Contextual Factors Affecting Physical Activity
The preceding chapter documented many long-term trends in the way the U.S. population lives, works, and travels that have sharply reduced the physical demands of daily life. The persuasive scientific evidence on the importance of physical activity for health presents a challenge: to increase physical activity in a highly technological society with a built environment that is already in place and has evolved over a long period of time. This chapter explores the socioeconomic and institutional context that has resulted in the current situation and holds the key to change. It starts with a discussion of the various factors that affect the individual’s choices about engaging in physical activity. The chapter then turns to the institutional and regulatory forces behind the decisions of planners, engineers, developers, elected officials, and others over the years that have shaped the built environment in place today.
FACTORS AFFECTING INDIVIDUAL CHOICE
As discussed in Chapter 1 (Figure 1-1), physical activity behavior is influenced by both individual characteristics and the social environment. Whether an individual is physically active depends on demographic characteristics such as gender, age, and ethnic background, and on socioeconomic characteristics such as education and income level. It also depends on at least three other factors, the latter two of which are external to the individual: (a) attitudes,
preferences, motivations, and skills related to the behavior; (b) opportunities or constraints that make the behavior easier or more difficult to perform; and (c) incentives or disincentives that encourage or discourage the desired behavior relative to competing activities. Each of these factors is discussed in turn in this section. Much of the discussion is based on self-reported survey data and focus groups. Relative to observational surveys, self-reported data often provide unreliable estimates because of problems with recall or the well-established tendency of survey respondents to give socially desirable rather than completely truthful answers (see Chapter 2). Results from focus groups cannot be generalized to the population at large. Nevertheless, self-reports and focus groups are the only way to obtain insight into attitudes and motivations that help explain behavior. This type of information is particularly important because the determinants of physical activity behavior are not well understood.
The Behavioral Risk Factor Surveillance System (BRFSS) and the National Health Interview Survey (NHIS) have revealed that physical activity levels of U.S. adults decline with age and are lower among women, ethnic and racial minorities, those with less education and low income levels, the disabled, and those living in the southeastern region of the United States (see Chapter 2).1 These results have been corroborated by numerous other studies.2 For example, younger age is positively associated with physical activity, as are university education and higher income levels. Although comparisons by race are often obscured by socioeconomic variables, some studies have shown that ethnic minorities, particularly African American and Hispanic women, are less likely to adopt and maintain active lifestyles. Other personal barriers to walking
and an active lifestyle cited in the literature include state of personal health and physical disability; lack of time, motivation, and energy; and lack of self-esteem. Further elaboration is not provided here because the committee has chosen to focus its discussion on physical activity behaviors linked with the built environment, such as nonmotorized travel and attitudes toward walking and cycling.
Attitudes, Preferences, Motivation, and Skills3
Several national surveys have been conducted in recent years to determine the public’s attitudes toward walking and cycling, as well as the frequency and purpose of these behaviors. Two of the surveys were sponsored by organizations that advocate walking and cycling—the Surface Transportation Policy Project and America Bikes. They found positive attitudes among respondents toward both walking and cycling and strong support for investments that would make communities more friendly to these modes (BR&S 2003; America Bikes 2003).
A national survey of walking and cycling sponsored by the National Highway Traffic Safety Administration and the Bureau of Transportation Statistics (BTS) and administered by the Gallup Organization during summer 2002 found that 8 of 10 respondents aged 16 or older had taken at least one walk of 5 minutes or longer in the past 30 days; fewer than 30 percent, however, reported having ridden a bicycle at least once (DOT 2003). When asked the primary purpose for walking trips, respondents most commonly cited exercise or health reasons (27 percent), personal errands (17 percent), and recreation (15 percent). The primary purposes for cycling trips were recreation (26 percent) and exercise or health reasons (24 percent).4 Survey results
should be interpreted with caution because of low response rates.5
Another survey, conducted as part of the BTS monthly Omnibus Household Survey (BTS 2003), queried adults aged 18 and older about walking and cycling, among other forms of transportation, during 2001–2002.6 [These results should also be interpreted with caution because of problems with response rates and sampling as detailed in a TRB report (2003).] Approximately 72 percent of those interviewed reported having walked, run, or jogged outside for 10 minutes or more at least once during the month prior to the survey (BTS 2003). Nearly 60 percent of those who walked, ran, or jogged (about 40 percent of all respondents) reported spending about 30 minutes on these activities an average of 13 days per month, as compared with the recommended minimum of 30 minutes per day of moderate-intensity activity on 5 or more days per week (see Chapter 2). Nearly 20 percent of respondents reported a longer duration of activity, but 40 percent reported no outside walking, running, or jogging (BTS 2003).7 Only 16 percent of adult U.S. residents reported cycling outside during the month prior to the survey—spending just over 1 hour per day cycling on an average of 6 days per month (BTS 2002).
The Omnibus survey also inquired about the reasons for walking and cycling. Slightly more than three-quarters of those respondents who walked, ran, or jogged reported that they did so
primarily for exercise or recreation. Another 15 percent walked for personal errands, and only 7 percent to get to work or as part of their job (BTS 2003, 1).8 Similarly, the primary reasons for cycling were for recreation (54 percent) or exercise (33 percent); only 6 percent reported commuting by bicycle to get to school or work or as part of their job (BTS 2002).
In sum, the surveys indicate that walking is more prevalent than cycling, but reported levels of walking appear to fall short of recommended daily guidelines. To the extent that Americans report walking and cycling, the primary reasons appear to be for exercise and recreation. These results correspond with the behavioral data from public health surveys discussed in the previous chapter showing a trend toward increased leisure-time physical activity.
Market research has also been conducted to probe the reasons for engaging in physical activity. Several studies cited by Kirby and Hollander (2004)9 found that adults’ dominant beliefs about moderate physical activity were that it results in feeling better or more energetic, helps reduce stress, and improves physical condition (e.g., feeling less out of breath, stronger). Focus groups with older Americans revealed similar beliefs.10 Notably absent from the survey and focus group results is any mention of the longer-term benefits of physical activity identified by the health community and summarized in Chapter 2, such as disease prevention. The positive health effects of physical activity may have been assumed by the survey and focus group respondents, but the results may also reflect the value placed by many people on more immediate benefits, such as those enumerated above. In any event, the market research
results underscore the importance of understanding the beliefs and attitudes of those whose behavior one wishes to reinforce or change. As marketers are well aware, beliefs and attitudes are likely to differ across subpopulations. For example, a single mother holding two jobs is likely to be motivated to become more physically active by information showing how physical activity can be fit into her busy daily routine, whereas a teenager is likely to be more motivated by information that physical activity will make her more fit and attractive. Thus, tailoring interventions to specific groups is likely to prove more effective than delivering mass messages about the benefits of being physically active.
Finally, while beliefs, attitudes, and preferences have a role in determining a person’s physical activity habits, cognitive and behavioral factors come into play as well. To become more physically active, for example, individuals can self-monitor the target behavior, learn how to set realistic and achievable goals, monitor progress toward those goals, identify barriers to achieving the goals, use problem-solving techniques to overcome those barriers, and identify and use peer and family social support to help achieve lasting behavioral change. Interventions using these methods, which are based on psychosocial theories and models such as social cognitive theory and motivational readiness, have been applied successfully in randomized, controlled clinical trials to evaluate methods of helping sedentary adults become more active (Kohl et al. 1998; King et al. 1998; Dunn et al. 1999). The committee is unaware, however, of published reports in which cognitive and behavioral interventions have been incorporated into designs that also encompass environmental and socioeconomic factors.
