Research on Women's Issues in Transportation, Volume 1: Conference Overview and Plenary Papers (2006)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

29 Community Design and Travel Behavior Exploring the Implications for Women Susan Handy, Department of Environmental Science and Policy, University of California, Davis In the face of growing levels of congestion and persis-tent air quality problems, planners increasingly seecommunity design as a way of reducing automobile dependence. Because of growing levels of obesity and the attendant health problems, public health officials have also turned to community design as a way of increasing physical activity. Proponents from both camps argue that higher population and employment densities, greater mixes of land uses, more gridlike street networks, and better transit service contribute to lower levels of driving and higher levels of walking, and they cite numerous studies to support their cases. But most studies focus on the population as a whole, and few studies so far con- sider the ways in which the effect of community design might differ for particular segments of the population given their particular travel needs. As evidence of the complexity of women’s travel accu- mulates, researchers have begun to explore what com- munity design means for women, both the possibility that community design adds to their travel burden and the possibility than it can help to ease that burden. Women face significant concerns related to family, health, and safety that complicate their daily lives; these concerns contribute to their need for travel and to the constraints they face in attempting to meet those needs. Communities designed so that women must drive long distances to work, to daycare, to shopping, or to medical appointments add to the time and cost of meeting their personal and household needs. In contrast, communities designed for shorter driving distances and for modes other than driving may offer women the option of reduc- ing the time and money they spend on travel. At this time, few questions have been answered and many questions remain, not only about the implications of community design for the travel of women but also about the relationship between community design and travel behavior more generally. As a step toward build- ing a research agenda on the implications for women of the relationship between community design and travel behavior, the available literature is reviewed here, origi- nal data analysis is presented, and outstanding issues are discussed for the following questions: • What is community design? • How does community design affect travel behavior? • How might these effects differ for women? • Where do we go from here? WHAT IS COMMUNITY DESIGN? Researchers do not always agree on a definition of com- munity design or even on the use of this term rather than the term “built environment” or the term “physical envi- ronment” or some other term. Community design and built environment are used interchangeably here and defined as consisting of three elements: land use, the transportation system, and design (1). Land use refers to the spatial distribution of activities throughout the com- munity, in other words, what kinds of activities are located where. The transportation system provides the physical connections between activities and determines the quality of those connections in terms of travel times, safety, comfort, and other characteristics. Design refers

to aesthetic qualities of the built environment and over- lies both land use patterns and the transportation sys- tem, particularly in terms of the design of buildings and the design of streetscapes, respectively. Other terms also need definition. Sometimes used interchangeably with the built environment, the physical environment can be defined as encompassing not just the built environment but also the natural landscape and the human use of public spaces, elements that have the potential to influence choices about travel behavior as well. Another important concept to consider is that of access. Access acts as a mediating or intervening variable between community design and travel behavior (Figure 1). In other words, community design determines levels of access to potential destinations, and the level of access then influences travel choices. Some researchers thus choose to measure access rather than community design itself in order to explain travel behavior. All of these con- cepts—community design, built environment, physical environment, access—can be measured at different scales, from the block to the neighborhood to the region. The idea that community design can change travel behavior is not a new one, though it did get a boost in the 1990s with the growing popularity of the New Urbanism movement. As articulated by the Congress for the New Urbanism, this movement advocates for com- munities designed “for the pedestrian and transit as well as the car” where “many activities of daily living … occur within walking distance, allowing independence to those who do not drive, especially the elderly and the young” and where “interconnected networks of streets [are] designed to encourage walking, reduce the number and length of automobile trips, and conserve energy” (2). The concept of transit-oriented development (TOD) puts forward a similar idea: “Moderate to higher density development, located within an easy walk of a major transit stop, generally with a mix of residential, employ- ment and shopping opportunities designed for pedestri- ans without excluding the auto” (3). More recently, the Active Living by Design movement has promoted “envi- ronments that offer choices for integrating physical activity into daily life” (4). Each of these movements assumes that community design can reduce driving and increase walking. Other movements in the planning field might also work to alter community design in such a way as to decrease driving and increase walking (Table 1). A grow- ing number of U.S. communities have adopted street connectivity ordinances, which require a more gridlike street network in new residential subdivisions (5); other communities are investing in pedestrian-bicycle bridges and tunnels to connect areas severed by freeways or divided by rivers (6). A loosely related set of programs that might be called Main Street programs aims to revi- talize neighborhood commercial areas [e.g., Metro in Portland, Oregon (7)], protect local shopping areas from big-box stores [e.g., the New Rules Project (8)], and rebuild suburban strip malls as community centers [e.g., the Local Government Commission projects (9)]. Traffic calming programs, popular throughout the United States and elsewhere in the world, redesign streets to better bal- ance the needs of pedestrians and bicyclists with the needs of vehicles (10). Safe-routes-to-school programs, such as the one funded by the California Department of Transportation (11), represent a specific form of traffic calming, one designed to increase the share of children who walk or bike to school. Finally, trails programs such as those promoted by the Rails-to-Trails Conservancy (12) have grown popular in communities of all sizes; although these trails are often best suited for recreation, they may also serve as useful routes for transportation- oriented walking or biking. 30 RESEARCH ON WOMEN’S ISSUES IN TRANSPORTATION Community Design Access Travel Behavior Attitudes & Preferences FIGURE 1 Relationships between community design, access, travel behavior, and attitudes and preferences.

