Slashdot Mirror
Utah State Prof Says Hybrids Don't Kill More Pedestrians
thecarchik writes "Preliminary data seemed to show that hybrids were more likely to be involved in pedestrian crashes or hit cyclists. But now EV enthusiast Mark Larson (he's also an Emeritus Professor of Spanish at Utah State University) has analyzed some additional data and found this not to be the case at all. He used 1994-2008 figures from the Fatality Reporting System maintained by the NHTSA and found that the rate of pedestrian fatalities has in fact fallen over that same period."
Do you have any idea what welding ramming spikes to my Prius will do to my mileage?
The summary fails to mention that the liked study only focuses on blind people. So blind pedestrians are no more likely to get hit by a hybrid then full sighted pedestrians.
"Maybe this world is another planet's hell"
Aldous Huxley
How about data analysis by someone who isn't an "enthusiast" and by someone who is qualified?
"...rate of pedestrian fatalities has in fact fallen over that same period" yes, we've been designing pedestrian safe bumpers and hoods in that period, cross walks are safer with better lights and audible warnings.
As someone who was clipped by a Prius in a parking lot when it was on battery, the damned things are quiet as hell and sneak up on you like a ICE powered automobile doesn't.
Well there's your problem right there. You can't identify the contribution due to hybrids by looking at the total. There are on the order of what, 100 million vehicles on the road, and maybe 1% of them are hybrids. So if pedestrian kills by the other 99% of vehicles drop by 1%, hybrids could be 99 times more deadly than them and you wouldn't notice from this guy's analysis.
What he said.
[Sir Garlon] is the marvellest knight that is now living, for he destroyeth many good knights, for he goeth invisible.
In other news my Biology teacher thinks that Beowulf is a Shakespearean play. Something tells me that a Spanish teacher isn't an expert on analyzing statistics, hence why he teaches Spanish, not math or science.
Taxation is legalized theft, no more, no less.
Preliminary data seemed to show that Al-Qaeda were violent terrorists. But now AQ enthusiast Mark Larson (he's also an Emeritus Professor of Spanish at Utah State University) has analyzed some additional data and found this not be the case at all. He used 1988-2001 crime data from the Uniform Crime Report and found that the murder rate in fact fell over that same period.
It's not a problem with the cars, it's with dumbshits who don't look where they fucking walk. I ride a bicycle everywhere and it makes zero noise even when freewheeling, not that it matters I can ding my bell til my thumb falls off and many won't hear it because of the ipod craze.
People step out on the road in front of me all the time, maybe not realizing the speed I'm moving at when they last looked up the road, but it's still their fault. There are two ways to deal with this problem: Screaming at the top of your lungs at pedestrians "get of da focken road jackass!" or alternatively, pedestrians can take responsibility for their own personal safety and look with their fucking eyeballs.
POKE 36879,8
What's most astonishing about this is that the linked article states that Larson's analysis has two problems. The only way I can figure you'd stop at two is that one and two are the only numbers you know. Or perhaps more astonishing is the fact that nowhere in this list of flaws did the author of the article see fit to point out that this is a completely meaningless analysis. Instead the author of this article, who obviously has even less experience analyzing and undertanding data than this Larson fellow, focused on two very peripheral and arbitrarily chosen points. If you want to see this kind of analysis done right, visit http://www.venganza.org/about/open-letter/.
For the benefit of the exceptionally clueless, let me just point out that this article failed to mention the most obvious and devastating flaw with this kind of analysis -- the critical assumption that no other factor could possibly have influenced pedestrian fatalities since 2000.
It concluded that hybrids like the Toyota Prius were involved in pedestrian crashes at a rate of 0.9 percent, half again as high as the 0.6 percent rate for gasoline vehicles. Hybrids were also twice as likely to have hit cyclists, at a rate of 0.6 percent versus 0.3 percent.
Okay, this is pretty clear - the original study.
But now EV enthusiast Mark Larson (he's also an Emeritus Professor of Spanish at Utah State University) has analyzed some additional data. He used 1994-2008 figures from the Fatality Reporting System maintained by the NHTSA. If silent hybrid vehicles posed a threat to pedestrians, he reasoned, then the number of pedestrian deaths should have risen since 2000, when the first hybrids were sold. There are now roughly 1 million hybrid-electric vehicles among the 300 million on U.S. roads. But in fact, despite increasing numbers of hybrids on the roads, the rate of pedestrian fatalities has in fact fallen over that same period.
Wait, what? There's kind of a gaping hole here folks... But reading on shows that this objective and reputable news site has some doubts of their own as to Larson's methodology. Phew.
We like Larson's analysis,
Yeah, I'll just bet you do.
but we would observe that it has two problems.
Oh? Pray tell...
First, it doesn't factor in Vehicle Miles Traveled, which is correlated with a fall in accident deaths.
Okay, sure.
Second, Larson really only addresses half the issue. Fatalities from accidents are one data point, but injuries would be another--and are far more common than deaths.
Oh, yeah -- you nailed it exactly! Oh, wait - no, no you didn't. My bad, it was a typo -- I meant to write "you just completely ignored the glaring hole in the methodology applied by this professor of Spanish Studies". You can see how I made such a mistake, can't you? It could have happened to anyone.
Here it is, because I have to say it even though it's pretty flippin' obvious: In spite of the fact that OVERALL accidents are going down, the percentage of accidents caused by EVs is higher than non-EVs -- and when you consider that EVs still make up a very minor portion of the vehicles on the road, that's a pretty disturbing trend. Or how about the premise of his "report": The overall fatalities have decreased, and the number of EVs on the road has increased -- therefore EVs clearly do not pose any additional threat over their louder counterparts.
Oh, wait, here's what happened: Nice "reporting" greencarreports.com. I am duly impressed.