Opportunities and Constraints
The results of the surveys reviewed in the previous section and those of other large health surveys presented in Chapter 2 indicate that the majority of Americans are not acting sufficiently on their inclinations to meet recommended levels of total daily physical activity. Personal motivation is one likely explanation, but it is instructive
to examine other possible factors—real or perceived—that may be preventing the desired behavior, with particular attention to the built environment as a potential barrier. It should be noted that, although walking and cycling are discussed together here, they generally involve different infrastructure and user characteristics. For example, in urban areas, cycling typically is forbidden on sidewalks and confined to certain streets or bicycle lanes that share the right-of-way with automobiles. Cycling on pedestrian paths can pose a danger for those who are walking. These differences should be kept in mind in interpreting survey results. For example, these differences are likely to make cyclists more concerned with infrastructure facilities for safety.
The Gallup survey discussed above revealed that the primary reasons for not walking or cycling were personal (disabilities or other health impairments), weather- or time-related, or equipmentrelated (did not own or have access to a bicycle) (DOT 2003). Environmental factors (no safe place to ride or walk) were mentioned by only a small fraction of respondents (approximately 3 percent) (DOT 2003). Three of four adults reported being “very” or “somewhat satisfied” with the design of their communities for pedestrian safety. Nevertheless, when asked to recommend changes in their communities, presumably to make walking safer, about one-third of those polled suggested providing pedestrian facilities, such as sidewalks, traffic signals, lighting, and crosswalks. Satisfaction with the cycling environment was considerably lower. Only half of those polled were “very” or “somewhat satisfied” with their communities’ designs for cycling safety. Nearly one-half of all respondents recommended new bicycle facilities, such as bicycle trails, paths, lanes, racks, traffic signals, lighting, and crosswalks. The survey results suggest that, even for those favorably disposed to walking and cycling, changes to the physical environment that would enhance the safety and ease of engaging in these activities could make a difference.
Results of other surveys suggest that environmental factors may play a more dominant role depending on the activity—for example, transporting children to school. As noted earlier, the private
vehicle has become the primary mode of school travel (Dellinger and Staunton 2002). Long distances, dangerous traffic, and crime have been mentioned as the main barriers to children walking and cycling more to school (Dellinger and Staunton 2002; BR&S 2003).11 In fact, children (aged 5 to 18) of parents who reported no barriers [16 percent of all respondents to the Centers for Disease Control and Prevention’s (CDC’s) HealthStyles Survey reported by Dellinger and Staunton] were six times more likely to walk or bicycle to school than those whose parents cited one or more barriers.
Interventions to mitigate such barriers can be effective. For example, the California Safe Routes to School Program has provided more than $40 million to municipalities and counties to improve the safety and viability of walking and cycling to school. Typical projects include sidewalk construction and improvements, pedestrian and bicycle crossings, and traffic controls to improve the safety of street crossings (Boarnet 2004). A before-and-after evaluation of projects associated with 10 schools across the state found that walking and cycling had increased, with larger effects if the project was along the child’s usual route to school (Boarnet et al. 2004).12 The Marin County Safe Routes to School Program is a good example of a comprehensive approach to reducing barriers for children walking and cycling to school that appears to be working (see Box 4-1).
Constraints and barriers to physical activity are perhaps best illustrated in those low-income neighborhoods where crime, disinvestment, and isolation can be major deterrents to walking and cycling for many residents. Low-income urban populations
The Marin County, California, Safe Routes to School Program
The Safe Routes to School Program in Marin County is one of the programs funded by the California Safe Routes to School Program. Marin County has established a grassroots program that is getting more children to walk and bicycle to school.
Program components include mapping of routes and infrastructure improvements to improve access to schools by walking or bicycling, special events and contest promotions, new concepts such as “walking school buses” and “bike trains” to generate and maintain the interest of the community, and a well-integrated communication and promotion strategy. Safe Routes task forces collaborate with public works and law enforcement staff to develop and implement an improvement plan, apply for funding, and effect improvements such as crosswalks and signage to make it easier and more convenient to walk and cycle to school. The California headquarters for the Safe Routes to School Program also provides materials, tips, and tools for community volunteers and organizations. These include a walkability checklist, sample letters to parents in 13 languages, a “guide to success” with instructions on how to create a walking school bus and a bike train, and a guide on how to create safe drop-off points for children walking to school (see www.cawalktoschool.com/dropoff_zones.php). In addition, the California headquarters partners with the state health department’s injury control center to give its safety messages even more credibility with parents.
Most important, the program appears to be working. At the second-year mark of the commencement of the program in Marin County, 15 participating public schools reported an increase in walking (64 percent), bicycling (114 percent), and carpooling (91 percent) and a decrease in private vehicles carrying only one student (39 percent) (Staunton et al. 2003).
SOURCE: Kirby and Hollander 2004.
exhibit the highest levels of walking and bus transit use13 for utilitarian travel out of necessity (Pucher and Renne 2003), but they engage in much less discretionary physical activity than other groups (see Chapter 2). Interventions such as the Sisters Together Program (see Box 4-2), which attempt to address issues of regaining control over one’s environment (e.g., safe walk routes) and combating isolation (e.g., walking buddies), may help overcome barriers to recreational physical activity for some low-income urban populations. Not all low- or moderate-income neighborhoods are affected by fears of crime, however. Physical inactivity of their residents must derive from other causes.
Concern for personal safety can also play a role in the use of pathways for walking and jogging in urban and regional parks. Surveys and focus groups have shown that adults, particularly older adults and female minorities, perceive unsafe footpaths and other recreational areas for exercise as deterrents to walking and other physical activity (Hahn and Craythorn 1994; King et al. 2000; Booth et al. 2000).
Crime and deteriorated neighborhoods are less likely to be an issue in rural settings, where natural scenery (open fields) and lightly traveled rural roads provide opportunities for walking and cycling. For the rural poor, however, isolation and long distances between destinations may limit these activities (Brownson et al. 2000 in Kirby and Hollander 2004).
Providing opportunities for walking and cycling may not be sufficient to change behavior, however, particularly for certain types of travel, such as commuting. Time constraints, long distances between destinations, and the mobility afforded by the automobile make traveling by personal vehicle the preferred option for many commuters. A recent study of commuting behavior in three neighborhoods in the San Francisco Bay Area—one urban and two suburban—attempted to separate the effects of household location preferences from the spatial characteristics of residential neighbor-
The Sisters Together Program
This obesity prevention pilot program supported by the National Institutes of Health (NIH) and the National Institute of Diabetes and Digestive and Kidney Diseases (www.niddk.nih.gov/health/nutrit/pdf/SisPrmGuide2.pdf) began by targeting young black women in three inner-city neighborhoods of Boston. The campaign focused on creating positive messages to generate normative change and involving existing community programs to build sustainability.
The Sisters Together initiative developed a coalition of programs and people in the community, targeting both healthy eating and moving more (www.hsph.harvard.edu/sisterstogether/move.html). In an effort to suggest activities that would resonate with their target audiences, program staff developed tips on dancing, not just walking: “Turn on your favorite music and dance to three songs a day three times a week. It gets your heart pumping, your body moving, and your mind feeling great.” A web page and brochure provided safe walking routes around the city. Radio public service announcements offered women a chance to sign up for a neighborhood walking group if they came to a 2-mile warm-up walking event. Making it easier for women to locate a walking buddy helped promote a positive social norm with regard to walking. The program’s Why Walk cites the top three benefits of walking validated by research—“Walking will … give you more energy, make you feel good, and help you relax.”
A traditional method—the bounce-back card—was used to obtain feedback from the target audience and partners on how the program was working and what could be improved. Community partners were engaged to be the sustaining force behind the program once NIH funding for the pilot project ended. Rudd et al. (1999) describe the community development model employed in this project, but no longer-term evaluation data could be located.