HOW DOES COMMUNITY DESIGN AFFECT TRAVEL BEHAVIOR? What effect might these changes in community design have on travel behavior? In setting out to answer this question, researchers have measured community design in various ways, most commonly as density, land use mix, distance to the nearest destination, other measures of accessibility, or neighborhood type (1). Travel behav- ior has also been measured in various ways, most com- monly by trip frequency, trip distance, mode choice, and total vehicle miles traveled (VMT). Most (but not all) studies control in some way for sociodemographic factors that might influence travel behavior, including income, automobile ownership, household size, employment status, presence of children, and sex. Many studies compare travel behavior in neighbor- hoods of different types by using analysis of variance techniques. In other studies, methods for testing for an association between community design and travel behavior while controlling for sociodemographic char- acteristics range from simple regression models to more sophisticated discrete choice models. Differences in measures and methodologies across these studies make direct comparisons of their results difficult; neverthe- less, certain patterns emerge. After reviewing more than 50 empirical studies, Ewing and Cervero (13) concluded that trip lengths are primarily a function of the built environment, trip frequencies are primarily a function of sociodemo- graphic characteristics, and mode choice is a function of both. Their meta-analysis of available data from these studies produced estimates of elasticities of –0.05 for the relationship between density and VMT. In other words, a 100% increase in density was associ- ated with a 5% decrease in VMT—a statistically sig- nificant though rather small relationship. In a review of studies of the link between the built environment and walking as a mode of transportation, Saelens et al. (14) concluded that “transportation and planning research supports the proposition that the physical environment is associated with physical activity in the form of walking/cycling for transport.” In Handy’s review of this literature, strong evidence was found of an association between accessibility and frequency of walking (1). But these studies generally assume a simple causal relationship between community design and travel behavior. A few studies have used a more complicated framework that accounts for the role of attitudes and preferences (Figure 1) and have found that attitudes and preferences play a more significant role in explain- ing travel behavior than community design (15, 16). These studies raise the possibility that the observed association between community design and travel behavior is spurious, that an association between atti- tudes and preferences and both community design and travel behavior creates the appearance of a causal rela- tionship between community design and travel behav- ior that does not actually exist (17). Researchers are now taking on the “self-selection” question and testing the degree to which residents who prefer to drive less or to walk more consciously choose neighborhoods that offer such opportunities. In studies of the link between community design and travel behavior, sex has been just one more variable to control for rather than the focus of analysis, despite the fact that it has been a significant factor in many studies. In the travel behavior literature, some studies have ana- lyzed travel at the level of the household rather than the individual (15, 18) or have used aggregate data at the level of the census tract or neighborhood (19, 20), in which case the effects of sex have not been considered. Analyses at the level of the individual show that after the effect of the built environment is accounted for, women make more trips by car (21, 22), make fewer walking trips (23, 24), are less likely to choose an alternative to driving for their trips (25, 26), but still drive less than men overall (27). These studies thus show a significant effect of sex on travel behavior but do not consider inter- action effects, that is, the differential effect of commu- nity design on travel behavior for women. 31COMMUNITY DESIGN AND TRAVEL BEHAVIOR TABLE 1 Planning Movements That Improve Alternatives to Driving Movement Variations Potential Effects Network connectivity Street connectivity ordinances Shorter distances to destinations Bike/ped bridges over freeways More direct routes to destinations Bike/ped bridges over rivers Main Street programs National, state, and local Main Street programs Stores and services within walking distances Anti-Big-Box ordinances Grayfield redevelopment of strip malls Traffic calming programs City traffic calming programs Increased safety and comfort for pedestrians Safe-routes-to-school programs Trails programs Transportation enhancements projects Separate facilities for bikes/peds Rails-to-Trails projects

