Local accessibility, pedestrian travel and neighboring: Testing the claims of new urbanism
 |
|
|
|
|
|
|
|
|
 |
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
Session:Wed., March 14 8:45-11:30am
|
|||||||||||||||||||||||||||
|
Local accessibility, pedestrian travel and neighboring: Testing the claims of new urbanism
|
|
|||||||||||||||||||||||||
Session:Wed., March 14 8:45-11:30am |
|||||||||||||||||||||||||||
|
INTRODUCTION A primary claim of new urbanism is that by placing daily amenities within walking distance of homes and creating a pleasant walking environment, there will be increased opportunities for people to walk in and around their neighborhood and people will take advantage of these opportunities. In turn, this increased pedestrian activity will create opportunities for chance encounters among neighbors at the least, and possibly the formation of friendships. This paper tests these claims, focusing primarily on the significance of local access to parks and/or neighborhood commercial districts. (Since many areas, including Portland, now consider basic pedestrian amenities as norms in development, the study controls for this new urbanism design element in order to focus on the more controversial issue of mixed-uses). The paper focuses on three relationships: that between local accessibility and pedestrian travel, between pedestrian travel and neighboring behavior, and between local accessibility and neighboring. Sociodemographic characteristics, personal attitudes toward pedestrian travel and neighbor interaction, and individual perceptions of the neighborhood environment are also included in the analyses. OVERVIEW OF THE LITERATURE Past studies have examined the relationships between physical neighborhood environments and individual behaviors from a variety of perspectives. Casual contact and interaction among neighbors, for instance, relates to the presence of semi-public spaces, particularly front porches (Abu-Gazzeh 1999, Bothwell, Gindroz and Lang 1998, Brown 1998, Langdon 1997, Skjaeveland and Garling 1997), to human-scale, people-friendly street designs (Appleyard and Lintell 1972, Bosselmann, Macdonald and Kronemeyer 1999, Plas and Lewis 1996), to the presence of trees and other vegetation (Kuo, Sullivan, Coley and Brunson 1998), to the accessibility of parks and other open spaces (Langdon 1997, Bothwell et al 1998), and to the spaciousness and arrangement of open spaces (Abu-Gazzeh 1999, Skjaeveland and Garling 1997). Pedestrian travel is higher in neighborhoods with adequate pedestrian environments (Handy 1992 and 1996b, Kitamura, Mokhtarian and Laidet 1997, Moudon, Hess, Snyder and Stanilov 1996), higher densities (Frank and Pivo 1995, Kitamura et al 1997), and access to local shopping and other retail (Ewing, Haliyur and Page 1995, Frank and Pivo 1995, Gordon and Peers 1993, Handy 1992, 1996a and 1996b, Kitamura et al 1997, Rutherford, McCormack and Wilkinson 1998, Shriver 1997, Steiner 1998). In addition to these objective variables, researchers also identify links between individuals’ own perceptions of the environment and their neighborhood-based behaviors. Skjaeveland and Garling (1997) find that perceptions of spaciousness within a residential environment—particularly when combined with objective measures of the same variable—contribute to higher levels of neighboring. And Handy (1996b) emphasizes the importance of individual perceptions of the pedestrian environment in the choice to walk. Most of these studies control for sociodemographic influences, yet few of them consider another personal variable—attitudes. According to Kitamura et al (1997), however, attitudes may have an even greater impact on travel behavior decisions than do environmental variables and need to be included in future behavioral studies. I did not find any studies on neighboring behaviors or social interaction that consider the implications of attitudes toward neighboring. This paper improves and expands upon this literature in three major ways. The first is by bridging two interrelated fields of research that have been working largely in isolation of one another. On the one hand is urban and transportation planning literature, focusing on the relationships between urban form and travel behavior; on the other hand is community and environmental psychology literature, with its focus on the relationships between physical environments and more socially-oriented behaviors. By bridging these fields, each can learn greatly from the findings and methodologies of the other, and by facilitating a collaborative approach to the understanding of our neighborhoods we can not only increase the speed with which we gain this knowledge but also receive a more complete picture. The second way this paper improves upon existing literature is by further examining three under- (or un-) researched areas—the relative role of attitudes and design elements in predicting behaviors, the link between environmental perceptions and behaviors, and the link between pedestrian travel and neighbor interactions. Finally, to avoid adding to the plethora of existing measurement tools, and the resulting lack of comparable data or proven tools, this study utilizes existing scales developed by Handy (1996b) and Skjaeveland, Garling and Maeland (1996). RESEARCH DESIGN Structured around the theory of new urbanism, the findings presented in this paper examine pedestrian travel behavior and neighbor interaction in eight neighborhoods in the Portland Oregon metropolitan area—four inner-city neighborhoods and four new subdivisions—of varying neighborhood designs. Individual-level data on neighborhood-based behaviors, personal attitudes, sociodemographic characteristics, and neighborhood perceptions are collected through household surveys. The survey instrument includes quantitative questions—used to test the study’s hypotheses—as well as more probing, qualitative questions—used to help explain and understand the quantitative findings. Neighborhood selection The neighborhoods used in this study were selected according to five criteria. The first two—access to retail and access to parks—define the study’s objective evaluations of the neighborhood’s public space structure. The remaining criteria—pedestrian-friendly environment, era of development, and median property value—serve as control variables.
