vrkelley

Weird stuff. Request for comment: https://www.sciam.com/article.cfm?cha...94A60E4D9A76B2

May 10, 2007
Strange but True: Helmets Attract Cars to Cyclists
Although you might not want to leave your protective gear at home, just know that if you do, drivers will be a lot more scared of hitting you.

Spring is in full swing now, and a number of the straphangers (read: subway riders) in New York City, as well as citizens in other locales, are getting new tubes and tires and dragging their bikes out of storage. Bicycle riding is the skill you reportedly never forget, but there's a raging debate about whether or not you should forget your helmet when you hop on your two-wheeler.

Last September a plucky psychologist at the University of Bath in England announced the results of a study in which he played both researcher and guinea pig. An avid cyclist, Ian Walker had heard several complaints from fellow riders that wearing a helmet seemed to result in bike riders receiving far less room to maneuver—effectively increasing the chances of an accident. So, Walker attached ultrasonic sensors to his bike and rode around Bath, allowing 2,300 vehicles to overtake him while he was either helmeted or naked-headed. In the process, he was actually contacted by a truck and a bus, both while helmeted—though, miraculously, he did not fall off his bike either time.

ADVERTISEMENT (article continues below)

His findings, published in the March 2007 issue of Accident Analysis & Prevention, state that when Walker wore a helmet drivers typically drove an average of 3.35 inches closer to his bike than when his noggin wasn't covered. But, if he wore a wig of long, brown locks—appearing to be a woman from behind—he was granted 2.2 inches more room to ride.

"The implication," Walker says, "is that any protection helmets give is canceled out by other mechanisms, such as riders possibly taking more risks and/or changes in how other road users behave towards cyclists." The extra leeway granted to him when he pretended to be a woman, he explains, could result from several factors, including drivers' perceptions that members of the fairer sex are less capable riders, more frail or just less frequent bikers than men.

Randy Swart, founder of the Bicycle Helmet Safety Institute (BHSI), says that studies such as Walker's run the risk of misleading cyclists as to the effectiveness of helmets. "The cars were giving him, on average, a very wide passing clearance already," he explains, noting that most vehicles typically stayed well over three feet from the bikes, rendering the 3.35-inch discrepancy to be insignificant. "If you really want the greatest passing distance, you should wobble down the road," looking as inept as possible, he adds.

Walker actually reanalyzed his data recently to counter this line of reasoning. "I assessed the number of vehicles coming within one meter [roughly 3.3 feet] of the rider, on the principle that these are the ones that pose a risk," he says. "There were 23 percent more vehicles within this one-meter danger zone when a helmet was worn, suggesting a real risk."

Dorothy Robinson, a patron of the Bicycle Helmet Research Foundation and a senior statistician at the University of New England in Armidale, Australia, published a 2006 review article in the BMJ (British Medical Journal) about regions in Australia, New Zealand and Canada that introduced legislation that spurred an over 40 percent increase in bicycle helmet use among their populaces. The newly instituted laws, she found, did not have a significant effect on bicycle accidents resulting in head injuries, the primary purpose of the gear. Her conclusion was "helmets are not designed for forces often encountered in collisions with motor vehicles" as well as that they "may encourage cyclists to take more risks or motorists to take less care when they encounter cyclists."

Coincidentally, around the same time as Walker announced his results, New York City released a report on bicycle deaths and injuries: 225 cyclists died between 1996 and 2005 on New York streets; 97 percent of them were not wearing helmets. Of these deaths, 58 percent are known to involve head injury, but the actual number could be as high as 80 percent. Comparing the helmet to a seat belt in a car, Swart of the BHSI says, "When you do have that crash, you better have it on."

Raiyn

Not this old chestnut again.

vrkelley

gah definately a candidate for a South Park episode...of course an anti-cyclist cager sent the article to me.

LittleBigMan

I don't see why helmet issues are taboo. It's already on everyone's mind.

I see it as an individual choice by adults, not a requirement. But although I'm not afraid to ride without a helmet, I continue to wear it for safety.

A helmet will not protect you from being crushed to death by a truck or car, but it can make a simple fall a non-event, instead of a concussion, or maybe worse.

(Still, I admit not wearing one is a nice experience, sort of like not wearing a condom--which I haven't done in a very, very long time...)

markhr

helmets don't kill people - people kill people

Bikepacker67

Strange but True: Helmets Attract Cars to Cyclists

Painful but True: Foreheads attract curbs.

I'll wear my helmet.

Raiyn

Because there's another current thread on the same topic on the same page.
https://www.bikeforums.net/showthread.php?t=297486

CB HI

Five or more threads on this study. It is starting down the path of the recycling impotence threads.

Bikepacker67

Boy, that joke went right over your head.

trickmilla

According to this story a helmet can actually prevent your head from being crushed.

From the article: "The truck ran over his head. "I didn't see it coming, but I sure felt it roll over my head. It feels really strange to have a truck run over your head." His helmet, a Giro, was crushed, but Lipscomb's head was fine."

trickmilla

On the article of this thread. I think it is important to point out that the science is totally schlocky and the conclusion that is drawn that wearing a helmet may increase danger just plain sloppy for a number of reasons:

1) cyclists have crashes that do not involve cars (mechanical failure, road conditions, operator malfunction)
2) some cyclists do not take more risks with helmets on
3) some accidents are caused by drivers that do not see cyclists at all
4) some bike/car crashes are intentional driver assaults
5) some bike crashes are caused by drivers that are drunk and or lose control of their vehicle

The researcher would have to support that enough fatal accidents are caused by drivers getting too comfortable and driving too close to cyclists to eclipse many other potential crash situations to support the above statement.
............................
I hope this doesn't devolve into another flame war. People need to make their own decisions for themselves based on their personal judgement and the facts. It doesn't do anybody any good for people to call each other stuipd or whatever because they have different ideas about safety.
I choose to where a helmet most of the time. I do so for my won reasons. I don't think any less of somebody who doesn't. If you are an adult you are able to make that decision for yourself.