HOW MIGHT THESE EFFECTS DIFFER FOR WOMEN? Community design might affect travel behavior differ- ently for women than it does for men for the basic rea- son that women’s daily lives are significantly different from men’s. In particular, women face different and often more pressing concerns related to family duties, health issues, and safety threats that relate to travel behavior. Community design often contributes to these concerns but also has the potential to help alleviate them by improving alternatives to driving. Although research on these issues is limited, data from a recent study by Handy et al. at the University of Cali- fornia, Davis (UC Davis), is suggestive of what further research efforts might find (28). These data, selections of which are presented in the following discussion, come from a survey of households in eight neighborhoods in Northern California, four traditional ones and four sub- urban ones. Community design in these two types of neighborhoods differs significantly, with traditional neighborhoods offering greater accessibility to basic ser- vices, measured both as the number of services within specified distances from home and as the distance to the nearest service establishment of each type. The survey was administered using a mail-out, mail-back approach that achieved a 25% response rate for a total sample of nearly 1,700 respondents. Categories of variables in the survey included travel behavior, perceived characteristics of the neighborhood, travel attitudes, and sociodemo- graphic characteristics. According to the survey results for perceived neigh- borhood characteristics, traditional neighborhoods offer greater accessibility to potential destinations, more social activity in the neighborhood, and a more attractive envi- ronment, thus potentially offering more opportunities for walking. Suburban neighborhoods, on the other hand, offer a greater sense of safety, another important factor influencing the potential for walking. In the fol- lowing sections, simple comparisons are presented between men and women and between women with chil- dren and those without children for traditional neigh- borhoods and suburban neighborhoods. Though the analyses presented here do not establish a causal rela- tionship, differences in travel behavior by type of neigh- borhood are suggestive of the impact of community design on travel behavior. These results point to interest- ing and important research questions. Family Concerns Despite working outside the home at higher rates than ever, women still bear more of the responsibility for household duties. The 2003 American Time-Use Survey showed that employed men worked about an hour more than employed women per day but that employed adult women spent about an hour more per day than employed adult men doing household activities and car- ing for household members (29). On an average day, 84% of women spent some time doing household activ- ities (such as housework, cooking, lawn care, or house- hold management), 55% reported doing housework (such as cleaning or laundry), and 66% reported doing food preparation or cleanup versus 63%, 20%, and 35% of men, respectively. Not only do these responsibilities restrict free time for women, but they are associated with additional travel needs. A report by the Surface Trans- portation Policy Project (30) found that two-thirds of all trips to chauffeur people around (such as driving chil- dren to soccer practice or an older parent to the doctor) are made by women. Family responsibilities are espe- cially burdensome for women in the “sandwich genera- tion,” with responsibilities for both dependent children and aging parents. Community design offers the possibility of easing these burdens at least a bit. If communities are designed so that necessary destinations such as schools, stores, and medical offices are closer to home, women will not need to travel so far to take care of their household responsibilities. If communities are also designed so that walking is a safe and viable option and if adequate tran- sit service is provided, dependent children and aging par- ents might be able to get where they need to go by themselves, reducing the burden on women to chauffer them. The reduced need for driving would save women time and money and, equally important, increase their flexibility. Good community design, in other words, can make it easier for women to conduct their everyday lives. Data from the UC Davis study suggest that suburban environments put a greater driving burden on women than traditional neighborhoods do. Consistent with pre- vious studies, the survey results show that women drive less than men—141 mi in a typical week versus 184 mi for men (Table 2). However, the differences between tra- ditional and suburban neighborhoods are significant, with women who live in traditional neighborhoods driv- ing 20 mi less than women who live in suburban neigh- borhoods, a difference of 15%. The data do not show whether this difference is a matter of choice or necessity, 32 RESEARCH ON WOMEN’S ISSUES IN TRANSPORTATION TABLE 2 Vehicle Miles Driven per Week All Respondents Traditional Suburban p-value All respondents 161 148 177 0.00 Women 141 132 152 0.06 Men 184 167 202 0.01 p-value 0.00 0.00 0.00 Women w/kids 161 132 180 0.04 Women w/o kids 135 132 139 0.51 p-value 0.03 0.82 0.01

