[Ken Kifer's Bike Pages]
ARTICLE: How Many Bicycle Commuters Are There in the USA?
Estimates of the number of commuting cyclists often run rather high. Recent data from the US Census provides much lower figures. However there are good reasons for considering these Census figures to be understated.
Questions What is the source of census data about bicycle commuters? How many bicycle commuters did the census estimate? What are some other sources for estimates of cycling commuters? How does the definition of bicycle commuter affect the numerical results? Are the only bicycle commuters those over 16 who travel to work by bike, or should we count those under 16, those going to school, and those using bicycles to run errands? How does the census fail to allow for the seasonal nature of bicycle commuting? How does it skew the results in favor of motor vehicle use? Why does the census undercount those who don't use motor vehicles? What do the census results prove?

Directories

Bike Pages Home Page

The Cyclist Lifestyle

Bike Commuting and Transportation

Bicycle Camping and Touring

Cycling Health and Fitness

Bicycling Advocacy

Bicycle Traffic Safety

Basic Skills for Cyclists

Cycling Humor and Tales

Bicycling Surveys and Statistics

Links to Other Cycling Sites

Comments on This Page

How Many Bicycle Commuters Are There in the USA?

During the year 2000 census, the US Census sent out a supplimental data sheet to 58,000 participants each month for twelve months, asking various questions including one about methods of getting to work for those over 16. As one of the options was via bicycle, this survey can be used as a source of information about US bicycle commuters. Based on this data, the US Census estimates that there are between 411,000 and 750,000 people over the age of 16 who ride bicycles more miles than any other vehicle to get to work during an average week.

I have seen much higher figures for the number of bicycle commuters than are given by the US Census results. The Bicycle Institute of America has claimed that 4.9 million ride bikes to work, and The Environmental Benefits of Cycling and Walking estimates 2.8 million bike commuters. [Note: after this was published, I have received the following information about these statistics: The BIA estimate was first given as 3.5 million and then increased by 10% a year; it is probably not based on hard data. The "Environmental Benefits..." estimate is for a range of 1.7 to 2.8 million commuters. It does not include those running errands. My source liked my other arguments, especially the undercounting of students.]

I believe that a great deal of the difference between the sets of data lies in the definition of "bicycle commuter." A broad definition would be "someone who uses a bicycle for utilitarian travel (that is, not recreation) on a reasonably regular basis. Using this definition, a cyclist under 16, a cyclist traveling to school, and a cyclist traveling to the supermarket would all be commuters, but according to the census definition they would not. To be listed on the census as a bicycle commuter the cyclist has to be over 16, and the cyclist has to be going to work, not to school or on an errand.

This method of defining bicycle commuters greatly reduces their numbers because, according to the Census, only about half of the people in the USA are over 16 and work. However, it seems to me that the proportion of those traveling by bicycle would be greater among those under 16 (who can't legally drive a car), those who are going to school instead of to work (colleges often place restrictions on the use of motor vehicles on campus, and students often can't afford a car), and those who are retired or unemployed (who would have good reason to avoid wasting money plus more time in which to travel by bicycle). We also might find some additional bicycle commuters among those who work at home and those who walk to work, who might depend on their bikes for running errands but not for getting to work.

Those aren't the only factors reducing the number of bicycle commuters in the Census total. Forms were sent out in each of the twelve months of the year asking people to indicate which type of transportation they had used during the previous week for the greatest number of miles. We all know that cycling is seasonal, as some people are unwilling to ride in the snow or the rain. Bicycle commuters are probably more willing to endure bad weather than recreational cyclists, but the numbers must decline quite a bit during the winter. Now you may feel that the high and low figures given below represent the difference between summer and winter. Not at all; those figures represent the degree of uncertainty, as only 1/4 of one percent of US families were given these forms. If weather was responsible for the difference, one would expect greater fluctuation with these figures in Alaska where winter weather would prevent bicycle commuting than in Alabama or Arkansas where the cyclist can commute every week of the year, but the opposite is true. Instead, the greatest amount of fluctuation occurs in the states where bicycle commuting occurs in only a few cities, thus producing greatly different results from month to month.