Using these criteria, neighborhoods were selected to represent four variations in the objective neighborhood evaluations—access to both parks and retail, access to park(s) only, access to retail only, and no local access. One established and one new neighborhood were selected for each category (see the neighborhood matrix in Table 1).
Dependent study variables Pedestrian travel behavior. Pedestrian travel behaviors consist of two variables: frequency of "strolling" walk trips and frequency of "destination" walk trips. These forms of pedestrian travel are separated because, according to Handy (1996b), each is associated with a different set of motivational factors. Neighboring behavior. To capture at least some of the multiple dimensions of neighboring (see Skjaeveland et al 1996), neighboring behavior consists of three separate variables: frequency of unplanned interactions (or "chance encounters") among neighbors, weak social ties (or the number of acquaintances one has within close proximity of their home), and supportive acts of neighboring (or the frequency with which one gives/receives assistance to/from their neighbors). (Latent and negative aspects associated with neighboring were also measured in the larger study but are not presented here.) Frequency of unplanned interactions is defined by the number of times in the past week respondents were involved in the following interactions with their neighbors: (a) waved or said hello, (b) stopped and chatted, and (c) invited them inside their home. The remaining variables—Weak Social Ties and Supportive Acts of Neighboring—are measured using Skjaeveland et al’s (1996) Multidimensional Measure of Neighboring (MMN) scale. See Skjaeveland et al (1996) for a description of this scale. Independent study variables Independent variables are grouped into 5 sets: sociodemographic variables, attitudinal variables, objective physical variables, subjective physical variables, and behavioral variables. These sets coordinate with the hierarchical regression Variable sets. Not all variables described here are included in every regression equation. Personal variables. Two sets of variables focus on characteristics of the respondents: sociodemographic and attitudinal. Both sets serve as control variables in the regression models. Sociodemographic characteristics include age group, gender, race, number and ages of children, and whether the respondent identifies as a "homemaker." Other sociodemographic variables also collected, but not included in the regression models due to a lack of variation among respondents, include approximate household income, home ownership, and a more detailed breakdown of race/ethnicity. Attitudinal characteristics measure respondents’ attitudes, on a 3-point scale, toward the importance of walking to daily activities, of interacting with one’s neighbors, and of feeling "at home" in their neighborhood. Neighborhood variables. Variables representing the physical environments of the neighborhoods include both objective evaluations of the neighborhood and respondents’ own subjective evaluations of the environment. The objective physical variables are defined by the neighborhood selection criteria. These include dichotomous variables for: (a) access to local retail only (no parks), (b) access to local parks only (no retail), (c) access to parks and retail, and (d) established neighborhoods. The first three represent neighborhood accessibility characteristics and are the focus of this study. The latter functions more as a control variable, controlling for differences that arise due to variations in the amount of time a neighborhood has had to develop, either socially or physically. The study includes three measures of respondents’ subjective evaluations of the neighborhood: satisfaction with local parks and satisfaction with local retail (measured using a 4-point scale of satisfaction) and "perception of walking in neighborhood" (measured using an 11-item scale developed by Handy 1996b). Behavioral variables. In order to examine the indirect link between neighborhood accessibility and neighboring behavior through pedestrian travel, the regression models for neighboring behavior include walk trip frequencies (destination and strolling) as independent variables. These variables enter the hierarchical models after the personal and environmental variables. Research hypotheses With regard to pedestrian travel, this study hypothesizes that, controlling for relevant sociodemographic factors, measures of walking trip frequencies and neighboring behaviors will be highest in the neighborhood with access to local parks and retail and lowest in the neighborhoods