SOURCE: Kirby and Hollander 2004.
hoods that help shape travel patterns (Schwanen and Mokhtarian 2004). The researchers found that, even after controlling for sociodemographic characteristics, mobility limitations, personality and lifestyle types, and travel attitudes, suburban-minded residents of the urban neighborhood (i.e., urban dwellers who preferred lower-density environments) commuted by private vehicle far more than their urban-minded neighbors (those who preferred higher-density environments such as the one in which they lived). Similarly, urban-minded suburban dwellers commuted by car about as often as their suburban-minded neighbors. However, the differences in commuting behavior across neighborhoods were greater than those within neighborhoods, which indicates that neighborhood structure itself has an autonomous effect on travel choices. Commuting by personal vehicle strongly prevails in suburban neighborhoods in which residents have fewer mode choices, longer distances to travel, and lifestyle preferences for low-density living. In urban neighborhoods where densities are higher, travel distances are often shorter, and travel options are greater, transit achieves a higher commute mode share than in suburban neighborhoods (Schwanen and Mokhtarian 2004).
Incentives and Disincentives
Lowering the cost of a desired behavior and raising the price of an undesired behavior can be an effective strategy for motivating behavior change. The choice to walk, bicycle, or combine either with transit may require such incentives and disincentives. For example, a combination of providing transit fare subsidies through the workplace and either cashing out14 or raising parking fees could help level the playing field between driving and taking transit and encourage greater transit use (Shoup 1994; Shoup 1997).15 (Of course, the
workplace must be accessible to transit for the employee to take advantage of the transit fare subsidy.)
In addition to monetary incentives, such strategies as reducing the time cost of physical activity—making it easier and more convenient to be physically active—can be effective. In the previously cited national survey conducted by BR&S on Americans’ attitudes toward walking, inconvenience (destinations being too far) and time were the primary reasons cited for not walking more (BR&S 2003). Likewise, a community intervention in Wheeling, West Virginia,targeting sedentary adults aged 50 to 65 found that time and schedule were the major stated deterrents to being physically active (see Box 4-3) (Reger et al. 2002 in Kirby and Hollander 2004). With the tagline “Isn’t it time you started walking?” the intervention attempted to make the case that walking is an activity easily accommodated and integrated into one’s daily routine. Similarly, one of CDC’s earliest campaigns to promote the benefits of moderate physical activity—with the slogan “Ready, Set, It’s Everywhere You Go”—sought to underscore that moderate-intensity activities could easily be part of the daily routine (see Box 4-4) (Kirby and Hollander 2004). Lack of evaluation of such projects for their effects on physical activity levels, however, makes it impossible to predict the benefits of such approaches.
Improving access by shortening distances between destinations is more difficult to address. Such a strategy requires moving one’s residence or employment or both, or locating facilities in closer proximity to one another—a topic discussed in the following section.
To create incentives for physical activity, one must also consider the competition for the desired behavior. For example, a competitor for engaging in recreational physical activity might be watching television (Kirby and Hollander 2004), although this need not be the case. Many individuals watch television, or could be encouraged to do so, as they walk a treadmill either in their homes or at sports clubs. Another, more challenging competitor is the car, particularly for utilitarian physical activity. As Schwanen and Mokhtarian (2004) found, improving nonautomobile mode choice options can help encourage transit use and related walking, but these modes
Wheeling, Virginia, Intervention for Sedentary Adults Aged 50–65
The work of Reger et al. (2002) is a good example of using marketing principles to design a behavior change effort. First, the researchers decided on a specific behavior—walking—for a specific target audience—sedentary adults aged 50 to 65 in Wheeling, West Virginia.
Initially, most of the programmatic effort was focused on promotion and price variables. Formative research had found that sedentary and irregularly active people and regular walkers share similar attitudes and normative beliefs but exhibit strong differences related to their perceived control over time and scheduling. The major “price” of walking for the sedentary adult was “time.” Thus, the ensuing promotional strategy was focused on perceived control issues and positioned walking as an activity that was easy to accommodate and integrate into one’s daily routine. A pithy tagline was developed to address the time issue: “Isn’t it time you started walking?”
The formative research also identified optimal promotional channels for reaching the intended audience. A combination of paid advertisements on television and radio and in newspapers was developed. In addition, non–mass media channels were tapped, such as the Wheeling Medical Society, physician prescriptions, work site wellness challenges, and community walking events. After initial campaign efforts, the researchers remained in contact with community participants, who suggested various improvements in community walking facilities. The mayor was engaged; a community task force was established; and collaboration with the National Park Service, the state Department of Transportation, and a local Rails-to-Trails group was initiated.
The following quotation illustrates the power of addressing perceived behavioral control:
My biggest surprise about walking was the fact that I actually could do it. When my brother found out how far I was going he talked me into racing. I didn’t think I could do it, but luckily I won my first race and from then on I was hooked. I loved the people, the atmosphere, and the challenge.
SOURCE: Kirby and Hollander 2004.
CDC’s “Ready, Set, It’s Everywhere You Go” Campaign
CDC launched one of the earliest campaigns to encourage moderate physical activity as opposed to “exercise,” which had a more high-intensity, time-demanding connotation. CDC’s “Ready, Set, It’s Everywhere You Go” campaign relied mainly on communication techniques to introduce the notion of moderate-intensity activity that could be part of an adult’s routine daily life. Formative research was conducted to validate audience segmentation and develop materials and promotional messages.
CDC produced a marketing kit for use by community-based organizations in their local efforts. It comprised three parts: (a) marketing strategies for physical activity, (b) ways to work with the media, and (c) the development of physical activity programs and events. The kit included a colorful poster and print ad emphasizing that people can be active doing routine activities such as yard work and walking the dog. The kit was designed to help health professionals and community-based organizations identify adults who wanted to become physically active and reach them with accurate and positive messages that had been tested with the same target audience. Target audiences specifically wanted materials that were family-friendly and conveyed the idea of having fun and being energetic.
By analyzing market research data and conducting focus groups and interviews, CDC developed a detailed picture of the intended audience. Research revealed that the majority of the intended audience was 18 to 45 years old, educated, middle-income, and female. Since 71 percent were married, 74 percent were employed, and 58 percent had live-in children, it was not surprising that these adults reported having little time for themselves after meeting their household, job, and family demands. Few of them considered themselves rugged or athletic; rather, they described themselves as interesting, friendly, caring, mature, fun, smart, honest, and content. As a whole, the target group members reported that they did not enjoy vigorous
“exercise.” However, they did view “physical activity” as fun and enjoyable and were pleased to learn that it is important to their health and well-being. Participants believed that internal motivation, pleasant and manageable activities, support from family and friends, and convenience would help them become more physically active.
A variety of other barriers stood in the way of physical activity. Participants in the research cited such barriers as long work hours, being tired at the end of the day, lack of confidence in their athletic ability, and family priorities. Program planners reasoned that helping people understand that physical activity is “everywhere” they go and easy to do could help them become more active. The “Ready, Set, It’s Everywhere You Go” materials, community kit, radio spots, and posters were audience tested—an important step in formative evaluation for marketing. The project has not been evaluated for its impact on changing behavior related to physical activity or healthy eating, however.
SOURCE: Kirby and Hollander 2004.
cannot always compete with the mobility and convenience afforded by the personal vehicle. Fortunately, total levels of physical activity matter, not whether an individual drives rather than walks or cycles to work, and even modest increases in total physical activity levels can have a positive effect on health (see Chapter 2). Nevertheless, it must be acknowledged that some individuals may view physical activity—even in small amounts—as unpleasant.
INSTITUTIONAL AND REGULATORY CONTEXT
Whereas the previous section examined the demand for physical activity, this section looks at the supply side of the link between the built environment and physical activity. Specifically, it examines
the institutional and regulatory arrangements and policies that, over time, have created the built environment in place today.16
Zoning and Land Use Ordinances
In the United States, local governments are responsible for developing comprehensive plans and establishing land use regulations that determine how a community will develop. The authority to create zoning and subdivision controls and building regulations, which have the force of law, is a powerful tool in establishing the design requirements and physical context of a community’s development. Most zoning regulations and subdivision controls regulate two factors thought to be closely linked to a community’s walkability and bikeability—development densities and mixing of land uses.
