles was 16.05; that comes to 160.5 per 1,000,000. In 1994, the number would be 232.1 per million cars, a drop of 69.2% from 1994 to 2007. Safer, but not enough to match the chart given by the IIHS.


There more data to study, but I'd say that the "Study" from the IIHS is already showing some real flaws.


Bob Diaz

[forestacademy]

Bob Likes Charts

Are you a bean couter Bob?

[John_H]

Beans would be simpler to sift through. I never like that data like this is obscured. "Deaths per 100kpeople" is based on population or drivers+passengers or passenger miles or....

[24-42]

Originally Posted by jwight

We could take this same scenario back to the late 1940s (MG-TF), 1950s (Austin Mini and VW), 1960s (Corvair and Falcon), 1970s (Pinto and Vega), etc.

Nothing's changed except today's small cars have much more in the way of active and passive safety features.

Minor, no offence but in the interest of accuracy

Production MGTF: 1953-1955


Donald LaFavor

[jwight]

Originally Posted by 24-42

Minor, no offence but in the interest of accuracy

Production MGTF: 1953-1955


Donald LaFavor

Good catch; would you believe the TC?

[BobDiaz]

Originally Posted by forestacademy

Bob Likes Charts

Are you a bean couter Bob?

My background is Electrical Engineering and I currently teach Electronics, Computer, and Robotics Classes at El Camino College. So, I guess that's why I'm into the technical details and charts...


Bob Diaz