with no local amenities. The study also hypothesizes that, at the individual level, walking trip frequencies will have a significant positive relationship with objective and subjective physical variables, and neighboring behaviors will have a significant positive relationship with objective and subjective physical variables and walking trip frequencies. Data collection Data were collected in the early fall months of 2000 using a mail-out/mail-back survey. Every single-family housing unit in each neighborhood received a 4-page survey—stamped with a numerical code to enable geocoding—for a sample size of 1454. (The study is intentionally limited to single-family units because only two of the eight neighborhoods have any multi-family housing). A total of 499 completed surveys were returned, for an overall response rate of 34 percent. Response rates are slightly higher in the older, traditional neighborhoods (37 to 41 percent response rates) than in the newer subdivisions (24 to 36 percent response rates), but for the purposes of this study each neighborhood is relatively well-represented. Research limitations The most significant limitation of this study is its narrow sociodemographic focus. The New Urbanism developments of the Portland region are nearly homogeneously populated by white, middle-income homeowners. In order to control to the greatest extent possible the influence of these non-design-related factors, it was necessary to match these neighborhoods with neighborhoods of similar sociodemographic characteristics. The study therefore under-represents minority residents, renters, and very low- and very high-income residents. Factors that contribute to community life in neighborhoods of a different ethnic or economic character may differ from those identified in this study and should be the focus of separate research. A second limitation is one common to many survey-based studies—the potential self-selection of respondents. The most effective way to minimize this issue is to maximize the survey response rate, which I attempted to accomplish by (a) using a four-stage mail-out/mail-back data collection process and (b) increasing the number of surveys mailed out in a given site by conducting a population survey rather than a sample survey. Unfortunately, the data collection process yielded an overall response rate of just 34 percent. On the positive side, however, these responses reflect 34 percent of the entire neighborhood populations rather than 34 percent of a sample. Neighborhood self-selection A common concern in studies of neighborhood-level phenomena, particularly those involving new urbanism communities, is that the findings are not indications of neighborhood design influencing or changing individual behaviors, but of the self-selection of residents into neighborhoods that allow them to continue their existing behaviors. Due to the cross-sectional nature of this study, it was not possible to compare the behaviors of respondents before moving into their present neighborhood to their current behaviors. The study does, however, ask respondents about their attitudes toward relevant neighborhood-based behaviors and sentiments, enabling a comparison of these attitudes across neighborhoods in order to detect possible self-selection. I emphasize the word "possible" because the survey cannot detect whether these attitudes were formed prior to or after moving into their current neighborhood. Analyses of covariance reveal that, controlling for significantly correlated sociodemographic characteristics, attitudes toward the importance of walking to daily activities do vary significantly across neighborhood accessibility types (F[3, 473=7.65, p<.01), with respondents from retail-accessible neighborhoods (either alone or in combination with park accessibility) placing significantly higher levels of importance on walking than respondents from neighborhoods with no local amenities. Attitudes toward the importance of interacting with one’s neighbors and of feeling "at home" in one’s neighborhood do not vary significantly across neighborhood accessibility types, but do vary significantly across neighborhood development eras. Respondents from established neighborhoods place a greater value on neighbor interaction and feeling a sense of place than do respondents from new subdivisions (F[1, 470]=8.84, p<.01 and F[1, 464]=5.25, p<.05, respectively). Overall, it appears that respondents may in fact be self-selecting into neighborhoods based on access to local amenities (if they think walking is important) or based on the presence of well-established social and/or physical environments (if they think neighbor