Zoning was introduced by urban reformers in the United States in the early twentieth century to help alleviate the impacts of urban overcrowding on disease and illness. New York’s Zoning Ordinance of 1916—the first enacted in the nation—was created for the express purpose of limiting development densities and thereby improving public health (Jurgensmeyer and Roberts 1998 in Meyer and Dumbaugh 2004). Early zoning regulations prohibited mixing of land uses to segregate those that would be incompatible, such as residential and high-polluting industrial uses. As they evolved, zoning laws also operated to reinforce economic and racial separation. Exclusionary zoning in wealthier communities restricted certain types of development, such as multifamily housing construction, and established stringent standards, such as minimum lot sizes or housing square footage, that had the effect of keeping housing prices high and thus excluding lower-income families (NRC 1999). Once such zoning limits were in place, they tended to be reinforcing. Households that moved to a community with single-family zoning viewed efforts to incorporate more affordable multifamily housing as a threat to their property values (Fischel 1999 in NRC 1999).
Municipal street designs and zoning requirements regarding parking have also had an important impact on the development of communities. Early municipal street designs incorporated in guidelines issued by the U.S. Federal Housing Administration in 1935 recommended that residential streets be designed to “discourage through traffic, have a minimum paved width of 24 feet, use cul-de-sacs17 as much as possible, and avoid excessive planting in the front yards to have a ‘more pleasing and unified effect along the street’ ” (FHA 1935). Municipal street design standards were also developed to take into account requirements for providing emergency services. Wide streets were believed necessary to accommodate the worst-case scenario—two high-rise ladder trucks jockeying for position on a dead-end street (Duany et al. 2000 in Meyer and Dumbaugh 2004).
Most community zoning codes require that a minimum number of parking spaces be provided per unit or per 1,000 square feet to accommodate the maximum demand for parking (Meyer and Dumbaugh 2004). In most cases, this number is greater than what is needed to handle “normal” demand and results in an oversupply of parking, particularly in suburban areas.
Taken together, zoning and land use controls can make it difficult to provide many of the characteristics associated with walkable and bikeable communities today (Meyer and Dumbaugh 2004). For example, low-density development often results in long distances between destinations, and research suggests that walking and cycling are highly sensitive to distance as compared with automobile travel, particularly travel for utilitarian purposes. Walking speeds are about 3 miles per hour (mph), and average bicycle speeds are about
8 mph, depending on topography. Some planners suggest that walkable communities should have destinations within roughly ¼ to ½ mile of the point of origin (Seneviratne 1985). Bicycle destinations can be located slightly farther—2 to 3 miles from the point of origin (Meyer and Dumbaugh 2004). These guidelines are simply rules of thumb—individuals may well be willing to walk and cycle longer distances—but they underscore the competition posed by faster transport modes.
Separation of land uses also tends to increase the distances between destinations and creates a monotonous environment that may not be conducive to walking or cycling. In today’s economy, the rationale for separating land uses is less compelling; many service-related work places are compatible with residential uses. Minimum parking requirements accommodate driving to most destinations and take up space that could be used for neighborhood amenities, such as parks and green spaces. Finally, wide residential streets with long straight sight lines and few trees contribute to vehicle speeding, creating a potentially dangerous environment for pedestrians and cyclists (Meyer and Dumbaugh 2004).
Urban Design Features
Design features are also thought to affect the form of community development and travel choices. Such features refer to both the aesthetic appeal and the function of buildings, streetscapes, and public spaces, which can be designed in ways that can encourage walking and transit use, particularly in the neighborhood, but also around work sites. Table 4-1 lists five such urban design features—density of development, land use mix, street connectivity, street scale, and aesthetic qualities—and describes how they can be measured.
The writings of Jane Jacobs in the 1960s (Jacobs 1961) and Kenneth Jackson in the 1980s (Jackson 1985) critiqued the loss of neighborhood scale and community life in the automobile-dependent suburbs created largely after World War II. Whyte and Appleyard’s studies of public spaces and livable streets in the early 1980s (Whyte 1980; Appleyard 1981) and architect Peter
TABLE 4-1 Examples of Design Features of the Built Environment
Description and Possible Measures
The amount of activity found in an area—usually defined as population, employment, or building square footage per unit of area and measured as people per acre or jobs per square mile. Floor–area ratio, the ratio between the floor space in a building and the size of the parcel on which the building sits, is another density measure.
Land use mix
The relative proximity of different land uses (e.g., homes, stores, offices, parks) within a given area—no standard measure.
The directness and availability of alternative routes from one point to another within a street network—measured by the number of intersections per square mile, average block length, and so forth.
The three-dimensional space along a street as bounded by buildings or other features—typically described as “human-scale” or “automobile-scale”—measured by the average building setback or by the ratio between building heights and street widths.
The qualities that contribute to the attractiveness or appeal of a place, such as the design of buildings (size and orientation of windows), landscaping, lighting and benches—the most intangible of the design features—more often described than measured.
SOURCE: Adapted from Handy et al. 2002, 66.
Calthorpe’s vision of more walkable and livable communities (Calthorpe 1993) gave rise to a set of design concepts collectively known as the “new urbanism.” The movement emerged in the late 1980s through architects who designed smaller, people-oriented communities with a small-town feel and a village scale. The goal was to establish a sense of community—often missing in newly developed neighborhoods—by creating human-scale housing and streets, mixing land uses, providing vibrant public spaces, and getting people out of their cars (Boarnet and Crane 2001). New-urbanist developments emphasize such design elements as front porches, sidewalks, and common public spaces as gathering places for community activities.
The claim is made that walking will increase if the activities of daily living (e.g., parks, neighborhood shopping) are within walk-
ing distance and linked to where people live and work by an interconnected network of streets, sidewalks, and paths (Handy et al. 2002). These goals can be achieved by straightening of streets to improve connectivity (i.e., use of grid street patterns), “calming” of traffic, more compact land uses with a diversity of destinations, and inviting street environments with amenities such as street furniture and plantings (Boarnet and Crane 2001). Transit use should also increase with more compact land use and clustering of shopping and housing near rail or bus nodes. The evidence supporting the effects of urban design features on walking, including walking to access transit facilities, is reviewed in the paper by Handy commissioned for this study (Handy 2004) and summarized in Chapter 6.
Private developers and lenders are ultimately responsible for the development and construction of local residences and commercial facilities. Recent surveys (Levine and Inam 2004; Smith-Heimer and Golem 2001 in Kirby and Hollander 2004) have revealed that real estate developers perceive considerable market interest in walkable communities and support developments with greater density and more mixed uses than regulations allow, particularly in inner-suburban areas. A panel discussion with California developers yielded evidence of good market support for transit-oriented development projects that involve mixed-use development near transit stations (Smith-Heimer and Golem 2001).
In view of these findings, why are there not more walking- and cycling-friendly developments—often called neotraditional or new-urbanist developments—and transit-oriented development projects in response to market demand, particularly since such developments are in short supply? Levine and Inam (2004) suggest that a large majority of developers perceive local zoning controls and the related costs of pursuing variances as the primary obstacles to neotraditional developments. For example, among developers
who proposed neotraditional developments for which variances were granted, the density was reduced by the community planning process for approximately 80 percent, mixed-use characteristics were reduced for nearly 50 percent, the housing types were changed for nearly 30 percent, the share of mixed-use development was changed for one-third, and changes were made in pedestrian or transit orientation for nearly 20 percent (Levine and Inam 2004). Other surveys have revealed that public resistance to densification and neighborhood opposition to mixed-use development are significant barriers to neotraditional projects (Logan et al. 2001 in Kirby and Hollander 2004). These findings indicate that, when faced with accepting higher densities or mixed-use development—changes that may be perceived as negatively affecting property values—not all consumers are as supportive of more walking-, cycling-, and transit-friendly communities as their survey responses would suggest.