though the latter seems more likely given the time con- straints that women face. For women with children, liv- ing in the suburbs exerts an even greater driving penalty—44 mi more per week than in traditional neigh- borhoods, a difference of 32%. These differences hold for nonwork driving as well (Table 3): women with chil- dren who live in the suburbs drive more for nonwork purposes than do women in the other three categories. These results point to important questions for researchers to address: What factors contribute to higher levels of driving by women in suburban environments? To what degree do these differences reflect a causal effect and to what degree do they reflect self-selection? More specific questions are also of interest: Does the increased burden on mothers reflect a lack of independence of their children in suburban environments? Research by McDonald (31) shows no evidence that the burden on mothers declines as population densities increase, sug- gesting that other factors may be at play. In contrast, research by Weston (32) suggests that adolescent girls are less likely to travel independently than are adolescent boys, putting a greater burden on their parents, particu- larly their mothers. What about women with constrained access to a car? Is their situation mitigated or exacerbated by community design? A study in Germany by Vance et al. (33) found little impact of community design as either a mitigating or an exacerbating factor. Is the situation dif- ferent in the United States, where getting around in sub- urban areas is largely dependent on driving? Health Concerns Levels of obesity are increasing in the United States, espe- cially for women and for older women. In 2000, 62% of U.S. women were overweight, a rate that jumped to 68% for women older than 60 (34). Being overweight con- tributes to numerous health effects, including cardiovas- cular disease, arthritis, breast cancer, gallbladder disease, and infertility, as well as stigma and discrimination. A decline in physical activity is contributing to the prob- lem. According to the Centers for Disease Control and Prevention (CDC) (35), more than 60% of women in the United States do not engage in the recommended amount of physical activity and more than 25% are not active at all. Even modest increases in moderate physical activity can help. Community design can play an important role in efforts to increase physical activity for women. If com- munity design brings destinations within walking dis- tance, women are more likely to choose walking over driving. The CDC recommends “environmental induce- ments to physical activity” such as trails and sidewalks as well as neighborhood watch programs to increase safety (35). Such facilities help to encourage recreational walking as well as walking to destinations. The aesthetic qualities of the environment and amenities such as shade, benches, and drinking fountains also make walking a more attractive choice. Indeed, studies of physical activ- ity show that enjoyable scenery is associated with more activity for women (36). The UC Davis study reveals significant differences in walking between traditional and suburban neighbor- hoods for women. In traditional neighborhoods, women walk to the store more than twice as often as in subur- ban neighborhoods; the differences hold for men as well (Table 4). The effect of children is significant: in tradi- tional neighborhoods, women with children walk less frequently than women without children. But they still walk more frequently than women living in suburban neighborhoods, with or without children. For strolling around the neighborhood, the differences are similar: women (and men) living in traditional neighborhoods stroll more frequently than women (and men) living in suburban neighborhoods (Table 5). However, for strolling the frequency is just 20% more in traditional neighborhoods rather than 135% more for walking to the store. For strolling, the effect of children is not sig- nificant: women living with children stroll just as fre- 33COMMUNITY DESIGN AND TRAVEL BEHAVIOR TABLE 3 Nonwork Vehicle Miles Driven per Week All Respondents Traditional Suburban p-value All respondents 91 84 98 0.02 Women 79 76 83 0.35 Men 103 93 113 0.05 p-value 0.00 0.03 0.00 Women w/kids 88 78 96 0.30 Women w/o kids 77 76 78 0.80 p-value 0.22 0.88 0.18 TABLE 4 Walks to Store per Month All Respondents Traditional Suburban p-value All respondents 3.5 4.9 1.8 0.00 Women 3.5 4.7 2.0 0.00 Men 3.5 5.1 1.8 0.00 p-value 0.9 0.3 0.4 Women w/kids 2.9 3.6 2.3 0.03 Women w/o kids 3.7 4.9 1.9 0.00 p-value 0.07 0.08 0.37 TABLE 5 Strolls per Month All Respondents Traditional Suburban p-value All respondents 9.0 10.1 7.7 0.00 Women 9.4 10.2 8.5 0.03 Men 8.7 10.0 7.2 0.00 p-value 0.2 0.8 0.1 Women w/kids 9.1 10.0 8.4 0.23 Women w/o kids 9.5 10.3 8.5 0.08 p-value 0.64 0.87 0.96

quently as women not living with children, whether they live in traditional or suburban neighborhoods. Even in traditional neighborhoods, however, the average fre- quency is just 10 times in 30 days, or about once every 3 days, less than recommended levels for exercise. These results suggest that community design affects walking as a mode of transportation differently than it does walking for exercise. These results raise many other questions for researchers. To what degree does community design explain levels of walking for women? The answer is not straightforward. An analysis by Clifton and Dill (37) of several different data sources finds that factors such as household responsibilities and resources, perceptions and concerns, trip purposes and comprehensive travel needs confound the relationship between community design and walking for women. Are higher levels of walking in traditional neighborhoods attributable to community design or to the preferences and attitudes of the women who live