New Survey Numbers Show Surprising, But Slight, Dip in Bike Commuting

Bicycling advocates say that the American Community Survey has never accurately measured bike commuting, because they don’t ask the right questions. That may be true, but the upshot is that a year that appeared to be a banner year for cycling ended up being kind of a dud, according to the ACS. The ACS recorded a slight dip in bike commute mode-share, from 0.55 percent in 2009 to 0.53 percent in 2010.

Though it may seem like bicycling is booming, the data, puzzlingly, says otherwise. Photo: ##http://greatergreaterwashington.org/mvjantzen/page/1##M.V. Jantzen / GGW##

“The methodology will always work against us,” said Darren Flusche of the Bike League. The ACS question about commuting goes like this:

How did this person usually get to work LAST WEEK? If this person usually used more than one method of transportation during the trip, mark (X) the box of the one for most of the distance.

“As long as that’s the question, it’s going to undercount cycling,” Flusche said.

After all, someone who rides twice a week doesn’t get counted for any cycling. Someone who bikes to the train station doesn’t get counted. Someone who works too far away to bike but does everything else on two wheels doesn’t get counted.

A somewhat more useful survey is the National Household Travel Survey, but that has only been done twice, with an eight-year gap in between. It counts all trips, not just commuting, but it can’t be broken down geographically.

The .02 percent “drop” in the ACS isn’t statistically significant, but anyone following the huge gains in bicycling over the past couple years would have expected to see a jump, if anything. Major cities like New York, San Francisco, Chicago, and Washington are competing for bike-friendly gold, installing bike lanes, bike corrals, and cycletracks at a dizzying rate. Bike-share systems are proliferating. Transportation Secretary Ray LaHood declared bikes on an “equal footing” with other modes. Federal funding for bicycling has risen dramatically above what it was just five years ago.

So what gives with the low number?

Flusche emphasizes that we don’t really have to go looking for excuses to justify the dip. After all, in the 70 biggest cities in the U.S., cycling rates held steady. In cities dedicated to improving cycling, rates rose.

And sometimes, you just have to take the long view: Bike commuting is up 39 percent in the U.S. since 2000 – and up 63 percent in the 70 biggest cities. That’s news to celebrate.

John Romeo Alpha at network blog One Speed: GO! did an interesting analysis of the data as well, finding that population density tended to have a proportional effect on bike commuting. He found that Seattle managed to have a high bike mode share without density and New York had high density without correspondingly high mode share, but in general, the 25 most populous cities had cycling rates that corresponded with their mode share.

Source: One Speed GO!

That doesn’t necessarily offer an explanation of the numbers, but it does help illustrate how a city can increase its bicycle mode share. The Bike League also noted that its awarded “Bicycle Friendly Communities” just keep winning over new cyclists, but it’s harder to see year-over-year. Since 2005, “the 38 Bicycle Friendly Communities among the 70 largest cities saw a 95 percent average increase in bicycle commuting” compared to 46 percent growth for the rest of them. Numbers like that put a .02 percent drop, in a survey with questions stacked against cyclists, in perspective.

  • Anonymous

    Great point about how commute surveys are designed. I always face this problem, since I spend about 1/3 of my daily commute on my bike and about 2/3 on the train. And since the train goes much faster than I do on my bike, even on days when I bike longer and take the train less so that the ratio of my *time* spent on the train and on my bike is the same, I still have to say “train” which I choose my mode of transit on a survey since that by covers the most distance. There is no doubt that this is a significant reason why the number of cyclists are bigger than surveys indicate.

  • Anon

    Is there a typo here? The spreadsheet you reference says that 3.5% of SF commuters go by bike.

  • The Greasybear

    There are two aspects to ACS bicycle commuting–the bike part and the employment part. I am part of the small drop in bike commuters right now, not because I switched modes but because I got laid off and don’t currently commute. Given the recent rise in cycling in major US cities among those who still work, it could very well be the small drop reported by the ACS is due to slightly fewer cyclists holding jobs.

  • right – isn’t ACS data only work trips?  Per National Household travel survey – for all trips – Los Angeles County has 1.86% done via bicycle and and over 20% of trips done via walking. !

  • Mark Elliot

    “Why the low number?” I think you’ve answered your question right here. The survey is not measuring what we need to measure, as you point out, so that’s a problem right there. Then there is the margin of error, which we don’t know anything about. It could be 5%. Given the relatively small share of cyclists (still) in NYC, and especially as counted in the ACS, one would expect a high margin of error for a small subpopulation like that. So not only is the .2% drop not significant, but even a 5% change would have to be considered with some skepticism were that the margin of error. And it’s not even measuring what we want to measure! We’ve got a savvy, data-hungry bunch of advocates here that need to be fed a better data diet.  

  • Anonymous

    Why would that be a typo? 3.5% seems about right for SF, or perhaps low. I used to live in Berkeley and cycle commuting was huge. Also, TONS of people took bikes on BART (including me), which would mean I wasn’t counted as a “cycle commuter”

    Note http://blog.sfgate.com/transportation/2010/10/07/san-francisco-oakland-move-up-national-list-of-bicycle-commuting-cities/

  • Yuri

    The survey needs to be changed to reflect a multi-modal commute. I bike 2 days a week to work and take the bus on 3 days.  The survey needs to reflect that.  Even if I rode my bike to a train station, the survey should reflect all modes for a commute, perhaps weighted by distance.

  • RickLA

    It would be great if the National Household Travel Study could be geolocated to Census tracts.

ALSO ON STREETSBLOG

Census: American Bike Commuting Up Nine Percent in 2012

|
Congratulations, America. We’re biking to work more than ever before. We’ve known for a while that Americans are driving less than they used to, even as the economy grows. And just about every quarter, the American Public Transportation Association delivers more stats about increasing transit ridership. Now the Census brings another measure of Americans’ shifting transportation […]

Low-Income Americans Walk and Bike to Work the Most

|
The U.S. Census Bureau just released its first-ever report exclusively on walking and biking [PDF]. Using data from the American Community Survey, the report shows how rates of active transportation vary by age, income, education, race, and the availability of a vehicle. It’s a lot more detail than the usual Census data release on how people get […]

Study: Federal Funding Means More Bike Commuting

|
Bicycling is at a tipping point in many American cities. Bike-share systems are multiplying rapidly, infrastructure that used to be seen as novel is now commonplace, and commuting rates are growing. There are many explanations for this cultural shift, but here’s one not to be ignored: federal funding. Georgetown Public Policy Institute student Marissa Newhall […]