interaction and/or sense of place is important). These data do not reveal, however, whether these attitudes are translated into actual behaviors; to analyze this relationship, these attitudinal variables are included in the regression models for pedestrian travel and neighboring behaviors and discussed later. ANALYSIS AND RESULTS Each of the dependent variables presented in this paper is examined using first an analysis of covariance (to detect neighborhood-level variations) and then a hierarchical regression model (to explain individual-level variations). Neighborhood-level analysis The analyses of covariance (ANCOVA) presented in Tables 2 and 3 test the hypotheses that (a) walking trip frequencies and (b) neighboring behaviors will be highest in the neighborhood with access to local parks and retail and lowest in the neighborhoods with no local amenities. For each ANCOVA, sociodemographic variables that correlate significantly with the dependent variable are included as covariates (see the Table footnotes for a listing of covariates). Group memberships are defined according to the neighborhoods’ local accessibility characteristics: Park and Retail Access (Group 1), Retail Access Only (Group 2), Park Access Only (Group 3), and No Local Access (Group 4). Differences in mean values are analyzed overall (with the F statistic) and among specific means. For the analysis among specific means, Group 4 (the neighborhoods with no local amenities) is specified as the control group; Groups 1, 2, and 3 are compared to Group 4 using a simple contrasts model. A second set of ANCOVAs (not presented in tables) are also conducted defining group memberships according to the neighborhoods’ era of development, with established neighborhoods representing Group 1 and new subdivisions representing Group 2. This enables us to detect whether the length of time a neighborhood has had to develop physically or socially is a source of variation in residents’ behaviors.
Two variables vary significantly across neighborhood accessibility types, but not across development eras—destination walk trips and frequency of unplanned interactions. Compared to the No Local Access neighborhoods, destination walk trips are significantly higher in the neighborhoods with retail access, either alone or in combination with access to parks. Unplanned interactions, on the other hand, are significantly higher in the neighborhoods with park access, either alone or in combination with retail access. Differences across development eras, but not across accessibility types, are significant for destination walk trip frequencies (established, Mean=3.12, MSE=0.17; new, Mean=1.97, MSE=0.22) and number of local weak social ties (established, Mean=4.74, MSE=0.06; new, Mean=4.30, MSE=0.08). In each case, mean values for the established neighborhoods are significantly higher than for the new subdivisions at the 95 percent confidence interval. Strolling trip frequencies do not vary significantly across neighborhood accessibility types or development eras. For the remaining variable—supportive acts of neighboring—the ANCOVA findings are less straightforward, with interaction plots revealing an interaction between development era and neighborhood accessibility type in predicting mean values for this variable. I explore this interaction further in the regression analysis for acts of neighboring. With regard to the hypotheses, these findings provide strong support for the hypothesis that walking and neighboring behaviors will be higher in the more accessible neighborhoods in just two cases: destination walk trips and frequencies of unplanned interactions. Some support for this hypothesis exists in the case of supportive acts of neighboring. Individual-level analysis For individual-level analyses, this study uses a hierarchical regression technique. This allows us to examine the relative effect of sets of variables (attitudinal, behavioral, etc.) in addition to the relative effect of individual variables. The order of the hierarchy is structured to first remove potentially confounding variables (in this case sociodemographic and attitudinal attributes, entered as variable sets 1 and 2 respectively) and then to reflect the causal priority of the remaining variables. These remaining sets enter the hierarchical regression equation so that—to the extent possible—no variable set entering later is a cause of an earlier set. Objective physical variables are entered first (as Set 3), followed by subjective physical variables (Set 4) and then, if applicable, behavioral variables (Set 5). Due to space constraints, however, this paper presents only the final models in tabular form.