there? A study in three Maryland communities by Clifton and Livi (38) found significant differences in attitudes and perceptions about walking between men and women that helped to explain differ- ences in walking. How does walking in the neighbor- hood fit into overall levels of physical activity for women? In an analysis of data from the National House- hold Transportation Survey, Helling (39) found that women drive to places to exercise (such as a gym or health club) less frequently than men; this result suggests the importance of walking in the neighborhood as a form of exercise for women but could also reflect more exer- cising within the home or less exercise overall. Safety Concerns Almost every survey of fear of crime shows that women are more concerned about their personal safety and feel less safe than men (40). This difference is especially prevalent in public places, including the transportation system. Because cars represent a private and protected environment, women perceive cars to be the safest mode and transit settings, including stops and stations as well as the vehicles themselves, to be risky (40). This percep- tion may eliminate alternatives to driving from consider- ation for many women and makes nondriving modes particularly burdensome for women who do not have the option to drive. Safety concerns thus compound the challenges associated with family and health concerns. For example, women who perceive neighborhoods as unsafe are less physically active (41). Community design can help in many different ways (40). Better lighting can mean greater comfort for walk- ing and transit, for example, as can design that elimi- nates dark spaces where potential attackers may hide. Wide sidewalks with a wide grass strip can greatly improve the walking experience. Community design that encourages more people to be out and about helps to increase perceived safety; architecture that promotes “eyes on the street” also helps. The presence of certain kinds of land uses—liquor stores or bars, for example— may decrease perceived safety for women. Creating safe places for women or other groups to “hang out” can also help. Safety for pedestrians can also be affected by the volume and speed of vehicle traffic; traffic calming programs that slow speeds and discourage excess traffic can thus help to increase real and perceived safety. If women feel safer walking and riding transit, they have a greater opportunity to drive less. In the UC Davis study, survey respondents were asked how true 34 characteristics were for their neighbor- hoods. Perceived characteristics related to safety that might influence walking differed between men and women and between women in traditional and those in suburban neighborhoods in interesting ways (Table 6). On average, women rated their neighborhoods higher for low crime, safety for children, neighbor interaction, and neighbors out and about; scores for safe walking, street lights, and quiet neighborhood did not differ between men and women. These results suggest that women perceive a better environment for walking than men do. Women in traditional neighborhoods gave higher scores for neighbor interaction and neighbors out and about but lower scores for street lighting and quiet neigh- borhood; these results suggest that traditional neighbor- hoods are better for walking in some ways but not in others. Women with children gave lower scores on low crime and safety for children than did women without children, possibly owing to higher standards for these characteristics rather than actual differences in their environments. Women with children in traditional neigh- borhoods gave lower scores on safety for children than women with children in suburban neighborhoods; this finding is consistent with the perception that suburbs are safer places for children. Survey respondents were also asked whether they agreed or disagreed with a series of statements related to transportation. Principal components factor analysis was used to reduce more than 30 statements to six fac- tors, including a factor that reflects an attitude that driv- ing is safe relative to other modes. On this factor, women scored somewhat lower on average than men. The score suggests that they feel less strongly that driving is safer than other modes (Table 7). In traditional neighbor- hoods, women agreed somewhat less than men that driv- ing is safest, but both men and women in traditional neighborhoods agreed much less than residents of subur- ban neighborhoods that driving is safest. The effect of children was not significant: women in suburban neigh- 34 RESEARCH ON WOMEN’S ISSUES IN TRANSPORTATION

