« Maybe they're from out of town? | Main | Bike frames in Rock Creek Park »


Feed You can follow this conversation by subscribing to the comment feed for this post.

As to the opinions of medical doctors, there is a long tradition in cycling of hopefully quoting them so that their opinion(s) will be relied on given their special knowledge (assuming one thinks it is special). See for example the 1892 "The Bicycle: Its Selection, Riding and Care" that spends many pages quoting medical doctors to convince the public that women riding was good for them. (And quoting ministers that it was OK from a religious point of view as well.)
See http://issuu.com/michael.neubert/docs/thebicycle

As for DC and driving training, WMATA bus drivers and DC Govt. heavy equipment operators do get supplemental training vis-a-vis pedestrians and bicyclists.

You probably should correct "87% of crashes". According to the press release there were only 2 crashes. The press release also talks about 6 near collissions, with the rest of the 54 "incidents" being incidents where something unsafe happens.

You might also want to put quotes around "study". The press release does not name an author or post a report, and the cooperating university's accident research center makes no mention of the study. So it is at least possible that the press release is an interpretation of an unpublished study written by the executive quoted, rather than a faithful abstract of what the study--when published--actually shows.

Note also that the Foundation executive also states that bicycle sales are higher than car sales in Australia--which is difficult to reconcile with the drop in cycling attributed to requiring helmets.

Oops. I meant that you might want to change "study out of Australia" to press release out of Australia

I'm going with the wording from the news story.

Well you haven't yet.

The data out of Australia is going to be a little skewed in favor of cyclists. Since cameras were mounted on cyclist helmets, it's safe to guess that the cyclists knew they were part of a safety study. The participating cyclists were probably on their best behavior, so of course they cycled by the book. I bet if you eliminated that prejudicial flaw from the study, bikes still come out ahead, just not by such a large margin.

Jim, you wrote "You probably should correct '87% of crashes'" when I wrote "87% of car-bike crashes". From the article "Drivers were at fault in 87 per cent of incidents with cyclists" I stand by my choice of language.

You also wrote that I should put quotes around the word study. From the article entitled "Study blames drivers for bike crashes," "The three-year study into cyclist safety..." and "The helmet camera study found..." So there again the article uses the word study, without quotes.

As for bike sales exceeding car sales, that's not a surprise. For one thing there are more potential cyclists since you can start biking at 4. There is also less of a market for used bikes, driving more new bike sales.

Very glad but surprised to see Dr. Pucher mentioned in this bolg- his ideas seem to meet with scorn from most of the commentors here.

I'd like to know more about that study out of Australia. Even with a small number of participants, 2 crashes and 6 near collisions over a 3 year study period, makes cycling look relatively safe.

That reminds me of a stat we don't talk about often enough: 5.1 years between collisions for the average DC driver, who has All State insurance.


All you need to know about the "study" is that its design is so flawed as to make it completely pointless. Anyway, one should immediately distrust any report of "study findings" that includes no details of methodology, nor a link to such, nor is it possible to find that onesself using google.

This article at least gives the basics of the methodology, and they are bad.

"13 adult cyclists in Melbourne were given helmet-mounted video cameras and asked to film 12 hours of commuting each over a four-week period."

Though it hardly seems necessary to explain what's wrong here, basically, everything is.

1) The study participants all know they are participating. (Not double blind... not even single blind).
2) The sample size is extremely small and certainly not at all representative of the population of cyclists at large in terms of age, travel patterns, skill, and so on.
3) The "conclusion" is based a very small number of data points. The statistical significance of this finding is none at all.
4) The fact a camera was mounted on each participant's head during the study certainly affects their behavior (since they know their actions are all being recorded). Nobody should be surprised to find that a cyclist with a camera on their head, knowing that indeed the purpose is to study accident fault, would be on their absolute best behavior. In fact I'm surprised that there were any incidents deemed to be the cyclists's fault at all.

I'd love to see the same "study" done except with 12 drivers. Would anyone be surprised to find that those drivers they selected just happened to be remarkably courteous and law-abiding when there was a camera on their dashboard?

I swear this is true - I was just thinking this morning about how somebody should invent an airless tire with some kind of foam inside.

The comments to this entry are closed.

Banner design by creativecouchdesigns.com

City Paper's Best Local Bike Blog 2009


 Subscribe in a reader