Overall, the regression model explains about one-third to nearly one-half the variation in Supportive Acts of Neighboring scores (R Square=0.47, p<.01), number of Weak Social Ties scores (R Square=0.39, p<.01), destination trip frequencies (R Square=0.33, p<.01), and frequencies of unplanned neighbor interactions (R Square=0.32, p<.01). The model is less useful, however, in explaining variations in strolling trip frequencies, accounting for just 11 percent of the variation in each case. Pedestrian travel. The number of destination trips an individual makes over the period of a week is linked most significantly to attitudinal factors (R Square Change=0.18, F Change=49.92, p<.01), primarily the importance they place on walking to daily activities, and objective environmental factors (R Square Change=0.11, F Change=17.41, p<.01), primarily retail access—either alone or in combination with park access. The sociodemographic and subjective environmental variable sets each increase the explanatory power by just 2 percent. Within the subjective environmental set, however, "perception of walking in neighborhood" is highly significant in the final model. Relative to the other variable sets, personal attitudes also explain the greatest share of strolling trip frequencies, but to a much lesser extent than in the case of destination trips, accounting for just 4 percent of the variation in trip frequencies. This is followed by the sociodemographic variable set, explaining an additional 3 percent of the total variation, and the objective and subjective environmental sets, each explaining an additional 2 percent. Looking at individual variables, only two are highly significant in the final model—the importance that one places on walking to daily activities and identifying oneself as a "homemaker." Each is related positively to strolling trip frequencies. Also significant, at the 95 percent confidence interval, are "perceptions of walking in neighborhood" (positive relationship), "retail access only" (negative relationship) and the presence of children ages 5 to 12 in a household (negative relationship). With regard to the stated hypotheses, these pedestrian travel findings provide strong support for the hypothesis that destination walk trip frequencies will have a significant positive relationship with objective physical variables, some support for the hypothesis that destination walk trip frequencies will have a significant positive relationship with subjective environmental variables, and no support for the hypothesized relationships between these environmental variables and strolling trips. Frequency of unplanned interactions. The frequency with which one has unplanned, or "chance," encounters with their neighbors (Table 5) is explained most significantly by the behavioral variable set, which includes strolling and destination walk trip frequencies and increases the total explanatory power by 10 percent (R Square Change=.10, F Change=30.37, p<01), above and beyond all other variables. Personal characteristics, however, are also highly significant: sociodemographic variables (Set 1) and attitudinal variables (Set 2) each contribute an additional 9 percent to the total explanatory power, significant at the 99 percent confidence interval. Contributing the least amount of explanatory power are the objective and subjective environmental variable sets, explaining just 2 percent of the total variation once personal characteristics are controlled. Individually, the variables of great significance in the final model include, respectively, strolling trip frequencies, the importance one places on neighbor interactions, having children ages 5 to 12 in the household, and having local park access (without retail). Significant at the 95 percent confidence interval are destination walk trip frequencies, perceptions of walking in one’s neighborhood, and being a "homemaker." All of these variables relate positively to unplanned interactions. Weak social ties. The total explanatory power of the model predicting number of weak social ties (see Table 5) is attributed almost entirely to personal characteristics. Sociodemographic variables, entered first, explain 14 percent of the total variation (R Square=0.14, F=8.67, p<.01). Of primary significance are age group (related positively to weak social ties) and the presence of children ages 0 to 4 (also positive). Attitudinal variables contribute an additional 20 percent (R Square Change=0.20, F Change=67.98, p<.01). Of primary significance in this variable set is the importance one places on interacting with their neighbors. The remaining variable sets—objective environmental, subjective environmental, and behavioral—together increase the total explanatory power by just 5 percent. Within these Sets, only two variables are significant in the final regression equation—strolling walk trip frequencies (Beta=0.13, p<.01) and perception of walking in one’s neighborhood (Beta=0.10, p<.05). Supportive acts of neighboring. The regression model for Supportive Acts of Neighboring also attributes most of its explanatory power to personal variables. The sociodemographic variable set explains 11 percent of the variation in this variable (R Square=0.11, F=7.10, p<.01), with length of residency and the presence of children ages 5 to 12 displaying the most significance, both positive, in the final model. The attitudinal set increases the explanatory power of the model—above and beyond the sociodemographic set—by 31 percent. Of particular significance is the importance one places on neighbor interaction (Beta=0.53, p<.01 in the final model). As a whole, the direct relationships between objective environmental variables and acts of neighboring contribute a nonsignificant amount of additional explanatory power. The interaction between "established neighborhood" and "retail access only" (entered in Set 6), however, is significant, even when all other variables are controlled (Beta=0.17, p<.01). This suggests a positive relationship between retail access only (no parks) and supportive acts of neighboring in established neighborhoods, but not in new subdivisions. (Possible reasons for this interaction are explored in the discussion.) The remaining variable sets—subjective environmental (consisting solely of the "perceptions of walking in neighborhood" variable) and walking behaviors—increase the explanatory power of the equation by, respectively, 3 percent (p<.01) and a nonsignificant amount. With regard to the stated hypotheses, the hypothesis that neighboring behaviors will be positively and significantly related to objective physical factors is somewhat supported for unplanned interactions and acts of neighboring, but not supported for weak social ties; the hypothesis that neighboring will be positively and significantly related to subjective physical factors is somewhat supported for all three neighboring behaviors; and the hypothesis that neighboring will be positively and significantly related to walking behaviors is strongly supported for unplanned interactions, somewhat supported for weak social ties, and not supported for acts of neighboring. DISCUSSION The results presented in this paper provide some support for each of the three relationships tested: the relationships between local accessibility and pedestrian travel, between pedestrian travel and neighboring behaviors, and between local accessibility and neighboring. The results also provide a strong indication, however, that there are other non-design factors, such as personal attitudes, that are also of significance and must be considered in future discussions and research efforts. Local accessibility and pedestrian travel In the case of destination trips, this study supports the relationship between local accessibility and pedestrian travel, with retail access (alone or in combination with parks) relating significantly and positively to the decision to walk to the store. In other words, residents do appear to be using their local retail—and walking to it—if it is available. This supports past findings by Handy (1992, 1996b), Shriver (1997), Steiner (1998), and Rutherford et al (1998). The decision to walk, however, also relates very significantly to the individuals’ attitudes toward walking, supporting the findings of Kitamura et al (1997), and their perceptions of the walking environment, supporting the findings of Handy (1996b). There is not strong support, however, for the relationship between local accessibility and strolling walk trips. First, the regression model developed does not adequately explain variations in the decision to stroll through one’s neighborhood; second, of the variation that it does explain (11 percent), most is attributed to personal attitudes and household characteristics. This latter finding supports Handy (1996b), which finds strolling trips to be related more to individual motivations than to neighborhood characteristics. In spite of the dominance of personal characteristics and attitudes in predicting strolling trips, two non-personal variables do relate significantly to the decision to stroll. These include a positive relationship between strolling and individuals’ own perceptions of the local walking environment, and a negative relationship between strolling and having local retail, but no parks. A likely explanation for this latter finding is provided by the strong positive relationship between retail access and destination trips—in other words, in neighborhoods with retail, walking trips are more likely to be destination trips than strolling trips. This supports the findings of Shriver (1996). Given the variation in trip purposes and in the regression models for each trip type, motivations for walking are surprisingly similar across neighborhoods. Motivations for walking (or not walking) are self-reported and measured using open-ended questions adapted from Handy (1996b). The top two reasons for walking are consistent across all accessibility types: the number one reason is for exercise, fresh air or relaxation (63 to 85 percent of total responses); number two is to walk with one’s children and/or dogs (31 to 45 percent of total responses). There are neighborhood differences, however, in the shares of destination-oriented motivations (walk to shops, etc.); respondents from retail-accessible neighborhood are (not surprisingly) more likely to report destination-oriented motivations than respondents from neighborhoods without access to a commercial area. Reasons for not walking are also similar across accessibility types, with the number one reason being that they are too busy (23 to 48 percent of total responses), and the second and third reasons falling primarily into the categories of health problems or physical limitations and/or poor weather conditions. The number two reason for the neighborhoods with park access but no retail access, however, is that there’s nowhere for them to walk to. This may be an indication of latent demand for walking. Pedestrian travel and neighboring behavior The results presented in this paper support the link between pedestrian travel and the number of unplanned interactions, or "chance encounters," an individual has with their neighbors, as well as the number of weak social ties an individual has near their home. In both cases, it is primarily strolling trips that contribute to these higher levels of neighboring, although destination trips do contribute somewhat to increased chance encounters. As with walking trip frequencies, however, the likelihood that an individual engages in these forms of neighbor interaction also relates very significantly to their attitudes toward neighboring. The