borhoods agree more strongly that driving is safest than women in traditional neighborhoods, whether or not they have children. These results suggest that attitudes about the safety of driving relative to other modes are determined more by neighborhood than by sex or pres- ence of children. It is possible that walking, biking, and transit are in fact less safe in suburban areas (making driving relatively safer), but it is also possible that resi- dents in the two neighborhood types have different per- ceptions of these modes despite similar levels of safety. Again, these results raise more questions than they answer. To what degree does perceived safety influence travel behavior for women? Do safety concerns differ depending on culture or ethnicity? How do safety con- cerns affect the quality of life for women, particularly those without the option to drive? To what degree do perceptions of safety match the reality of safety? What aspects of community design contribute to feelings of safety? Much of the research in this area focuses on safety concerns related to transit and explores ways of improving safety for women and others. Loukaitou- Sideris and others have studied how community design influences crime and safety in the vicinity of transit stops (42, 43) and Loukaitou-Sideris more recently offers approaches to addressing the fear of victimization that women feel in public places (40). But much of this research focuses on strategies for improving safety other than community design. Carter (44) discusses gender- sensitive solutions to improving safety for women who use transit, including technological, staffing, and cultural approaches. Anderson Bomar (45) looks at the use of technology to address system security and evaluates the effectiveness of these approaches for addressing the safety concerns of women. WHERE DO WE GO FROM HERE? On the topic of community design and travel behavior, answers are scarce and questions abundant (17). On the topic of how the relationship between community design and travel behavior differs for women, answers are rarer. Yet the questions for women are in many ways more interesting and potentially more important. Women face numerous concerns related to family, health, and safety that create significant transportation needs and con- tribute to critical transportation constraints (Table 8), and community design may prove important in efforts to ease these burdens. But many questions remain about the role of community design in shaping travel behavior for women in different situations and about the potential effectiveness of community design approaches to addressing their concerns. To answer the questions outlined here, researchers need to move beyond cross-sectional designs that simply compare travel behavior in one type of community with that in another (17). Longitudinal studies that track changes in travel behavior associated with changes in community design that occur as women move from one place to another provide a more solid basis for under- standing the causal effect of community design. Interven- tion studies that look at changes in behavior after a 35COMMUNITY DESIGN AND TRAVEL BEHAVIOR TABLE 6 Perceived Neighborhood Characteristics Low Safe Safe for Street Low Quiet Neighbor Neighbors Crime Walking Kids Lights Traffic Neighborhood Interaction Out All respondents 3.64 3.42 3.45 3.00 2.60 3.01 2.82 2.97 Women 3.78 3.44 3.53 3.03 2.58 3.00 2.90 3.02 Men 3.48 3.40 3.36 2.96 2.61 3.02 2.73 2.91 p-value 0.00 0.41 0.04 0.17 0.51 0.57 0.03 0.02 Women in traditional 3.75 3.43 3.50 2.94 2.57 2.91 3.01 3.17 Women in suburban 3.81 3.46 3.55 3.13 2.60 3.10 2.76 2.83 p-value 0.68 0.69 0.69 0.00 0.65 0.00 0.03 0.00 Women w/kids 3.51 3.46 3.24 3.03 2.60 3.07 2.84 3.06 Women w/o kids 3.86 3.44 3.62 3.02 2.58 2.97 2.92 3.01 p-value 0.03 0.79 0.01 0.88 0.79 0.18 0.58 .059 Women w/kids in tradional 3.46 3.44 3.10 2.92 2.48 3.00 2.95 3.21 Women w/kids in suburban 3.55 3.47 3.34 3.12 2.68 3.12 2.76 2.95 p-value 0.67 0.76 0.08 0.19 0.16 0.34 0.33 0.11 NOTE 4-point scale from 1 (not at all true) to 4 (entirely true). TABLE 7 Perception That Driving Equals Safety All Respondents Traditional Suburban p-value All respondents –0.01 –0.27 0.29 0.00 Women –0.06 –0.33 0.29 0.00 Men 0.03 –0.20 0.30 0.00 p-value 0.07 0.06 0.93 Women w/kids –0.02 –0.34 0.22 0.00 Women w/o kids –0.07 –0.33 0.33 0.00 p-value 0.60 0.93 0.26 NOTE: Factor score based on 4 attitudinal statements; factor score has mean 0 and variance of 1.

specific change in community design, such as the imple- mentation of a traffic calming program, are also a more effective way of establishing causality. In either type of study, attitudes and perceptions must be accounted for to establish that the causal relationship between community design and travel behavior is real. A better understanding of causal relationships is important in building the evi- dence base for community design policies in general and those targeted to the needs of women in particular. To answer the questions outlined here, researchers also need to move beyond quantitative studies. Qualita- tive approaches can be used to explore not just travel behavior but also the travel needs, constraints, attitudes, and preferences that shape behavior and can produce new insights into the role of community design as facili- tator, constraint, or both. Qualitative studies focused on women can provide deeper insights into these questions than quantitative studies alone. The available research on the relationship between community design and travel behavior provides a starting point for more quali- tative exploration, and qualitative exploration can then provide the basis for further quantitative work focused on the ways in which community design helps to address (or to make worse) family, health, and safety concerns. With all the questions that remain, it is not possible at this point to say that community design is either the problem or the solution. But the available research does reveal that community design increases the opportunities for alternatives to driving, for increased physical activity, and for improved personal safety. It can be said that these increased opportunities, regardless of their impact on behavior, are themselves a good thing for women. REFERENCES 1. Handy, S. L. Critical Assessment of the Literature on the Relationships Among Transportation, Land Use, and Physical Activity. Resource paper for Special Report 282: Does the Built Environment Influence Physical Activity? Examining the Evidence. TRB; Committee on Physical Activity, Health, Transportation, and Land Use, Institute of Medicine, National Academies, Wash- ington, D.C., 2005. http://trb.org/downloads/sr282 papers/sr282Handy.pdf. 2. Congress for the New Urbanism. Charter of the New Urbanism. 2005. http://www.cnu.org/aboutcnu/index .cfm?formAction=charter. 3. Statewide Transit-Oriented Development Study: Fac- tors for Success in California. Business, Transportation, and Housing Agency, California Department of Trans- portation, Sacramento, 2002. http://transitoriented development.dot.ca.gov/PDFs/TOD%20Study%20Exe cutive%20Summary.pdf. 4. Active Living by Design and Public Health. Active Liv- ing by Design, Chapel Hill, N.C., 2004. http:// www.activelivingbydesign.org/fileadmin/template/docu ments/primer/Primer_Low_Res.pdf. 5. Handy, S. L., R. G. Paterson, and K. Butler. Planning for Street Connectivity: Getting from Here to There. Amer- ican Planning Association, Chicago, Ill., 2003. 6. Handy, S. L. Amenity and Severance. In Handbook of Transport and the Environment (D. Hensher and K. Button, eds.), Elsevier, Ltd., Oxford, 2003. 7. Main Street Handbook: A User’s Guide to Main Streets. Metro, Portland, Oreg., 1996. 8. The Hometown Advantage: Reviving Local Business. New Rules Project, 2005. http://www.newrules.org/ retail/index.php. 9. Model Projects, Local Government Commission. Local Government Commission, 2004. http://www.lgc.org/ freepub/land_use/models/anaheim.html and http://www .lgc.org/freepub/land_use/models/santana_row.html. 10. Traffic Calming: State of the Practice. Institute of Trans- portation Engineers and Federal Highway Administra- tion, 1999. http://www.ite.org/traffic/tcstate.htm#tcsop. 11. Safe Routes to School. California Department of Trans- portation, Sacramento, 2005. http://www.dot.ca.gov/ hq/LocalPrograms/saferoute2.htm. 12. Rails to Trails Conservancy Home Page. 2005. http:// www.railtrails.org. 13. Ewing, R., and R. Cervero. Travel and the Built Envi- ronment: A Synthesis. In Transportation Research Record: Journal of the Transportation Research Board, 36 RESEARCH ON WOMEN’S ISSUES IN TRANSPORTATION TABLE 8 Summary of Research Questions Family Concerns Health Concerns Safety Concerns Concerns for women High levels of household work Low levels of physical activity Low levels of real and perceived safety in public places Implications for travel Dependence on driving Benefits from more walking Transit and walking not as safe as driving Chauffering duties Potential role of Can reduce the need for driving Can increase opportunities Can increase safety for transit community design for walking and walking Cross-cutting questions How important is community design relative to other factors in influencing travel choices for women? How does the role of community design differ by culture or ethnicity for women? How does community design help or hinder women without access to cars? What factors influence which communities women choose or are forced to live in? What kinds of community design policies can effectively address these concerns?

No. 1780, TRB, National Research Council, Washing- ton, D.C., 2001, pp. 87–113. 14. Saelens, B. E., J. F. Sallis, and L. D. Frank. Environmen- tal Correlates of Walking and Cycling: Findings from the Transportation, Urban Design, and Planning Litera- tures. Annals of Behavioral Medicine, Vol. 25, No. 2, 2003, pp. 80–91. 15. Kitamura, R., P. L. Mokhtarian, and L. Laidet. A Micro- Analysis of Land Use and Travel in Five Neighborhoods in the San Francisco Bay Area. Transportation, Vol. 24, 1997, pp. 125–158. 16. Bagley, M. N., and P. L. Mokhtarian. The Impact of Residential Neighborhood Type on Travel Behavior: A Structural Equations Modeling Approach. Annals of Regional Science, Vol. 36, No. 2, 2002, pp. 279–297. 17. Special Report 282: Does the Built Environment Influ- ence Physical Activity? Examining the Evidence. TRB; Committee on Physical Activity, Health, Transporta- tion, and Land Use, Institute of Medicine, National Academies, Washington, D.C., 2005. 18. Krizek, K. Residential Relocation and Changes in Urban Travel: Does Neighborhood-Scale Urban Form Matter? Journal of the American Planning Association, Vol. 69, No. 3, 2003, pp. 265–281. 19. Friedman, B., S. P. Gordon, and J. B. Peers. Effect of Neotraditional Neighborhood Design on Travel Char- acteristics. In Transportation Research Record 1466, TRB, National Research Council, Washington, D.C., 1994, pp. 63–70. 20. Cervero, R.. Mixed Land-Uses and Commuting: Evi- dence from the American Housing Survey. Transporta- tion Research, Vol. 30A, No. 5, 1996, pp. 361–377. 21. Crane, R., and R. Crepeau. Does Neighborhood Design Influence Travel? A Behavioral Analysis of Travel Diary and GIS Data. Transportation Research, Vol. 3D, No. 4, 1998, pp. 225–238. 22. Boarnet, M., and R. Crane. The Influence of Land Use on Travel Behavior: Specification and Estimation Strate- gies. Transportation Research, Vol. 35A, 2001, pp. 823–845. 