Financial institutions can also be a barrier. Despite some developers’ perception that neotraditional developments can be profitable and the findings of studies confirming that this is in fact the case (Eppli and Tu 1999 in Kirby and Hollander 2004),19 institutional lenders are risk averse. Typically, they look for projects that are compatible with other developments in the local market (i.e., that meet local zoning and subdivision controls). Financing of mixed-use developments can be particularly problematic because many lenders have experience in dealing with only one type of development (Meyer and Dumbaugh 2004; Kirby and
Hollander 2004). An informal survey of institutional lenders in the Atlanta, Seattle, and Boston markets conducted by Meyer and Dumbaugh (2004), for example, revealed that lenders are not averse to neotraditional developments as long as such developments are not expressly prohibited by local zoning and are not the first such development in an area. The presence of profitable existing neotraditional developments in a local market and evidence of other public and private investment in a transit-oriented development, of course, increase the acceptability of similar new projects (Smith-Heimer and Golem 2001).
Relaxing zoning and financing barriers to enable more neotraditional development for those who would like to locate in such communities would require changes on many fronts—not the least of which would be to educate the public, elected officials, and the real estate community in how these communities can work and be implemented. The more it can be shown that such communities can be profitable and not reduce surrounding property values, the more acceptable they will become (Meyer and Dumbaugh 2004). In addition, instead of overturning long-standing zoning regulations and ordinances, it may be easier to win support through more targeted approaches, such as overlay districts20 and incentives (Meyer and Dumbaugh 2004; see Box 4-5 for two examples).
School Design and Location
Decisions about school design and location are largely independent of the processes that drive other forms of community development. Planning for educational facilities is the responsibility of local school boards, which are typically composed of elected representatives. Because nearly all school boards are semiautonomous, they—not local governments that have a strong interest in the
Two Approaches to Relaxation of Zoning Regulations and Controls
Overlay districts. Changing a community’s land use zoning is often a difficult political undertaking. One of the approaches used to provide a higher level of urban design while maintaining the underlying zoning is to use overlay zones targeting specific development characteristics. A good example is Portland, Oregon’s, Light Rail Transit Station Zone (Portland Metro 2000). This overlay zone “allows for more intense and efficient use of land at increased densities for the mutual reinforcement of public investments and private development. Uses and development are regulated to create a more intense built-up environment, oriented to pedestrians, and ensuring a density and intensity that is transit supportive.” Actions include prohibition of parking garages within a specified distance of a station, a 50 percent reduction in the minimum number of parking spaces required within 500 feet of a light rail alignment, and the requirement of a high level of streetscape landscaping.
Neotraditional development incentives. Restructuring long-standing land use ordinances that have been the basic approach to community development is also difficult. A more appealing approach for encouraging neotraditional development and use of nonmotorized transportation is to provide incentives to both developers and communities. For example, in specified districts, developers could receive income tax credits for certain types of development, reductions in permit fees and other procedural requirements, and relaxation of other zoning requirements that might save the developer money. Regional planning agencies could reward communities that provided approvals for neotraditional developments. In the San Francisco Bay Area, for example, the metropolitan planning organization provides a certain amount of money to a community for every bedroom constructed within a certain distance of a transit station. These funds can be used by the community for any purpose. By using incentives, policy makers participate in the development market, but not in the traditional regulatory way.
SOURCE: Meyer and Dumbaugh 2004.
overall development of a community—make decisions about school design and location.
The trend in school design has been to develop bigger schools to lower costs through economies of scale. The large land requirements recommended by the standards-setting organization for school facilities21 make incorporating these schools into existing communities difficult. The trend is to locate the facilities on large suburban tracts, which necessitates driving or busing students from surrounding communities. Because local school boards are responsible only for on-site circulation and not for access from the surrounding street network, means of accommodating walking and cycling to school are rarely planned as part of a school construction project. Similarly, school bus routes and safe access to bus stops are under the purview of local school boards; little coordination with local community planners is involved (Meyer and Dumbaugh 2004).
Some states and communities have begun to reevaluate the desirability of requirements that encourage the building of large new school campuses (EPA 2003). For example, South Carolina recently eliminated state-mandated acreage requirements for new schools that may make it easier for existing schools to be renovated. Neighborhood school initiatives in Wisconsin and Minnesota have resulted in retaining many elementary schools within walkable neighborhoods.
More coordination between local school boards and local government planners could help in addressing issues of school access and optimum school location. At a minimum, initiatives such as the previously discussed Safe Routes to School Program could help encourage more walking and cycling to school or walking to school bus stops.
Modeling of Transportation Needs
All major metropolitan areas are required by the federally supported planning process to have a regional transportation model
for analyzing network investment needs and alternatives (Meyer and Dumbaugh 2004). These models provide a very aggregate picture of regional travel and were not intended to handle the level of detail required to analyze or predict pedestrian or cycling trips. Major roads, such as freeways and arterials, are represented by network links in the models, but most local roads where pedestrians walk and bicyclists ride are not. Many walking and cycling trips are relatively short in distance, occurring within a traffic analysis zone. Statistical averages are normally used to represent intrazonal travel. This can understate the impact of mixing land uses or improving pedestrian ways within a city block in promoting walking because averages ignore any variation around the mean. Overall, regional transportation models generally do a poor job of representing non-motorized travel, which can understate the potential role of pedestrian facilities and bicycle paths as well as land use strategies in promoting walking and cycling trips (Meyer and Dumbaugh 2004).
The typical regional transportation forecasting model consists of four major steps: trip generation, trip distribution, mode choice, and trip assignment. The first step, trip generation, is a function of exogenously determined demographic patterns and economic activity in a region. The remaining three steps, which are followed sequentially, simply allocate trips among alternative destinations in trip distribution, alternative travel modes in mode choice, and alternative highway (and transit if appropriate) routes in trip assignment (TRB 1995). Trip assignment is based primarily on minimizing travel time through an iterative process that feeds back to mode choice, and sometimes to trip distribution, in an effort to equate initial with final travel time estimates. The outputs of the model are vehicle and passenger volumes on highway and transit routes, respectively.
The recent interest in policies supporting pedestrian and bicycle travel has led some metropolitan areas, such as Portland, Oregon, and the San Francisco Bay Area, to make advances in incorporating pedestrian and bicycle travel into their transportation models. They are still the exception rather than the norm, however. Furthermore, few metropolitan areas have integrated land use and travel demand
models, so that the effects of changes in urban form and design on travel behavior cannot be determined (Meyer and Dumbaugh 2004).
Models are important inputs to policy makers, but they represent only one element in the policy process. European cities typically have better bicycle and pedestrian facilities than most of their U.S. counterparts, but this is not an outcome of better models. Other policies, such as higher gasoline prices that discourage automobile travel, have likely played a far greater role in investment in nonmotorized facilities.
Roadway Infrastructure Design
A highly standardized approach to roadway infrastructure design has also played a major role in determining the design and development of communities (Meyer and Dumbaugh 2004). The design manuals used by highway and traffic engineers—the “Green Book” of the American Association of State Highway and Transportation Officials (AASHTO), which provides guidance on roadway design,22 and the Manual on Uniform Traffic Control Devices, which contains uniform standards for traffic control devices—date back as far as the late 1920s and 1930s. The primary focus was, and continues to be, on automobile and truck travel (Meyer and Dumbaugh 2004), although highway engineers are being encouraged by AASHTO and the Federal Highway Administration to interpret the guidelines more flexibly to better accommodate nonmotorized travel.
