Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration
Office of Planning, Environment, & Realty (HEP)

Interim Report to the U.S. Congress on the Nonmotorized Transportation Pilot Program SAFETEA-LU Section 1807

Chapter 5: Results of Phase 1 Data Collection

Previous Return to TOC Next


This chapter is a summary of research conducted by the University of Minnesota's Humphrey Institute of Public Affairs and the Center for Transportation Studies, in collaboration with NuStats, in support of the NTPP. A full description of the research is provided at The research team designed and implemented surveys to collect travel behavior data to establish a baseline or "before" information on travel by bicycling and walking in the four pilot communities (and in the control site of Spokane, Washington10). The research team will use this baseline data in comparisons to "after" data that it will collect with the same surveys in fall 2010 to identify changes in travel behavior in the pilot communities. This chapter summarizes the survey and analysis methods used in this phase of work and presents key characteristics of walking and bicycling behavior as they relate to NTPP objectives. The following results consider only the transportation patterns of adults primarily because of barriers to collecting survey information directly from minors.

Key Results

Across the four pilot communities, non-auto mode share (which includes travel by bicycle, walking, and public transit) for adults ranges from 8.5 percent in Sheboygan County to 29.3 percent in Minneapolis (Table 5.2). Survey data indicate that on a given day, about 2 to 4 percent of adults in the NTPP communities bicycle for utilitarian purposes (which includes bicycling to work or to other destinations, but not for recreation), while about 15 to 35 percent of adults walk (these are ranges across the communities). The average daily distance for bicyclists is about five to eight miles; for walkers it is about 1.5 to 2.0 miles. For both modes, about 30 to 40 percent of bicycle or walking work commute trips would otherwise have been made by driving. About 95 percent of bicycling and walking trips to other destinations would otherwise have been made by driving.11

Given these basic results, the research team arrived at a measure of "avoided driving" that can be attributed to walking and bicycling. The research team will use changes in this measure to identify broader changes related to program goals including environmental quality and energy use. While there was considerable variation across the communities, the two modes combined to replace approximately 0.25 to 0.75 mile of driving per day, per adult resident. About 70 percent of this avoided driving was due to walking. While walking trips are shorter, far more people walk than bicycle on any given day.

This use of walking and bicycling for utilitarian purposes reduces driving by about 1 to 4 percent, depending on the community. Because of the large populations involved and the constant nature of this rate of reduction over time, this seemingly modest contribution leads to significant long-term results: the research team estimates that the current total reduction in driving in all four program communities, over the course of an entire year, is in the range of 156.1 million miles.12 This total excludes recreational trips and trips by children because the study primarily focuses on utilitarian trips.13

Research Design

The research team designed two surveys - a short mail survey and a longer Internet or telephone followup survey - and conducted them between September 2006 and January 2007. From the original 31,120 mailed surveys a total of 4,457 were completed, yielding a 15 percent response rate. Of respondents who returned the mailed surveys, 1,514 completed followup surveys, for a 34 percent response rate. While most of the followup survey was the same for all respondents, one section was devoted to a more detailed exploration of a single "reference trip" as well as some mode-specific attitudinal questions. The "reference trip" is a term that refers to one mode-specific utilitarian trip per followup survey about which the research team asked several additional questions (e.g., trip distance and destination and perceptions along the route).

Table 5.1 shows the response rates for each community and the reference trips by mode. While the research design called for 100 responses/reference trips for each mode in each community to meet statistical objectives, this proved difficult and was ultimately not possible within time and budget constraints.14 To control for possible external effects that could affect all of the communities (such as a dramatic change in gas prices or economic conditions), the research team identified and surveyed a control city, Spokane, Washington.

Table 5.1: Counts of Survey Responses by Community

Community Self-mailer Full Survey Reference Trip Mode
Transit Bicycle Walk Auto
Columbia 797 313 50 73 104 86
Marin 891 272 70 52 100 50
Minneapolis 837 343 123 62 104 54
Sheboygan 972 297 26 70 101 100
Spokane 960 289 66 50 100 73
Total 4,457 1,514 335 307 509 363

The research team surveyed specific geographic areas for each community. The survey was conducted within the city boundaries of Minneapolis and Columbia and within the county boundaries of Sheboygan and Spokane. For Marin, the survey focused on a specific list of census tracts representing the eastern urbanized portion of the county where projects will be concentrated.