On the other hand this study is just plain BAD SCIENCE and that is something worth discussing.

JohnBrooking

I'm not a scientist, but isn't it sloppy science to conduct an experiment with yourself as the only subject? It surprises me that Scientific American would give it credence.

JohnBrooking

Okay, looking at the original, I see that it is filed under the "Weird Science" heading. And hey, isn't that guy in the picture riding against traffic? I can't tell if that's a bike lane or just a shoulder. And he's drawing conclusions about bike safety?

Keith99

Yes and no. It is beyond poor science to dol this and say if gives an answer. It is very acceptable science historically where a lot of science has started to do things like this and say it raises the question.

All the reasearcher did was notice that it seemed that on average cars passed him closer when he had a helmet on. He then decided to measure to be sure it was not just his perception. His feelings were confirmed, the question is now raises. Idiots writing subsequent articles have turned a question into an answer.

musician

what if a cyclist without medical insurance and without a helmet crashes and falls, and has a major brain injury -- don't our taxes pay for their recovery? what if an unhelmeted cyclist falls, is injured -- if we share an insurance company, i'm helping to pay for their choices.

if one chooses to go without helmet, then dies in a crash (and suppose a helmet would have saved them), what of family and friends left behind?

so, yes, it is an individual choice. but it may have profound effects both immediate and unseen upon others.

just saying that if one takes a personal risk (which i am all for, don't mistake me), they should be prepared for that risk and also ready to accept any consequences.

Pat

The methodology of the study is obviously flawed. Finding a difference is not any big deal. You take two means, even drawn from an identical population, you will not get the identical numbers.

The researcher did not apply a statistical test to find out if differences in his results could be explained solely by simple chance. The general statistical requirement is that you need a chance smaller than 5% that the results could be explained by chance to accept that something statistically significant is going on. Actually, if you have major effects, the probabilities go off the test's scales at something like a .1% chance or .01% chance. The fact that this guy did not apply statistics suggests to me that he is a completely novice researcher and he really does not know about methodology.

John E

Why are we subjected to this again?

My rule is simple: always wear a helmet when cycling, and always ride as though you just realized that you accidentally left the helmet at home.

chipcom 

Have been for over 40 years. What if pigs fly?

Satyr

I always love when someone like a psychologist thinks he is a statistician.

Satyr

Scientific American is not a peer-reviewed science journal. It's just a popular science magazine, so it's not really out of line for them to publish something like this.

skanking biker

Thats all you need to know. Any "conclusions" from a "study" conducted by a "scientist" who observes the experiment, analyzes the data AND participates in the experiment as a "guinea pig" are not worth the lab coat the "scientist" purchased from the Dollar Store.

It is a known fact that the results of an experiment are seriously flawed when the participant knows the object of the experiment. Likewise, data collected and analyzed by someone familiar with the purpose of the study is inherently suspect.

I'm not saying it isn't true. I am just saying the "experiment" is not scientifically valid.

chipcom 

I wish people would quit dredging up this 'study'. It was good for a laugh the first time, but I'm running out of ways to make riding with a wig funny.

Brian Ratliff

Yep. SA is a subject reporting magazine covering science related subjects. Kind of like National Geographic. The journals which are peer reviewed and publish scientific work are numerous and, for the most part, completely unknown outside the academic world. Now, if the article was published in the Transportation Research Record where it'd have to be peer reviewed, I'd give him some credence.

That said, perhaps this author is right. Perhaps there is a correlation between helmet useage and passing distance. It passes my smell test, though to form any lasting conclusions I'd need to see the data. Now, the question on my mind is: what is the correlation between passing distance and safety? Everyone assumes that closer passing indicates reduced safety, but most people's experiences are skewed because the main instances of close passing happen in a shared narrow lane situation. But if there is a clear path for two vehicles side by side, say a car in a full width lane and a bicycle in a bike lane, does passing distance correlate with safety?

Brian Ratliff

You know, validity is on a sliding scale. It is definitely less valid than a well funded, double blind experiment, but probably more valid than Forester's sidepath test, i.e. he quantified a variable and took a direct measurement of it. It is probably just a cheap little experiment to shallowly test a hypothesis in order to set the stage for a larger, more controlled (and hopefully better funded) experiment. It is certainly not the last word on the subject, and if it published in SA, the author doesn't even intend it to be the last word. But it passes my smell test, at least. Not to say that I will go about riding around without a helmet because of it though, but it is plausable that the "look" of the cyclist, to some extent at least, influences the passing distance a typical driver will give him or her.

I mean, the money for a study has to come from somewhere, and most research grants, in my experience, require that the researcher already have a good understanding of the phenomena under study. It is kind of an oxymoron, that. I work at an independent R&D engineering company, and the joke is that, to do the proposal, one must already have done the research we are trying to get funding for! And there is a grain of truth to it too.

Brian Ratliff

I take everything back. It was academically published and passed a peer review:

That it was published means his methods were adequately documented and controlled and his findings were statistically significant.