STUDY: Hybrids more likely to be involved in crashes with pedestrians, bicycles
A recent study by the National Highway Traffic Safety Administration finds that hybrid vehicles are more likely to be involved in accidents with pedestrians and bicyclists under certain scenarios. According to state-level accident data, 77 of 8,387 hybrid vehicles (that's .9 percent) were involved in crashes with pedestrians and 48 (.6 percent) were found to have been in accidents with bicyclists.
By way of comparison, 3,578 of 559,703 non-hybrid vehicles (.6 percent) were involved in pedestrian accidents and 1,862 (.3 percent) were involved with bicyclists. Tellingly, the NHTSA's data shows that hybrid vehicles are twice as likely than non-hybirds to be involved in pedestrian or bicyclist accidents at low speeds when the internal combustion engine is not running.
These statistics are not a complete representation of all accidents nationwide, and the NHTSA is quick to point out that additional research is necessary before any final conclusions can be made. Still, this is valuable data that "should serve as a guide when designing future HEV pedestrian and bicyclist crash prevention programs."
[Source: NHTSA - PDF via Consumer Reports]
Reader Comments (Page 1 of 1)
Snoopy 7:02PM (11/08/2009)
Lotus and Harman should have some fun with this.
Reply
jpm 7:03PM (11/08/2009)
I wonder how much this inconclusive "duh, who would have thought..." study cost the taxpayers?
If you ask me, the benefits from hyrbrids/EVs will easily eclipse the trivial BS that studies like this point out.
Reply
Rick 7:12PM (11/08/2009)
quiet cars, distracted drivers and nutty cyclists running stop signs - what could possibly go wrong?
Reply
AutoCritical 8:20AM (11/09/2009)
I;ve always found this as an interesting dilemma.
People should still be looking around I guess! - Looks like some time for public re-education!
Rick 7:13PM (11/08/2009)
maybe someone should put this dollar girl thing into some kind of spam filter
Reply
Joeviocoe 7:19PM (11/08/2009)
I noticed a strange statistic. When I visited southern California (I'm from Florida), the number of Prius' on the road increased 5000% and the number of cyclists and pedestrians went up too.
Since the Prius is essentially a cultural statement for a more natural society, doesn't it make sense that it would be driven more in neighborhoods with a lot of joggers, cyclists and pedestrians???
At low speeds, even modern gasoline engines noise is silent compared to the MUCH louder sound of tires on the pavement. Last I checked, the Prius was not a hovercar.... yet!
Reply
polo 8:09PM (11/08/2009)
Exactly. Who woulda thought that a car that appeals to "green culture" would be affiliated with incidents more likely to occur in cities that embrace said "green culture" with bike and walk trails.
Next thing you know they'll do a study that says off-road vehicles more likely to need a car wash.
nrb 9:36PM (11/08/2009)
77 accidents is hardly statistically significant enough to draw any kind of meaningful conclusion. The real margin for error (not the BS margins they like to quote) is probably larger than the entire conclusion.
Reply
Ben 10:01PM (11/08/2009)
Joeviocoe: great call!
As a cyclist and driver my first question would be who is at fault. I'm all for running over those idiot cyclists who ride around with their ipods cranked up, assuming that anything big enough to hurt them will make noise. My bike is quieter than a Prius, and if they get hit by a Prius rather than by me on my bike, then I regard that as a good thing. If they're not using their eyes, they're endangering us all: let them die.
Where to go from here? If the NHTSA cared about safety, they'd ban SUVs and cellphones. Most everything else they look into is noise.
Reply
john 10:55PM (11/08/2009)
Although the study seems to indicate hybrids are more likely to be involved in crashes with pedestrians and bicyclists, the margin of error on this study is way too close. When such a small number of hybrids are studied, a difference of two or three cars could throw the percentage off quite a bit.
So while I would agree that the study shows clearly that hybrids are more likely to be involved in crashes with pedestrians and cyclists - I will have to take the actual percentage figures with a grain of salt...and I don't want this to be an excuse to put big noise makers on hybrids! Furthermore, I'd like to see a study compare some quiet luxury cars (some of which are just about as quiet as a hybrid) with some of the louder cars, and see how the results compare.
