Lies, damned lies and statistics

As Twain – or was it Disraeli? — suggested, statistics can be used to support virtually any argument, valid or not.

And often, both sides of the same argument.

For instance, numerous New York publications and organizations have used a recent Hunter College study to support their contention that cyclists in the city are a bunch of dangerous, out-of-control scofflaws.

They cite statistics showing that 37% of New York cyclists didn’t stop at red lights, while another 28.7% stopped briefly before going through the light. And that 13% rode the wrong way, against the flow of traffic.

Sounds pretty damning.

But look at it another way, and it shows that 63% of cyclists did stop at red lights, and less than half of those continued through after determining it was safe to proceed. And an overwhelming 87% of cyclists did not ride the wrong way, despite the city’s numerous one-way streets.

Statistics.

I relied on them on the AirTalk program the other day to rebut the popular contention that all cyclists run red lights — citing the Hunter College study, as well as studies from London and Melbourne that each showed a compliance rate of well over 80%.

Other studies allowed me to show that cyclists aren’t the only ones who break the law, such as over a third of drivers routinely run stop signs in residential areas and seven out of 10 drivers exceed the speed limit.

Of course, the validity of the statistics depends on the quality of the research. And like Schroedinger’s Cat, the results can depend on the observer.

For instance, a study by the California Highway Patrol showed that cyclists were at fault in 60% of collisions, while research by a Toronto professor said cyclists were at fault less than 10% of the time. Cyclists would argue that the CHP study offers proof of bias in investigating bicycling accidents; drivers would contend that the Toronto professor’s interpretation of the data was biased because he is a cyclist himself.

I’ve often struggled to find valid statistics for cycling in Los Angeles — like how many people in L.A. ride bikes, what routes are most popular, where they would ride if more people felt safer on the streets, what percentage observe stop signals, and how much cyclists contribute to the local economy.

With over 40 universities in Los Angeles County — including a number of world-class research institutes — it should be easy for the LACBC or the LADOT to find one willing to help design and conduct a study that would answer those questions, and countless more. And give us a detailed picture of where we are now and where we need to go.

Of course, some baby steps have been taken. Earlier this year, the LACBC called on volunteers to conduct the city’s first bike count, establishing a baseline we can use to measure further growth in ridership.

Local cyclists are anxiously awaiting the results — some with rapidly diminishing patience, if my emails are any indication.

Now Alexis Lantz, a Master’s student in UCLA’s Department of Urban Planning — and an intern with the LACBC — is conducting a cycling survey on behalf of the Los Angeles Sustainability Collaborative.

It may not give us all the answers we want. But it could provide a good snapshot of how and why we ride.

You can find it by clicking here. Participation is limited to Los Angeles residents over the age of 18, and the deadline is December 15th.

So why not take few moments to complete the survey?

I just did.

………

Westwood is about to get its own Neighborhood Council; could a bikeway through the LA Country Club be far behind? Westwood is also home to a new student bike group, the UCLA Bicycle Coalition. SoCal’s wayfaring bike couple pedal through Yosemite — and they have the breathtaking photos to prove it. Speaking of photos, who knew a Missouri sky could look so good? Now that the Gold Line extension is open, you’ll need a place to park your bike. Maybe you missed Mr. and Mrs. Cindy Crawford riding along the beach last weekend. A Cupertino cyclist is killed in an apparent right hook or left cross. San Francisco wants to know where you ride, and they’ve got just the app to do it. Would stricter penalties help protect vulnerable road users? If you’re riding with an Easton EA30 stem, get off the bike and stop riding now; then again, if you’re riding now, you shouldn’t be reading this. For a change, the Feds want to increase bike and pedestrian funding. Coming soon: your very own I Pay Road Tax jersey. Finally, the most stirring eight minutes of bicycling video you’re likely to see today. And the music’s not bad, either.

3 comments

  1. TheTricksterNZ says:

    I admit I had to check my lovely steed last night. I’ve got the EA50′s rather than the 30′s. Want to get 90′s though just so I can pretend to be cool.

  2. stezzinator says:

    Leading institutions don’t survey cyclists for a reason: it’s exceedingly difficult to get publishable results. To trust any results, you need to get a random, representative sample of the population. That’s a problem with most transportation surveys, but a much bigger problem with cyclists.

    You want to survey motorists? Easy. Call a number at random, and you’ve probably found one. You can’t do that with a bus rider (too expensive to call everybody looking for a few), but there’s a place where they gather: buses. Cyclists are rare AND diffuse, so you don’t even have the gathering spot option.

    As a result, you can never look at a sample and have any confidence that it’s representative. Roadies, party riders, transpo riders by choice and transpo riders by necessity are different. Get any bias in your sample, and your results are trash.

    No one’s cracked that nut. Even the cyclist (corner) count recently performed only tells you about the corners counted. So until there’s a defensible, financially feasible sampling method,I don’t expect any institutions to be lining up to cyclist surveys.

  3. bikinginla says:

    Actually, cyclists aren’t all that rare. Previous studies have shown that between 38% and 44% of Americans ride bikes, so at least every third call should connect with someone who rides.

    If you want to gather input from a variety of riders, all you have to do is conduct your studies at a variety of bike shops, from co-ops like Bike Kitchen and Bikerowave, to mid-range shops like REI and Performance, and high-end shops like Helen’s and Cynergy.

    Behavior is even easier to measure. For instance, to get a number of how many cyclists run red lights, all you have to do is select a random sampling of intersections where running the light is a viable option, then count how many stop and how many run it.

    As this post indicated, similar studies have already been conducted in London, Melbourne and New York. It’s time we had some solid numbers for L.A.

Leave a Reply

Your email address will not be published. Required fields are marked *

17 + 8 =

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

%d bloggers like this: