PSU Prof shares results of bicycling behavior research

Bike to Work Day

Does infrastructure — like this
bike lane — influence people’s
bicycling behavior?
(Photo © J. Maus)

How does the the built environment influence bicycling behavior? What routes do cyclists take? Will cyclists go out of their way to use bike lanes and other bike-specific infrastructure?

Those are the questions Portland State University professor Jennifer Dill has set out to answer. Dill — who teaches at PSU’s Center for Transportation Studiesembarked on a bicycling behavior research study back in January and last week she began to share some of her initial findings.

To learn more about where (and why) people ride, Dill attached at GPS unit on the bikes of 164 Portlanders for seven days. The participants logged over 1500 total bike trips. The GPS device recorded the purpose of each trip and Dill followed-up with an online survey to fill in missing gaps of information.

The research is ongoing, but Dill has already presented early findings of the study at a Congressional briefing in Washington D.C. and at a seminar at PSU last week.

Cover slide of Dill’s presentation.
Download it here (PDF)

Dill was asked by the Robert Wood Johnson Foundation (they funded the research) to speak at a briefing in front of members of Congress sponsored by the Congressional Bike Caucus as part of their recognition of Bike to Work Week. In her statement to that group (which she shared with me), Dill made the case that bike-specific infrastructure is essential if “bicycling for everyday travel can be a viable option in large cities.”

According to Dill, one of the primary reasons for Portland’s success in getting people to go by bike was that our bikeway network grew 250% between 1991 and 2007. Before 1991, she says, the number of Portlanders who regularly rode a bicycle was under 1%. Now that number is somewhere between 4 and 6.5% citywide (depending on what source you use).

Also, during that same period of bikeway construction, the number of people riding bicycles over our four main downtown bridges increased over 400%.

In Washington D.C., Dill told lawmakers that it’s time to go beyond bike lanes;

“Achieving high rates of bicycling for transportation requires a broad range of infrastructure. Bike lanes, the common focus of many cities’ efforts, are just one part of what’s necessary to significantly increase bicycling as a transportation mode. Other facility types, including off-street paths and bicycle boulevards, provide an environment that may be particularly important in attracting new cyclists.”

BTA Bike Boulevard Ride

Bike boulevards provide a
serene and safe experience.

Dill’s research shows that half of the bicycle trips took place on roads with bike lanes, off-street paths, or that were designated bicycle boulevards. However, she also points out that those facilities represent only about 15% of the infrastructure available to cyclists in the Portland

She also found that about one out of every ten miles biked was on a bike boulevard, event though those streets make up one percent or less of the infrastructure. Similarly, while one out of seven miles biked was on an off-street path; these make up only about two percent of the infrastructure available to cyclists.

That means people vastly favor routes with bike-specific infrastructure, or streets that have been specifically engineered to prioritize bike traffic.

Here are some other interesting tidbits from Dill’s findings:

  • The average trip speed was around 10mph. This is the same speed PDOT estimated when calculating times-to-destination on their bicycle boulevard signs.
  • Only 6% of participants reported that they rode purely for exercise. 32% listed “home” and 26% listed “work” as the reason for their trip.
  • The median bike trip length was three miles compared to 4.1 miles for work trips.
  • “Infrequent” riders use minor streets more than “frequent” riders (38% to 31% respectively). On the other hand, infrequent riders use bike boulevards less. This might point to the need for stronger education and outreach about where bike boulevards are.
  • 32% of all trips included another adult (55% of trips made by women).

For more on Jennifer Dill’s bicycling behavior study, an mp3 file and the presentation slides from her recent seminar at PSU are available here.

Jonathan Maus (Publisher/Editor)

Jonathan Maus (Publisher/Editor)

Founder of BikePortland (in 2005). Father of three. North Portlander. Basketball lover. Car owner and driver. If you have questions or feedback about this site or my work, feel free to contact me at @jonathan_maus on Twitter, via email at maus.jonathan@gmail.com, or phone/text at 503-706-8804. Also, if you read and appreciate this site, please become a supporter.

Subscribe
Notify of
guest

29 Comments
oldest
newest most voted
Inline Feedbacks
View all comments
a.O
a.O
15 years ago

This is really important research. Looks like it\’s going to provide some support for continued and increased funding of bike infrastructure. Thanks for sharing.

Todd B
Todd B
15 years ago

Thanks Dr. Dill, RWJ and the volunteers for sharing this work.

brewcaster
15 years ago

I think my heart is going to explode with how much I love this city.

Elly Blue (Columnist)
15 years ago

This is so fascinating.

When looking at these stats about trip length, keep in mind that a trip is just one way, or one leg of the trip. So the average trip is 3 miles each way, or 6 total for a there-and-back trip. And it looks like only 32% of trips (not people!) were reported as trips returning home, which means a considerable number of trips included multiple stops.

So when you consider that survey participants are biking as few as 6 – 8 miles a day at a 10mph average, that gives you *at least* an hour to 90 minutes worth of physical exercise on days when you bike, without (at least for 94% of trips) doing it for the sake of exercise.

Which is a complicated way of saying that pizza and donuts are indeed a viable form of alternative fuel.

Vance
15 years ago

Thanks for the link to the Prof\’s page. I gave this a once over and see that a significant portion of this study is dedicated to vetting instruments for data collection, used in the study. My skepticism regarding surveying the human mind for empirical evidence remains.

Notice I didn\’t say bicycling is bad. Notice I am not challenging the validity of this study and it\’s findings. I just feel that it is important to point out, though, that it is pretty easy to stack-the-deck, and produce results that favor currently fashionable political agendas, when doing this kind of work.

I am all for work like this. I really am. Most important to me is the effort to preserve the integrity of the results, which is clearly being done. By way of, \’for instance\’, how is it determined users prefer bike-boulevards over other modes, when survey participants could easily reside on one. Therefore skewing the results of a survey pertaining to number of miles traveled. Can some one kindly point out the aspect of instrumentation that takes things of this nature into account?

Interested in learning here. Don\’t really feel like laying under the usual dog-pile. I\’d rather retract the question. Please.

mac
mac
15 years ago

Vance please feel free to compare slides #5 and 6 for a general understanding of the relationship between study participants residences and bike boulevard locations. A relatively quick examination suggests to me that there does not seem to be a significant correlation between residences and bike boulevards.

BURR
BURR
15 years ago

164 cyclists seems like a pretty small sample group and how they were selected seems like a fairly important question to me.

How do we know the 164 cyclists who took part in this study represent the full range or cross-section of cyclists in Portland, or if they are more representative of a self-selecting group of cyclists more inclined to use bicycle \’facilities\’?

Adams Carroll (News Intern)
15 years ago

\”How do we know the 164 cyclists who took part in this study represent the full range or cross-section of cyclists in Portland\”

BURR,

no one is pretending this is a representative sample. these are preliminary findings as I\’ve stated and the research is ongoing.

BURR
BURR
15 years ago

This seminar will present results from the BikeGPS study that collected data from Portland area bicyclists using GPS technology. The study collected data from over 150 cyclists for seven days each during 2007, resulting in detailed information for over 1,500 bicycle trips throughout the urban area. The GPS data provides detailed information on the amount, location, and speed of bicycle travel and allows us to answer questions about route choice. For example, how much to people ride on roads with bike lanes, on bike boulevards, or paths? Do these patterns vary by gender, age, or other factors?

It sounds to me like she is attempting to reach some pretty definative conclusions based on a potentially biased group of participants.

Vance
15 years ago

Thanks mac #6. I really appreciate the civility. Now, I viewed slide #5, and had a quick look-see at the instrumentation. The map, in the slide itself, could certainly provide the data pertaining to bike-boulevard proximity, but I do not see in the presentation specifics on how this proximity is being accounted for statistically speaking. Which of course I wouldn\’t expect to see in a presentation, or early-findings report either one. I don\’t see a slide #6, so I couldn\’t make the comparison you suggested. I will take you at your word, of course.

This study is clearly going to be incorporated into policy. One reason I\’m interested, is in forecasting what the impact this sort of thing is going to have on selfish old me. To date, most of the changes occurring on Portland streets are proving to restrict my access to the public right-of-way; and flood the streets with riders who are much more of an inconvenience, and danger, to me than any motorist ever thought of being. That coming from some one whose sole means of transpo for the last 20 years IS a bike.

Another reason is that I\’ve heard of Professor Dill, and have heard some amazing things about her methodology, and instrumentation; and I\’m trying to pick up a little free, secondary education. I\’m leery of conducting surveys, and calling it science. I\’m leery of using tax-payer resources for things that may, or may not, be real. Please do not mistake my lack of enthusiasm for opposition. Trepidatious perhaps, opponent no. Ignorant, definitely.

Thanks again.

bikieboy
bikieboy
15 years ago

Elly said it already, but I\’ll echo: this is fascinating.

Dave Sohigian
15 years ago

Although the sample size may seem small here, I learned something truly staggering from a friend of mine who is with the city transportation office. He attended this talk and mentioned that the national survey for transportation also includes data for bikes. It only has 700 trips in as a representation of cycling activity – and this is the standard information used by city planners all over the country. Even though 1500 trips for Portland may not be representative it is a heck of a lot better than 700 for the entire country!

Matt Picio
15 years ago

There were a lot more than 164 applicants for the study, but only 164 were selected. Some sort of criteria was used in the selection process, but I can\’t recall what it was (I was one of the applicants who was rejected).

Unfortunately, it\’s difficult to make a study of this size truly representative of the biking population at large. In any case, this study still has a lot of value, even if the methodology behind it isn\’t 100% bias-free or objective. (which I\’m not saying that it isn\’t – it could very well be 100% bias-free and objective, I don\’t know)

In my case, I spend a great deal of time on designated bike routes and especially bike boulevards. I live on a bike boulevard currently, but when I lived in Clackamas County, I was nowhere near one, and I still frequently used other bike infrastructure when it was available, especially the Springwater Trail. I\’m not exactly representative, though. Also, while I usually use bike boulevards, I\’m known to occasionally ride down major streets like 39th, Sandy, or MLK (and sometimes 82nd) in order to get somewhere quickly.

This is a valuable study, and I hope a more detailed and broader-based followup study occurs in the near future.

Greg Raisman
Greg Raisman
15 years ago

You\’re right, Dave.

Jennifer noted that the sample in this study is twice as large as the National Household Travel Survey. http://www.bts.gov/programs/national_household_travel_survey/

The NHTS is a major national survey to gage personal travel in the entire country and Jennifer\’s sample is twice as large for just our area.

E
E
15 years ago

well the results seem obvious to me, so I\’m glad she got some numbers to support them. Even with all the stuff Portland has done, it still doesn\’t \”feel\” all that safe, especially to a less-experienced rider. For more people to ride, there has to be safer options – obviously bike lanes are not the preference of cyclists for that feeling of safety. If they\’re not being used, find what will be used and build it. Money spent on things cyclists don\’t like doesn\’t accomplish the goal of increasing ridership – it just is a nice thing to put on your resume when you run for election.

zilfondel
zilfondel
15 years ago

\”This might point to the need for stronger education and outreach about where bike boulevards are.\”

They should replace the tiny bike boulevard street markings with the giant bike sharrows that they have in NW Portland. And they should be so big that nobody can miss them – like six feet wide.

E
E
15 years ago

Burr – I just have to comment on your questioning the validity of a study with 164 participants. Although I\’m not a statistician, if it were a truly randomized sample, the sample size can be surprisingly small. And sure, all statistics have to be looked at carefully. If you are really questioning it, I\’m sure you could find a full description of her study and find out for yourself the validity.

John Russell
15 years ago

Having just completed an AP Statistics course, I don\’t presume to be an expert by any stretch of the imagination. However, I do know that for most the tests of significance performed for comparing various proportions, only a sample large enough to give ten of each result is needed.

For this to be completely valid, however, one would need either a simple random sample or a sample drawn from a population in which the parameter of interest is being measured. The data are not given, so we can\’t tell for sure, but we should be able to assume that any professor is going to be conducting a survey with a properly random sample.

Basically, so long as the professor hasn\’t erroneously collected these data, we can say that these data should produce reliable results.

I hope this might clear up the confusion about the sample size.

Blue commute
Blue commute
15 years ago

I was one of the cyclists chosen for this study as an infrequent cyclist. They had a harder time finding people that were not avid cyclists at first, so they tried to target people later who didn\’t ride their bike all of the time. I want to thank the master students who helped Professor Dill, and worked as the outreach to all of the participants, they were great. I knew when I had the GPS unit that they were getting an incredible amount of data, and I very much looked forward to the analyses. There is still so much they could extrapolate from this study. Once again, this is something that makes me happy to live in Portland!

peejay
peejay
15 years ago

Is it just me, or is Vance behaving with much more civility lately? Oh, I still disagree with most of what he says, but he\’s kind of – dare I say? – contributing to the discussion.

That said, Vance, are you actually advocating discouraging bike use for novices so that you can enjoy your ride more? That\’s a little selfish, if you ask me.

Scott Mizée
15 years ago

@BURR #7:

I was also at the presentation and was one of the cyclist participants last spring. Professor Dill spent a significant amount of time at the beginning of her presentation explaining that this was not a representative sample. However she also referenced the National Household Travel Survey (as other readers have mentioned) and pointed out that this data set, however limited is still incredibly \”rich\” compared to anything else available in the United States.

rlk
15 years ago

Alternative fuels indeed! Love the way you think, Elly… I\’d add blue cheese dressing (OMFG!) to the list as well. 😉

Rode home from work early today, and was supremely amused at how many bike commuters are out there…oouf!

Michael M.
15 years ago

PDX seems to be turning into Survey City lately. Saturday morning at the Hollywood Library someone from PSU was collecting info on how patrons got there and where they came from. Today downtown a transportation consultant was taking a survey on people\’s mode of transportation for whatever brought them downtown and the reasons for those choices. I was locking my bike in front of the Columbia Sportswear store when he approached me, so it was pretty obvious how I\’d traveled.

Donald
Donald
15 years ago

OK, who gave gave Vance a scholarship to attend stat class? And who\’s in charge of pop quizes?

But seriously: Vance, you are so totally my favorite high school drop out of the day.

I do take exception though as I have a biological affinity to the safety in numbers theory and I feel that more riders of all skill levels raise the safety factor for all of us two tired wheelfolk.

Ain\’t no bike rider ever scared me as much a teen driver in a lowered Civic showing his latest text message to his back seat passengers.

(Although I did have a close call with a bakfiet full of toddlers captained by a cell phone chatting hotmom this morning on the steel bridge.)

Ginseng
Ginseng
15 years ago

This is a good study by Professor Dill. Now what? Her closing slide states\” Use the data to improve Metro\’s ability to model bicycling\”…

Does Metro model bicycle data…I think not.

My opinion of the research is that it is beginning to build upon a lot of the research that examining public health, urban infrastructure, and the goals of compact urban development. Her approach of actually \’mapping\’ the routes of the individuals and be to compare them across the region on the maps certainly displays that portland cyclists (at least the ones sampled) are busy folks.

By mapping the actual routes used by cyclists, planners at the City and Metro can now visualize how extensive our roads and paths are used, both on dedicated and non-dedicated routes. They can \”begin\” to understand why and where people travel. And rationalize how to fund projects giving scarce resources.
The map on slide 7 shows a fundamental difference in travel behavior. \”Where\” do people bike…looks like they like gridded street pattern, with shorter blocks and lower volumes of traffic…Just look at downtown and SE and compare that to the suburban patterns of Washington county. The bikers in the suburbs seem to get funneled onto arterials…just what DIll identified as a type of barrier.

The unfortunate reality of this study is that it is currently trapped in academics. And the place to attack is not with Dill and her sample size or methodology…RWJ Foundation would not fund shoddy research. The sample and methods are fine. Take a shot at Metro.

Planners at Metro should find it useful, but the data is not part of their tranportation model. So, the data does not impact the projected usage statistics on modal splits in the future…and therefore the modal and the Managers have the planners focused on projects like the Columbia River Crossing or the lame widening Highway 217 project. Projects that are focused on autocentric development. The map alone demonstrates the need for increase cycling infrastrucute across the region. In the city for choice, efficiency and safety…in the burbs for connectivity and choice.

In fact, the Metro transportation model (this model drives the priorities and funding for transportation projects) falls apart for 2 specific reasons. The first is that it is old….the data is based on 1995 (or so) sample…Dills research shows the region is vastly different that in 1995. And the most important thing about mode choice…the Metro model does not account for the price of gas.

This type of research should make the Transportation Departments at Metro and the City take a serious look at their own data (and how old and useless it is) and how contemporary travel behavior data needs to be integrated into a more useful model. If the academics are doing this…why isn\’t Metro. They are handing out the project money!

Vance
15 years ago

peejay #20! Hey thanks, I really appreciate that. Jonathan, and tonyt are responsible, not I. They reached out, I accepted, and here we sit, typing away! Without their intervention, I\’d still be counter-blogging.

That said, Vance, are you actually advocating discouraging bike use for novices so that you can enjoy your ride more? That\’s a little selfish, if you ask me.

I can see where you could get that from both past comments I\’ve made, and today\’s too. Chalk that up to poor communication skills, and high-personal-stakes for me. Let me pose this hypothetical: Are you saying that you support attracting new riders to this thing we do, and all other riders and their needs should be ignored?

Of course not, I hope. Same is true for me. I may criticize, but that should not be construed as opposition. If my criticism lapses into opposition, or if I fail at communicating my ideas correctly then I would hope that somebody says something, instead of leaping to conclusions. I\’m only human, and whether you\’all will allow me to feel justified in my resentment or not, I\’m still quite bitter and angry.

Please exercise empathy before judging me. I\’ve been here, on my bike, in the rain, for a very long time. I\’ve bled all over these streets, quite literally. I was not in a position, through no fault of any one\’s but my own, to cope with the Portland Culture Rush. I found myself surrounded by strangers in the blink of an eye.