23. Greenwald, M. J., and M. G. Boarnet. Built Environ- ment as Determinant of Walking Behavior: Analyzing Nonwork Pedestrian Travel in Portland, Oregon. In Transportation Research Record: Journal of the Trans- portation Research Board, No. 1780, TRB, National Research Council, Washington, D.C., 2001, pp. 33–42. 24. Handy, S. L., and K. J. Clifton. Local Shopping as a Strategy for Reducing Automobile Travel. Transporta- tion, Vol. 28, No. 4, 2001, pp. 317–346. 25. Cervero, R., and C. Radisch. Travel Choices in Pedes- trian Versus Automobile Oriented Neighborhoods. Transport Policy, Vol. 3, 1996, pp. 127–141. 26. Kockelman, K. M. Travel Behavior as Function of Accessibility, Land Use Mixing, and Land Use Balance: Evidence from San Francisco Bay Area. In Transporta- tion Research Record 1607, TRB, National Research Council, Washington, D.C., 1997, pp. 116–125. 27. Handy, S. L., X. Cao, and P. Mokhtarian. Correlation or Causality Between the Built Environment and Travel Behavior? Evidence from Northern California. Trans- portation Research, Vol. 10, 2005, pp. 427–444. 28. Handy, S. L., P. L. Mokhtarian, T. Buehler, and G. Col- lantes. Residential Location Choice and Travel Behav- ior: Implications for Air Quality. California Department of Transportation, Sacramento, 2004. 29. American Time-Use Survey Summary. Bureau of Labor Statistics, U.S. Department of Labor, 2004. http://www.bls.gov/news.release/atus.nr0.htm. 30. High Mileage Moms: The Report. Surface Transporta- tion Policy Project, Washington, D.C., 2002. http:// www.transact.org/report.asp?id-184. 31. McDonald, N. C. Does Residential Density Affect the Travel “Gender Gap”? In Conference Proceedings 35: Research on Women’s Issues in Transportation, Trans- portation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, pp. 68–75. 32. Weston, L. M. Gender Differences in the Travel Behav- ior of 13-, 14-, and 15-Year-Olds and Role of Built Envi- ronment (abstract only). In Conference Proceedings 35: Research on Women’s Issues in Transportation, Trans- portation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, p. 76. 33. Vance, C., S. Buchheim, and E. Brockfeld. Gender as a Determinant of Car Use: Evidence from Germany. In Conference Proceedings 35: Research on Women’s Issues in Transportation, Transportation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, pp. 59–67. 34. AOA Fact Sheets: Women and Obesity. American Obe- sity Association, 2004. http://www.obesity.org/subs/ fastfacts/obesity_women.shtml. 35. Physical Activity and Health: A Report of the Surgeon General. Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, Atlanta, Ga., 1996. 36. King, A. C., C. Castro, S. Wilcox, A. A. Eyler, J. F. Sal- lis, and R. C. Brownson. Personal and Environmental Factors Associated with Physical Inactivity Among Dif- ferent Racial-Ethnic Groups of U.S. Middle-Aged and Older-Aged Women. Health Psychology, Vol. 19, No. 4, 2000, pp. 354–364. 37. Clifton, K. J., and J. Dill. Women’s Travel Behavior and Land Use: Will New Styles of Neighborhoods Lead to More Women Walking? In Conference Proceedings 35: Research on Women’s Issues in Transportation, Trans- portation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, pp. 89–99. 38. Clifton, K. J., and A. D. Livi. Gender Differences in Walking Behavior, Attitudes About Walking, and Per- ceptions of the Environment in Three Maryland Com- 37COMMUNITY DESIGN AND TRAVEL BEHAVIOR

munities. In Conference Proceedings 35: Research on Women’s Issues in Transportation, Transportation Research Board of the National Academies, Washing- ton, D.C., 2005, Vol. 2, pp. 79–88. 39. Helling, A. Connection Between Travel and Physical Activity: Differences by Age and Gender (abstract only). In Conference Proceedings 35: Research on Women’s Issues in Transportation, Transportation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, p. 77. 40. Loukaitou-Sideris, A. Is It Safe to Walk Here? Design and Policy Responses to Women’s Fear of Victimization in Public Places. In Conference Proceedings 35: Research on Women’s Issues in Transportation, Trans- portation Research Board of the National Academies, Washington, D.C., 2005, Vol. 2, pp. 102–112. 41. Centers for Disease Control and Prevention. Neighbor- hood Safety and the Prevalence of Physical Inactivity: Selected States, 1996. Morbidity and Mortality Weekly Report, Vol. 48, No. 7, 1999, pp. 143–146. 42. Loukaitou-Sideris, A. Hot Spots of Bus Stop Crime: The Importance of Environmental Attributes. Journal of the American Planning Association, Vol. 65, No. 4, 1999, pp. 395–411. 43. Loukaitou-Sideris, A., R. Ligget, H. Iseki, and W. Thur- low. Measuring the Effects of Built Environment on Bus Stop Crime. Environment and Planning B, Vol. 28, No. 2, 2001, pp. 225–280. 44. Carter, M. Gender Differences in Experience with and Fear of Crime in Relation to Public Tranport (abstract only). In Conference Proceedings 35: Research on Women’s Issues in Transportation, Transportation Research Board of the National Academies, Washing- ton, D.C., 2005, Vol. 2, p. 100. 45. Anderson Bomar, M. Technology as a Strategy for Addressing Personal Security Concerns of Women on Public Transit (abstract only). In Conference Proceed- ings 35: Research on Women’s Issues in Transportation, Transportation Research Board of the National Acade- mies, Washington, D.C., 2005, Vol. 2, p. 101. 38 RESEARCH ON WOMEN’S ISSUES IN TRANSPORTATION