Design guidelines have evolved over the years, and many can accommodate the designs advocated for nonmotorized travel (Meyer and Dumbaugh 2004). The primary barrier appears to lie in how the standards have been implemented. The methods used to evaluate facility design requirements and performance, which are described in the following paragraphs, often tend to emphasize the needs of motorized travel at the expense of other modes (Meyer and Dumbaugh 2004).
Roadway design starts with a functional classification of highways into two environments—urban and rural—with three classes of roads under each—local roads, collectors, and arterials (AASHTO 2001).23 Embedded in this framework are the competing concepts of access—the ability to travel to and from properties located adjacent to the travel way—and mobility—the ability to travel with a reasonable level of performance (i.e., at uncongested and reliable speeds) (Meyer and Dumbaugh 2004). Both of these concepts are oriented to vehicular travel with little attention to the nonmotorized traveler, who typically travels at low speeds for short distances.
Once a road has been classified, the design speed, or the speed at which motorized vehicles can travel safely on the road, is prescribed (Meyer and Dumbaugh 2004). The design speed then determines the other geometric elements of the road, such as its curvature. With the exception of local streets, the AASHTO Green Book exhorts the engineer to “use as high a design speed as practical” (AASHTO 2001, 67). Thus, many features compatible with pedestrian and bicycle travel, such as lower vehicle speeds and trees adjacent to the travel way, are viewed as limiting vehicular throughput and creating potentially dangerous obstacles should a vehicle leave the road, particularly on higher-speed roads (Meyer and Dumbaugh 2004).
Another decision involved in road design is determination of the “design vehicle,” or the vehicle type that requires the greatest amount of maneuverability on a road. For example, if buses or trucks are expected to use a road, lane widths, turning radii, traffic signal timing, and the like will be oriented to these vehicle types. The resulting design facilitates higher operating and turning speeds for smaller passenger vehicles, which escalate the danger for pedestrians and cyclists who share the roads, and increases street-crossing distances for pedestrians and bicyclists (Meyer and Dumbaugh 2004).
The demands placed on municipal street design to accommodate emergency vehicles and the implications for walking and cycling have already been discussed.
Vehicle-oriented performance measures affect both facility design and planned improvements. The level of service (LOS) is used to describe how a transportation facility is performing. It ranges from LOS A, defined as free flow where traffic volumes are low and there is little or no restriction on traffic flow, to LOS F, characterized as highly congested with stop-and-go traffic (AASHTO 2001). A desired LOS is used as a performance criterion in designing a facility and is then incorporated into development site guidelines, local comprehensive plans, and state policies. Existing roads that perform at or below the desired LOS are candidates for capacity enhancements whose primary objective is improved vehicular performance.
It is difficult to change both the vehicular orientation of road design and performance evaluation that favors higher ranges of design standards, which are equated with “better” and “safer” performance, and standardized approaches perceived by engineers as reducing liability claims (Meyer and Dumbaugh 2004). Nevertheless, AASHTO and the Federal Highway Administration have encouraged engineers to take advantage of existing guidelines by designing more flexibly to accommodate such objectives as nonmotorized travel on certain types of roads (see Meyer and Dumbaugh 2004 for further detail). Targeted approaches, such as context-sensitive design24 and special design districts, make it possible to design roads to accommodate adjacent land uses and incorporate nonmotorized users for specific areas and projects without changing the entire underlying system of road classification and design criteria (see Box 4-6 for examples). Care must be taken to implement such projects with the safety of all users—motorized and nonmotorized—in mind so as not to increase the risk of crashes.
Two Examples of More Flexible Transportation Infrastructure Design Approaches
Context-sensitive design. Many state transportation departments are moving toward a more flexible project design process known as context-sensitive design or, more broadly, context-sensitive solutions. This movement began in the late 1990s, when several states launched initiatives to define better ways of designing roadways. Perhaps one of the best definitions of context-sensitive design is found in a technical memorandum from the Minnesota Department of Transportation: “Context sensitive design is the art of creating public works that are well accepted by both the users and the neighboring communities. It integrates projects into the context or setting in a sensitive manner through careful planning, consideration of different perspectives and tailoring designs to particular project circumstances” (Minnesota Department of Transportation 2000). Such efforts are beginning to focus attention on those aspects of infrastructure design in sensitive community contexts that enable greater flexibility in implementing design standards.
Special design districts. Rather than relying on the ability of design professionals to arrive at the desired design ranges, some areas have attempted to circumvent the standardized roadway classification system through the creation of special design districts that indicate the desired dimensions for specific roads. Portland, Oregon, known for its progressive pedestrian orientation, included pedestrian districts as part of its original 1977 Arterial Streets Policy. These districts include special design criteria specifically addressing pedestrian travel (City of Portland 1998).
SOURCE: Meyer and Dumbaugh 2004.
Another common approach to accommodating nonmotorized travel is traffic calming. Originating in Europe, these measures are designed to slow traffic speeds in residential neighborhoods and near schools and pedestrian ways through self-enforcing physical devices. Examples are vertical deflections (speed humps and bumps and raised intersections); horizontal deflections (serpentines, bends, and deviations in a road); road narrowing (via neckdowns and chokers); and medians, central islands, and traffic circles (Loukaitou-Sideris 2004). The Institute of Transportation Engineers has developed suggested design guidelines for traffic calming measures encompassing applications, design and installation issues, potential impacts, and typical costs (ITE 2004).
Finally, more creative use of the cul-de-sac could be considered. Cul-de-sac patterns providing greater connectivity could achieve more of the benefits of the street grid pattern while retaining the cul-de-sac’s higher levels of privacy, safety, and quiet and lower construction costs (Southworth and Ben-Joseph 2004). For example, designing residential communities that connected cul-de-sacs and loop streets through a system of pedestrian and bicycle paths would provide better access to parks, schools, and neighborhood shops (Southworth and Ben-Joseph 2004). Retrofitting existing suburban cul-de-sac developments could prove more difficult,25 but “safe pathways” could be designed by using a combination of existing public rights-of-way, sidewalks, and street space in some closer-in suburbs.26
Transportation Infrastructure Financing
Transportation infrastructure financing has been a major factor in the development of the current transportation system. In particu-
lar, funding restrictions on use, matching shares, procedural requirements, and design standards all have had important influences on project outcomes. In general, nonmotorized transportation modes and, to a lesser extent, transit have not fared well in traditional programs and policies (Meyer and Dumbaugh 2004).
Funding arrangements differ across transportation modes. Highways have a well-established financing system with a long history of federal assistance, primarily from gas tax revenues set aside in the Highway Trust Fund. Local street and county road improvements, however, are financed from local revenues. Transit funding is a federal and local, and increasingly a state, responsibility. Nonmotorized transportation modes are primarily locally financed.
Different funding arrangements provide different incentives and constraints. For example, for many years the emphasis of federal-aid transportation programs was on highways, and matching requirements for state and local funds mirrored this emphasis. Federal funds financed 90 percent of Interstate highway construction, but only 50 to 80 percent of the cost of constructing transit facilities. In addition, projects using federal funds had to incorporate federally required design criteria. For many projects, this meant building an improved facility—adding more capacity for vehicular travel, for example—rather than simply replacing the existing facility as it was.
State and local funding arrangements vary widely by jurisdiction. For example, state constitutions restrict the majority of state gas tax revenues to highway expenditures. These projects rarely include pedestrian-oriented improvements, such as sidewalks, which are considered the responsibility of local governments or individual landowners (Meyer and Dumbaugh 2004).