Reply
Matt234 11:02PM (11/08/2009)
This was not an expensive study. There is one author, and it probably took a couple of months to analyze and write up, at most. Most of the work was probably in translating data from different states into one common dataset.
Either way, I am skeptical (if not a bit annoyed), and I will not buy the results as stated. I am not an epidemiologist... but let's pick this study apart. I don't think you need to be an expert to see some of the inherent flaws in the study. (warning, rant coming up.)
Let's get started. First of all (in itself a fatal flaw) it is an observational study. That means you have to adjust for everything under the sun to have any chance of finding the true answer. But this study makes use of very limited information. For example, it does not mention adjusting for age, gender, if it is an urban/rural accident, the car's rear visibility, and so on. Presumably there are a gazillion other unmeasured confounders.
Adjusting for very few things is a "cheap" way to get statistical significance if you have too few events. In simple terms, it just means you are really sure about the few things that went into the calculations. It does *not* mean that you know everything you need to be very sure. But it looks like you are really right. Looks great to most news reporters, especially if presented while wearing a white lab coat (just kidding).
The list of problems goes on. From the paper "The Honda Insight was considered but excluded from the analysis due to the fact that the earlier model years of Honda Insight always operate using their internal combustion engines, even at low speed" They then go on to include several cars as "hybrids" when they are not able to drive on electric alone (e.g. Honda). If removed from the "cases", the results would again probably lose statistical significance. And supposedly they included the Corolla Hybrid in the "case" group. Huh? Since when was there a Corolla Hybrid? I'm guessing they don't know that much about cars.
My conclusion is that unless they actually do a randomized trial it will never be possible to prove anything one way or the other. A randomized trial will never happen, of course, due to the practicality and cost.
Alternatively, electric/hybrid cars would have to get *a lot* more common so the number of events are bigger. In 10-15 years we might be able to know based on observational data. But please be aware that I just made that number of years up.
Finally, I'm a bit disappointed in my favorite green autoblog. Where does this sentence come from? "Tellingly, the NHTSA's data shows that hybrid vehicles are twice as likely than non-hybirds to be involved in pedestrian or bicyclist accidents at low speeds when the internal combustion engine is not running". Tellingly? Twice as likely? The study doesn't say that. Did you guys just make that up?
And about using this paper as a basis for policy? Don't get me started.
(end of rant)
Ironically, what the study is trying to prove might actually be true - I just don't think this study does a good job of proving it. I guess they did the best job they could with what they had... anyway, it should not have been published as is. They should certainly not have suggested that it should be used to influence policy.
Reply
Nick 12:03AM (11/09/2009)
I've got a study:
The toxic emissions of all ICE cars kill more people than all Priuses combined.
Reply
Paul 1:00AM (11/09/2009)
I wonder what the age profile of the pedestrians and cyclists who get run into by Hybrids in EV mode look like?
From my experience of years spent driving around CBDs (central business districts) most children use more caution when crossing the road than many adults who are so arrogant as to 'dare' you to hit them. At low speed it IS possible in most cases for drivers to brake to avoid stupid pedestrians, but playing 'chicken' is obviously a risky strategy.
These old dogs will have to learn new tricks, ie LOOK before you cross like school kids are taught to do.
Reply
Matt234 8:17AM (11/09/2009)
Um... I think when they say "visually impaired", they referred to blind folks, not people who dress badly.
Mark 2:59AM (11/09/2009)
Accidents with bikes and pedestrians are more likely in urban areas. Hybrids are more likely to be found in urban areas.
Whilst it's still worth considering the dangers* of near-silent running, the correlation isn't surprising.
* It's not dangerous to run silently of course. What's dangerous is crossing the road without looking.
Reply
Tohe 3:44AM (11/09/2009)
Once they take all those combustion driven vehicles out of our streets we might be able to hear the quieter ones. Or hey maybe we could just use our eyes and a little common sense.
Reply
625k.inc 12:01AM (11/20/2009)
You' all do realize that HR 734 and S 841 are not limited to hybrids and electric cars, right? Read section (3), (8) and you'll notice is not limited to just hybrids. It is a minimum noise standard for all cars.
I'm beginning to rethink my opposition. If we can just make the streets loud enough ...
Bob Wilson
Reply