Mode Shares

Table 5.2 provides mode share and total daily mileage per person by mode for utilitarian trips (trips made to a destination). Non-utilitarian trips (trips made solely for recreation or exercise) are discussed in a later section.

Table 5.2: Share of Total Person Trips by Mode and
Average Daily Mileage per Person by Mode

  Auto Walk Bicycle Transit
Community Vehicle % Rideshare % Average Miles % Average Miles % Average Miles % Average Miles
Columbia 86% 2.2% 15.1 8.6% 0.30 1.5% 0.10 2.2% 0.21
Marin 82% 1.4% 23.6 11.8% 0.40 1.8% 0.22 3.2% 1.37
Minneapolis 69% 2.2% 20.7 17.6% 0.55 2.0% 0.23 9.7% 2.23
Sheboygan 89% 2.4% 22.3 6.6% 0.16 0.7% 0.06 1.2% 0.11
Avg. for Pilots15 82% 2.1% 20.4 11.2% 0.35 1.5% 0.15 4.1% 0.98
Spokane 85% 2.0% 25.9 8.5% 0.25 0.8% 0.08 4.1% 0.88

Walking and Bicycling Leading to Reduced Auto Use

Given the objective of determining the amount of driving that is being avoided, the analysis focuses on trips that might have been made by driving if not by walking or bicycling. To determine the number of adults walking and bicycling for utilitarian purposes, the research team used the percent of respondents on the self-mailer who reported walking and bicycling to a place "yesterday." The research team assumed that all of these walking and bicycling trips were for utilitarian purposes and not solely for recreation or exercise. Children were not surveyed because of privacy concerns. The research team assumed that adult responses provided useful information about travel by children.

The research team used two sources of information to estimate total daily walking and bicycle mileage. One method uses average trip lengths from followup survey respondents with mapped walking and bicycling reference trips multiplied by the average number of daily trips by walking and bicycling. The research team estimated this number using the percent of respondents to the mailed survey who reported riding a bicycle or walking to a place the day before they completed the survey. The other method calculates average distance in miles based on total daily walking and bicycling travel times and uses an assumed average speed of 3 mph for walking and 10 mph for biking to estimate daily distance. The research team used the midpoint between these two estimates in the following analyses.

Between the two modes, the total estimated reduction in auto travel is in the range of 0.5 miles per adult resident per day. This is in the context of average levels of auto travel in the range of 15 to 25 miles per day per person, across the communities. Based on the research, the use of nonmotorized modes appears to reduce the amount of total auto travel in these communities by 1 percent to 4 percent, establishing a baseline from which to derive post-program comparisons (Table 5.3).

Table 5.3: Percentage Reduction in Auto Travel

Community Estimated daily driving per adult (miles) Daily walking and bicycling per adult (average in miles) Percentage reduced
Columbia 15.10.453.0%
Marin 23.60.672.8%
Minneapolis 20.70.823.9%
Spokane 25.90.311.2%

Table 5.4 shows the total amount of auto driving for utilitarian purposes in miles that are annually substituted by walking and bicycling in the NTPP communities.16 Because the survey was conducted at the end of the good weather season in most of the communities, the research team assumed that the daily bicycling and walking rates represent an average for the entire year (365 days). While winter rates will be lower than the average, summer rates will be higher, which led the research team to conclude that it is reasonable to assume that the two will offset each other.

Table 5.4: Total Annual Estimated Reduction in Miles of Auto Travel Due to Walking and Bicycling

Community Average
Columbia 11,033,324
Marin 48,286,503
Minneapolis 88,887,977
Sheboygan 7,894,232
Total 156,102,036
Spokane 16,380,21217

Access to transit

A related question about mode substitution involves trips by transit. One possible impact of improved walking and bicycling conditions might be in providing better access to transit. If, for example, a person chooses to drive to work because it is too hard to find parking at a transit stop, improved nonmotorized access might help to eliminate a car trip by substituting it with a combined walk-transit or bicycle-transit trip. Table 5.5 shows what mode people used to get to a transit stop.

Table 5.5: How did you get to the transit stop?