In addition, by asking the person answering the census to pick the method of transportation that involved the greatest number of miles, the census is favoring the motor vehicle over the bicycle. Say a person spends three hours a week traveling to work by bicycle and two hours a week traveling to work by car; which probably involves the greater number of miles? Or if someone travels 10 miles round trip to work four days a week and 50 miles round trip one day a week, on which day is the person most likely to not use the bicycle? In the US, figures based on the number of miles are preferred over those based on the number of trips or the number of days because mile-based figures make motor vehicle travel seem more important. If those taking the test had been asked which method of transportation involved the greatest number of trips, the results would have been different.

One final form of bias still exists -- the US Census itself. While some people feel that only an insignificant part of the population is not counted, I know that I was not counted in 1970, 1980, and 1990, even though I was a tax-payer during each of those years. Why? Well, rather than using the post office, the census depends on temporary workers who are more prone to visit houses with nice driveways than to visit small apartments or dwellings without driveways. Because of their choice of dwellings to visit, census workers customarily undercount minorities, the old, the young, and those without automobiles.

However, I feel the following data from the US Census still does have value, even though I consider it to be a drastic undercount. How many times have you heard it said, "There's no reason to consider bicycle riding as a method of reducing pollution, as nobody really uses a bicycle for transportation"? Well, here is census data establishing that hundreds of thousands of people over the age of 16 use bicycles to get to work more than they use any other vehicle or method of travel.

As to what the true figure for bicycle commuters should be, I am still uncertain. I think we could easily double those figures and more by including those who use bicycles for transportation other than work. We would also increase the number by asking about numbers of trips rather than miles. We would also increase the number of bicycle commuters if the census would really count everyone rather than pretend it has counted everyone. Finally, we would increase the number by making some allowance for inclement and winter weather. But there is no way to calculate what those results might be, so when one source says 2.8 million and another source says 4.9 million, they might both be right but using quite different definitions.

The Census Data

State Low Est. High Est.
Alabama 0 3,506
Alaska 1,095 3,847
Arizona 17,806 32,362
Arkansas 0 2,342
California 106,134 134,022
Colorado 8,449 22,207
Connecticut 1,051 4,463
Delaware 759 2,403
DC 2,818 4,656
Florida 39,284 56,932
Georgia 2,081 6,083
Hawaii 1,932 3,642
Idaho 4,048 9,418
Illinois 17,877 31,269
Indiana 5,873 26,775
Iowa 3,501 8,173
Kansas 2,861 8,705
Kentucky 1,353 4,415
Louisiana 4,859 12,577
Maine 860 3,352
Maryland 2,680 7,086
Massachusetts 10,880 21,898
Michigan 16,746 27,122
Minnesota 11,481 19,771
Mississippi 69 1,495
Missouri 2,783 8,357
Montana 3,056 5,184
Nebraska 1,845 4,221
Nevada 1,839 7,525
New Hampshire 10,850 17,896
New Jersey 11,160 25,618
New Mexico 2,490 7,074
New York 25,428 40,242
North Carolina 1,081 3,529
North Dakota 1,266 2,430
Ohio 6,432 13,338
Oklahoma 3,067 11,737
Oregon 14,010 22,982
Pennsylvania 8,852 16,686
Rhode Island 470 2,004
South Carolina 1,280 5,590
South Dakota 904 2,062
Tennessee 455 4,491
Texas 12,011 20,123
Utah 4,487 7,675
Vermont 230 706
Virginia 4,438 9,540
Washington 19,949 34,601
West Virginia 161 1,979
Wisconsin 7,161 13,521
Wyoming 1,433 2,839
Total 411,635 750,471
Elsewhere

The US Census Factfinder, the source of this information.

Carfree Households and Non-Auto Commuting Data from the 2000 Census Supplementary Survey. Jim Gregory's article uses a percentage evaluation on a metropolitan basis and gave me the desire to write this article.

Comments | SECTIONS: | The New World | Writing | Thoreau | Home | Bike Pages |
DIRECTORIES: | Lifestyle | Commuting | Touring | Health | Advocacy | Traffic | Skills | Humor | Survey | Links |
IN SURVEY: | Safety | Politics | Touring | Health | Commuter |
http://www.kenkifer.com/bikepages/survey/commuter.htm | Copyright © 2002 Ken Kifer | May 28, 2002