Local governments have assumed many responsibilities for transportation financing, including nonmotorized modes. For example, many larger communities finance transit operations with sales tax set-asides approved by voter referendum. Bicycle paths and pedestrian facilities (e.g., street overpasses) are largely a local responsibility or the responsibility of individual landowners (e.g., sidewalks). Local governments can finance such improvements through local taxes or impact fees on new developments but are often reluctant to
do so because of political backlash. These strategies shift costs directly to local residents (Meyer and Dumbaugh 2004).27
Since passage of the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA), the playing field between highways and transit has been leveled significantly. Certain highway funds can be “flexed” for transit and other nonhighway uses, and project matching shares for transit and highways are the same. In addition, several new programs were created that can help finance pedestrian and bicycle projects. One of the principal new funding sources for nonmotorized transportation is the Transportation Enhancements Program, which restricts 10 percent of Surface Transportation Program funds allocated to the states to such improvements as pedestrian and bicycle facilities and roadway beautification (Meyer and Dumbaugh 2004). The Congestion Mitigation and Air Quality Improvement (CMAQ) Program, also created by ISTEA, is aimed at improving metropolitan air quality. Projects such as bicycle, pedestrian, and transit improvements that encourage shifts from single-vehicle travel, thereby reducing vehicle emissions, are eligible for CMAQ funding (Meyer and Dumbaugh 2004). Another source of funding, particularly for enhancing bicycle and pedestrian safety, is the 402 program administered by the National Highway Traffic Safety Administration. Because sidewalks, intersection markings, and bicycle facilities can all be used to improve transportation safety, such projects are eligible for 402 funding (Meyer and Dumbaugh 2004). Finally, opportunities exist to incorporate pedestrian facilities and bicycle paths as part of other projects eligible for federal funding at minor additional cost.
AASHTO American Association of State Highway and Transportation Officials
BR&S Belden, Russonello & Stewart
BTS Bureau of Transportation Statistics
EPA U.S. Environmental Protection Agency
FHA Federal Housing Administration
DOT U.S. Department of Transportation
ITE Institute of Transportation Engineers
NRC National Research Council
TRB Transportation Research Board
AASHTO. 2001. A Policy on Geometric Design of Highways and Streets, 4th ed. Washington, D.C., Jan.
America Bikes. 2003. Polls Show Americans Want Better Biking. News release. Washington, D.C., May 5.
Appleyard, D. 1981. Livable Streets. University of California Press, Berkeley.
Barnes, P. M., and C. A. Schoenborn. 2003. Physical Activity Among Adults: United States, 2000. Advance Data from Vital and Health Statistics, No. 333, May 14.
Boarnet, M. G. 2004. The Built Environment and Physical Activity: Empirical Methods and Data Resources. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, July 18.
Boarnet, M. G., and R. Crane. 2001. Travel by Design: The Influence of Urban Form on Travel. Oxford University Press, Inc., New York.
Boarnet, M. G., K. Day, C. Anderson, T. McMillen, and M. Alfonzo. 2004. Urban Form and Physical Activity: Insights from a Quasi-Experiment. Presented at the Active Living Research Annual Conference, Del Mar, Calif., Jan. 30.
Booth, M. L., N. Owen, A. Bauman, O. Clavis, and E. Leslie. 2000. Social-Cognitive and Perceived Environment Influences Associated with Physical Activity in Older Australians. Preventive Medicine, Vol. 31, No. 1, pp. 15–22.
Brown, J. D. 1992. Benefit Segmentation of the Fitness Market. Review. Health Marketing Quarterly, Vol. 9, No. 3–4, pp. 19–28.
Brownson, R. C., and T. K. Boehmer. 2004. Patterns and Trends in Physical Activity, Occupation, Transportation, Land Use, and Sedentary Behaviors. Department of Community Health and Prevention Research Center, School of Public Health, St. Louis University. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, June 25.
Brownson, R. C., R. A. Housemann, D. R. Brown, J. Jackson-Thompson, A. C. King, B. R. Malone, and J. F. Sallis. 2000. Promoting Physical Activity in Rural Communities: Walking Trail Access, Use, and Effects. American Journal of Preventive Medicine, Vol. 18, pp. 235–241.
BR&S. 2003. Americans’ Attitudes Toward Walking and Creating Better Walking Communities. Washington, D.C., April.
BTS. 2002. Bicycle Use Among Adult U.S. Residents. OmniStats, Vol. 2, Issue 6, U.S. Department of Transportation, Dec.
BTS. 2003. Pedestrian Travel During 2002. OmniStats, Vol. 3, Issue 1, U.S. Department of Transportation, July.
Calthorpe, P. 1993. The Next American Metropolis: Ecology, Community, and the American Dream. Princeton Architectural Press, New York.
City of Portland. 1998. Portland Pedestrian Master Plan. Office of Transportation, Engineering, and Development.
Collette, M., G. Godin, R. Bradet, and N. J. Gionet. 1994. Active Living in Communities: Understanding the Intention to Take Up Physical Activity as an Everyday Way of Life. Canadian Journal of Public Health, Vol. 85, No. 6, pp. 418–421.
Dellinger, A. M., and C. E. Staunton. 2002. Barriers to Children Walking and Cycling to School—United States, 1999. Morbidity and Mortality Weekly Report, Vol. 51, No. 32, Aug. 16, pp. 701–704.
DOT. 2003. National Survey of Pedestrian and Bicyclist Attitudes and Behaviors—Highlights Report. National Highway Traffic Safety Administration and Bureau of Transportation Statistics.
Duany, A., E. Plater-Zyberk, and J. Speck. 2000. Suburban Nation: The Rise of Sprawl and the Decline of the American Dream. North Point Press, N.J.
Dunn, A. L., B. H. Marcus, J. B. Kampert, M. E. Garcia, H. W. Kohl III, and S. N. Blair. 1999. Comparison of Lifestyle and Structured Interventions to Increase Physical Activity and Cardiorespiratory Fitness: A Randomized Trial. Journal of the American Medical Association, Vol. 281, pp. 327–334.
EPA. 2003. Travel and Environmental Implications of School Siting. EPA-231-R-03-004. Washington, D.C.
Eppli, M. J., and C. C. Tu. 1999. Valuing the New Urbanism. Urban Land Institute, Washington, D.C.
FHA. 1935. Subdivision Development. Circular No. 5, Washington, D.C., Jan. 10.
Fischel, W. A. 1999. Does the American Way of Zoning Cause the Suburbs of Metropolitan Areas to Be Too Spread Out? In Governance and Opportunity in Metropolitan America (A. Altshuler, W. Morrill, H. Wolman, and F. Mitchell, eds.), National Academy Press, Washington, D.C., pp. 151–191.
Fridinger, F., S. Kirby, E. Howze, W. Holmes, B. Latham, and B. Leonard. 1996. Nutrition and Physical Activity Communications at the Centers for Disease Control and Prevention. Social Marketing Quarterly, Vol. 3, No. 2, pp. 14–15.
Hahn, A., and E. Craythorn. 1994. Inactivity and the Physical Environment in Two Regional Centers. Health Promotion Journal of Australia, Vol. 4, No. 2, pp. 43–45.
Handy, S. 2004. Critical Assessment of the Literature on the Relationships Among Transportation, Land Use, and Physical Activity. Department of Environmental Science and Policy, University of California, Davis. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, July.
Handy, S., M. G. Boarnet, R. Ewing, and R. E. Killingsworth. 2002. How the Built Environment Affects Physical Activity: Views from Urban Planning. American Journal of Preventive Medicine, Vol. 23, No. 2S, pp. 65–73.
ITE. 2004. Traffic Calming for Communities. www.ite.org/traffic. Accessed June 1, 2004.
Jackson, K. T. 1985. Crabgrass Frontier: The Suburbanization of the United States. Oxford University Press.
Jacobs, J. 1961. The Death and Life of Great American Cities. Random House, New York.
Jurgensmeyer, J., and T. Roberts. 1998. Land Use Planning and Control Law. West Group, St. Paul, Minn.