Community Bicycle/Walk Drove/Rode Sample Size
Columbia 89%11%47
Marin 45%55%64
Minneapolis 88%12%116
Sheboygan 84%16%25
Spokane 78%22%65

Recreational bicycling and walking

While the survey primarily focused on utilitarian travel, a number of questions addressed non-utilitarian, or recreational, travel. While the information in Table 5.6 is useful in identifying increased activity among those who walk or bicycle relatively infrequently, other questions provide additional information about those who already participate in these activities on a regular basis. Several questions closely mirror the Behavioral Risk Factor Surveillance System (BRFSS) administered by the CDC. Respondents were asked whether they walk or bicycle for at least 10 minutes at a time during a "typical" week. Those who do were then asked how many days per week they typically engage in each of these activities and for how many minutes on a typical day (Tables 5.7 and 5.8). These tables will provide a baseline against which to compare related changes in travel behavior in 2010.

Table 5.6: The most recent time respondent used a bicycle for any purpose or walked for recreation or exercise

  Within Past Week* Within Past Month Within Past 3 Months Within Past Year Not in the Past Year
Community Bicycle Walk Bicycle Walk Bicycle Walk Bicycle Walk Bicycle Walk
Columbia 10% 50% 19% 68% 27% 75% 38% 80% 62% 20%
Marin 14% 67% 22% 77% 30% 82% 38% 86% 62% 14%
Minneapolis 11% 60% 27% 79% 38% 86% 48% 90% 52% 10%
Sheboygan 6% 50% 17% 67% 28% 75% 37% 82% 63% 18%
Avg. for Pilots18 9% 55% 19% 71% 29% 78% 38% 83% 63% 17%
Spokane 6% 48% 12% 64% 21% 72% 28% 76% 72% 24%

* Totals in the first eight columns are cumulative, i.e., "within past month" includes the total from "within past week."

Table 5.7: Frequency and duration of bicycling during "typical" week

  Frequency Duration
0 days 1 to 3 days 4 to 5 days 6 to 7 days 10-29 minutes 30-59 minutes 1 Hour or more
Columbia 86% 10% 3.2% 1.3% 15% 42% 43%
Marin 78% 17% 4.9% 0.6% 13% 24% 63%
Minneapolis 79% 16% 4.4% 0.8% 13% 31% 55%
Sheboygan 77% 14% 6.0% 2.8% 22% 33% 45%
Avg. for Pilots8 80% 14% 4.5% 1.3% 14% 33% 53%
Spokane 82% 13% 4.4% 1.2% 8% 37% 55%

Table 5.8: Frequency and duration of walking during "typical" week

  Frequency Duration
0 days 1 to 3 days 4 to 5 days 6 to 7 days 10-29 minutes 30-59 minutes 1 Hour or more
Columbia 17% 31% 27% 25% 29% 41% 30%
Marin 9% 28% 30% 33% 18% 44% 38%
Minneapolis 11% 24% 29% 35% 21% 37% 42%
Sheboygan 24% 28% 21% 26% 25% 39% 37%
Avg. for Pilots8 16% 28% 27% 29% 24% 40% 36%
Spokane 21% 31% 25% 23% 26% 39% 35%

Environmental, energy, and health impacts

Using these estimates of baseline travel behavior and the corresponding survey data in 2010, the research team will estimate the environmental, energy, and health impacts of the NTPP. The composition of the personal vehicle fleet can be determined from vehicle registration and other sources. U.S. Environmental Protection Agency (EPA) mileage estimates for the different models can then be applied to calculate gallons of fuel saved. Similar EPA tools for estimating urban emissions (based on miles driven and speed) will be used to calculate the avoided quantities of various types of pollutants. The research team also will estimate the health benefits of increased physical activity from the NTPP using survey data on the number of walkers and bicyclists and the frequency and duration of their activities.


The research team dealt with a broad range of issues inherent in this type of data collection, including: the impact of weather conditions; the need to over-sample to increase the number of survey respondents who bicycle and walk; and reporting biases among respondents. The research team provides more detail on these issues at

While there is considerable variation across the communities, walking and bicycling for utilitarian purposes combined to replace approximately 0.25 to 0.75 mile of driving per day, per adult resident. About 70 percent of this avoided driving was due to walking. While walking trips are shorter, far more people walk than bicycle on any given day.

The research team estimates that walking and bicycling for utilitarian purposes reduce driving by about 1 to 4 percent, depending on the community. Because of the large populations involved and the ongoing nature of this reduction, this seemingly modest contribution leads to significant long-term results: the total reduction in all four program communities, over the course of an entire year, is estimated to be in the range of 156.1 million miles of avoided driving.19

This total excludes recreational trips and trips by children.

Previous Return to TOC Next
Updated: 12/10/2014
HEP Home Planning Environment Real Estate
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000