I was a messenger here from 1985-1994, and then off and on through-out the rest of the 90\’s. I was deeply involved in an effort to unionize that job. The failed attempt resulted in my being black-listed from every industry job, and contact, in the entire area, for me and a handful of other dedicated. Just when I was getting the pieces of that fiasco put together, 2.5 million people moved here in the space of 4 years.

I\’ve been an outsider, in my own hometown, ever since. Sour grapes is all. If that makes me a bad person, so-be-it. I\’m trying my level best to integrate, to re-learn the social, and political landscape that this community has become. Any way you cut it though, I was just sort of riding along, minding my business, when all of a sudden…

No excuses. I\’ve been screaming my resentment at total strangers, my bad. I\’m trying to demonstrate that I\’m not useless in this thing, just really mad, and feeling quite unappreciated. At least a small part of why this whole thing is happening where it is, is due to some fanatics getting out in the weather, and the traffic; and kicking, biting, clawing, and screaming their way onto the public right-of-way. If you had any idea how much money I\’ve spent on traffic citations…

Ironically, the entire movement has done their level-best to ostracize, and alienate these folks who paved the way. I\’m not the only bitter local crying in their milk about getting shut-down by the tourists, you know. To date, I\’m one of the few trying to let bygones…, and get down to the business of making the Portland cycling community a model the entire country uses as an example of how to do it right.

With that said, consider. Bringing novice riders into this thing, too soon, or without a vetted safety infrastructure, could result in unnecessary loss of life. Bringing novice riders into this thing increases the possibility of inflaming already hostile public sentiment. These ain\’t cars. These ain\’t gonna replace cars. People can get killed on these things. What good is that going to do anybody? You\’re reading opposition into statements that are meant to be a warning, and nothing more. Bottom line, it should be remembered that a novice rider is, above all, a novice rider; and we wouldn\’t want to pressure folks into a hospital-gurney, would we?

When I say, \”But what about me?\”, I really mean that it is great to grow your industry and all; but what of those who\’ve sacrificed, and bled? I\’ve earned a voice in this, and it rankles to be marginalized, and ignored just because I advocate careful planning. The tacit demand has been placed upon me to accommodate. Well, dogawnit, I already paid for my ticket now let\’s get the show started!

I can string a sentence together in the language I was born speaking. I\’m not stupid. Additionally, I\’ve forgotten more about urban-cycling than most people can learn in a lifetime. Strap a digital-cam to my body, and let me do my thing. I\’ll show you some things that will prove to be invaluable as we progress this thing through its paces. Or, write me off the first time I take a toe off the party-line. It\’s really up to you\’all now.

Sorry about the book. Again. And thanks peejay, that\’s pretty awesome of ya!

Vance
15 years ago

Donald #24. Thanks to you too! I\’m quite proud of my drop-out status, thank you very much! It\’s important to remember that I\’m an uneducated dolt, way more than I\’m a cantankerous old-coot. Quite a lot of both truthfully, but questioning stuff is my mots-operandi, and sometimes I like to throw elbows. Whateryagonnado, except apologize, and move on?

For the record, preaching bikes to me while swathed in a viscous, thick, layer of body-fat has been known to set me off. Preaching bikes to me, while your car sits leaking crankcase lubricant into the ground-water, has been known to set me off. Preaching bikes to me, when I\’ve been on one for more than 20 years, and treating me like I don\’t exist, has been known to set me off. I\’m really discovering I\’m sort of anti-preach-at-Vance about a lot of things concerning bikes. Something I\’ve realized is my responsibility, and not you\’alls. But if you ain\’t walking the talk, allow me to warn you in advance, my sense of entitlement will, at least, rival your own.

Donald
Donald
15 years ago

You keep on rockin in a free world, Vance.

I got graduated from Jeff in 84. Spent two years driving delivery downtown, biking in from Mt Tabor everyday. (Rembember the old guys on the red cruisers delivering blueprints?)

You\’re right, it\’s a brave new world out there. I can\’t decide some days if I miss the past or if I\’ve just romantasized it. Either way, it\’s still a freaking bike jam on Vancouver/Williams on sunny days and so I hear from where you\’re comin\’.

_DA

BURR
BURR
15 years ago

Wider bike lanes are in order for Vancouver/Williams, lower Hawthorne and a lot of other areas where the number of cyclists have now exceeded the capacities of these bike lanes.

One thing\’s for certain, study or no study, the number of cyclists in PDX is increasing faster than the local and regional governments can keep up.