King, A. C., J. F. Sallis, A. L. Dunn, D. G. Simons-Morton, C. A. Albright, S. Cohen, W. J. Rejeski, B. H. Marcus, and M. C. Coday. 1998. Overview of the Activity Counseling Trial (ACT) Intervention for Promoting Physical Activity in Primary Health Care Settings. Medicine and Science in Sports and Exercise, Vol. 30, pp. 1086–1096.
King, A. C., C. Castro, S. Wilcox, A. A. Eyler, J. F. Sallis, and R. C. Brownson. 2000. Personal and Environmental Factors Associated with Physical Inactivity Among Different Racial Ethnic Groups of U.S. Middle-Aged and Older-Aged Women. Health Psychology, Vol. 19, No. 4, pp. 354–364.
Kirby, S. D., and M. Hollander. 2004. Consumer Preferences and Social Marketing Approaches to Physical Activity Behavior and Transportation and Land Use Choices. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, April 20.
Kohl, H. W. III., A. L. Dunn, B. H. Marcus, and S. N. Blair. 1998. A Randomized Trial of Physical Activity Interventions: Design and Baseline Data from Project Active. Medicine and Science in Sports and Exercise, Vol. 30, pp. 275–283.
Kotler, P., N. Roberto, and N. Lee. 2002. Selecting Target Markets. In Social Marketing: Improving the Quality of Life, Sage Publications, Thousand Oaks, Calif., pp. 112–115.
Levine, J., and A. Inam. 2004. The Market for Transportation–Land Use Integration: Do Developers Want Smarter Growth Than Regulations Allow? Transportation, Vol. 31, pp. 409–427.
Logan, G. T., L. Frank, T. M. Noell, C. Leersen, and P. Engelke. 2001. Overcoming Barriers to Smart Growth: Local Government Outreach Summary. www.smartraq.net/pdfs/public%20officials.pdf.
Loukaitou-Sideris, A. 2004. Transportation, Land Use, and Physical Activity: Safety and Security Considerations. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, June.
Meyer, M. D., and E. Dumbaugh. 2004. Institutional and Regulatory Factors Related to Nonmotorized Travel and Walkable Communities. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, July.
Minnesota Department of Transportation. 2000. Design Policy—Design Excellence Through Context Sensitive Design. Technical Memorandum 00-24-TS-03. St. Paul.
NRC. 1999. Governance and Opportunity in Metropolitan America (A. Altshuler, W. Morrill, H. Wolman, and F. Mitchell, eds.), National Academy Press, Washington, D.C.
Portland Metro. 2000. Light Rail Transit Station Zone. Chapter 33. 450, Title 33, Oregon.
Pucher, J., and J. L. Renne. 2003. Socioeconomics of Urban Travel: Evidence from the 2001 NHTS. Transportation Quarterly, Vol. 57, No. 3, Summer, pp. 49–77.
Reger, B., L. Cooper, S. Booth-Butterfield, H. Smith, A. Bauman, M. Wootan, S. Middlestadt, B. Marcus, and F. Greet. 2002. WHEELING WALKS: A Community Campaign Using Paid Media to Encourage Walking Among Sedentary Older Adults. Preventive Medicine, Vol. 353, Sept., pp. 285–292.
Rudd, R. E., J. Goldberg, and W. Dietz. 1999. A Five-Stage Model for Sustaining a Community Campaign. Journal of Health Communication, Vol. 4, No. 1, pp. 37–48.
Schwanen, T., and P. Mokhtarian. 2004. What Affects Commute Mode Choice: Neighborhood Physical Structure or Preferences Toward Neighborhoods? Journal of Transport Geography (forthcoming).
Seneviratne, P. N. 1985. Acceptable Walking Distances in Central Areas. Journal of Transportation Engineering, Vol. 3, No. 4, pp. 365–376.
Shoup, D. 1994. Cashing Out Employer-Paid Parking: A Precedent for Congestion Pricing? In Special Report 242: Curbing Gridlock: Peak-Period Fees to Relieve Traffic Congestion (Vol. 2), National Research Council, Washington, D.C., pp. 152–199.
Shoup, D. 1997. Evaluating the Effects of Cashing Out Employer-Paid Parking: Eight Case Studies. Transport Policy, Vol. 4, No. 4, Oct., pp. 201–216.
Sloan, K. S. 2001. Physical Activity and 50+: Preliminary Findings to Support Effective Communications and Outreach. Presented at the Conference on Communicating Physical Activity and Health Messages: Science into Practice, Centers for Disease Control and Prevention and Health Canada, Dec. 8–11.
Smith-Heimer, J., and R. Golem. 2001. Summary of Panel Discussions with California TOD Developers. California Transit Authority, Statewide Transit-Oriented Development Study, Technical Appendix, pp. 128–139.
Southworth, M., and E. Ben-Joseph. 2004. Reconsidering the Cul-de-Sac. Access, No. 24, Spring, pp. 28–33.
Staunton, C., D. Hubsmith, and W. Kallins. 2003. Promoting Safe Walking and Biking to School. American Journal of Public Health, Vol. 93, No. 9, pp. 1431–1434.
TRB. 1994. Special Report 242: Curbing Gridlock: Peak-Period Fees to Relieve Traffic Congestion. National Research Council, Washington, D.C.
TRB. 1995. Special Report 245: Expanding Metropolitan Highways: Implications for Air Quality and Energy Use. National Research Council, Washington, D.C.
TRB. 2003. Special Report 277: Measuring Personal Travel and Goods Movement: A Review of the Bureau of Transportation Statistics’ Surveys. National Research Council, Washington, D.C.
Wankel, L. M., and W. K. Mummery. 1993. Using National Survey Data Incorporating the Theory of Planned Behavior: Implications for Social Marketing Strategies in Physical Activity. Journal of Applied Sport Psychology, Vol. 5, No. 2, pp. 158–177.
Whyte, W. H. 1980. The Social Life of Small Urban Spaces. Conservation Foundation, Washington, D.C.
Designing Research to Study the Relationship Between the Built Environment and Physical Activity
A more rigorous understanding of the extent to which the built environment is a factor in individuals’ choices about physical activity is important in designing effective policies and interventions to address the decline in such activity. A review of the theory and data available to guide research on the links between the two reveals that conceptualization and measurement of the relevant environmental factors are a relatively new area of inquiry. A more complete theoretical framework is needed to provide the basis for formulating testable hypotheses, suggest the variables and relations for study, and help interpret study results.
Research designs emphasizing longitudinal approaches are particularly relevant for studying the potential causal relationship between a given aspect of the built environment and the desired behavior (i.e., more physical activity). With few exceptions, however, such studies are not evident in the research conducted to date. The issue of self-selection bias has only recently been incorporated into research designs. Both longitudinal and cross-sectional studies should use analytic approaches that help distinguish the extent to which an observed association between the built environment and physical activity reflects the characteristics of the built environment versus the attitudes and lifestyle preferences of those who choose to live in an environment with particular characteristics (e.g., walking and bicycle paths).
To date, most available research in this area has focused on cross-sectional analyses. The primary limitations of this research approach have been a poor understanding of the variables to include, which in turn reflects a deficiency of good theory, and the
lack of well-developed measures of the relevant attributes of the built environment at the appropriate geographic scale. The latter can be traced to inadequate data, a function of the relatively immature stage of the research.
Measures of physical activity have been the focus of considerable research and are better developed than measures of the built environment. On the other hand, large surveys that measure physical activity and health have been focused primarily on leisure-time physical activity and do not provide information on the location of that activity. Thus, the researcher cannot determine total levels of physical activity or identify where the activity has occurred so these data can be linked with those on the characteristics of the built environment. At a minimum, geocoding the data collected in several of the large surveys on physical activity and health could facilitate linking these rich data sets with information on the built environment. Greater use of technologies that provide automated and objective measures to help verify the accuracy and enhance the precision of self-reported survey and diary data is already possible.