likelihood of an individual engaging in supportive acts of neighboring (on either the receiving or giving end), on the other hand, does not relate significantly to either type of walking behavior. It does, however, relate significantly to attitudes toward neighbor interaction, with an individual who feels that interacting with his or her neighbors is important being more likely to engage in acts of neighboring than one who does not. Local accessibility and neighboring behavior Even when individuals’ walking behaviors are controlled, there are still significant relationships between having parks and/or retail in one’s neighborhood and neighboring behaviors. This suggests that these destinations may serve as a place of contact for neighbors, regardless of how they get there. For instance, unplanned interactions are higher among residents who live in a neighborhood with a local park. This provides quantitative support for the assumption that parks serve as a place for neighbors to gather, or to watch their children (or dogs) play—and that residents, at least in these Portland neighborhoods, are using them for such purposes. And in neighborhoods without parks, the presence of local retail relates to higher frequencies of a second neighboring behavior—supportive acts of neighboring—at least in the case of the established neighborhood. Why this relationship does not hold true for the new subdivision is difficult to assess. A possible explanation is that the shops (or at least some shops) in the established neighborhood have developed, over time, as a gathering place or source of neighborhood information, whereas the shops in the new subdivision have not yet made that connection to the community. A second possibility is that the shopping area near the new subdivision draws its customers from a much larger geographic area than does the retail strip in the established neighborhood, reducing or eliminating its "neighborly" feel. Interestingly, individuals’ subjective evaluations of their neighborhood’s walking environment also relate significantly to higher instances of acts of neighboring, weak social ties and chance encounters, even when controlling for actual walking behaviors. This suggests that people-friendly street environments may be more conducive to neighboring, even if people are not using the streets for walking. Pedestrian environments might, for instance, encourage usage of the street space as a gathering place or as a place for children to play. This finding would support those of Appleyard and Lintell (1972). Neighborhood self-selection and behaviors In this section, I return to the question raised earlier in this paper about whether individual attitudes—which do appear to vary by neighborhood—translate into corresponding behaviors. If they do, there may be support for the claim that residents are self-selecting neighborhoods that support their desired lifestyles and that these residential choices—as opposed to variations in neighborhood design—are the contributing factor to behavioral differences. (Again, however, this support if it exists cannot be conclusive as we still cannot determine whether these attitudes formed before or after moving into the neighborhood.) Based on the distribution of attitudes across neighborhood types, we would expect pedestrian travel to be higher in the accessible neighborhoods and neighboring to be higher in the established neighborhoods. As the ANCOVA results indicate, these expectations are met in the case of pedestrian travel for destination purposes, but not for strolling. These expectations are also not met in the case of most neighboring behaviors, with number of weak social ties being the one exception. Neither frequency of unplanned interactions nor supportive acts of neighboring are higher in the established neighborhoods, controlling for relevant sociodemographic variables. Trip substitution For those interested in the transportation implications of increased pedestrian travel, as opposed to the social implications, the survey also incorporated a question (developed by Handy 1996b) that asks respondents what they would have done if they had not been able to walk on their last pedestrian trip. Across all neighborhood accessibility types, respondents were most likely (by far) to have "driven to the same place" if they could not have walked. The second choice, again across all accessibility types, also involves driving, but these respondents would driven to a different place. These findings strongly support the notion that pedestrian trips are in fact replacing automobile trips, and support the recent findings of Person (2001). CONCLUSIONS In summary, one might draw from these findings the following conclusions:
I would like to remind readers that this study was unable to capture variations based on cultural differences, income differences, or differences in homeownership, and that these conclusions should therefore not be extended to neighborhoods with a large ethnic or cultural minority population or with a large share of renters or low- or high-income residents without further study. In addition to cultural and economic differences in community life and the use of public space, a number of other questions arise from the research presented here. First, what factors do relate to an individual’s decision to stroll through their neighborhood? Is it owning a dog? Being health-conscious? Working in a stressful environment and thereby needing a "release"? Given the strong relationship between strolling and interacting with one’s neighbors, this question should be of interest to community developers. Second, why do residents of traditional neighborhoods place a higher value on the importance of neighboring and feeling "at home" in their neighborhood, but not actually engage in neighboring behaviors more frequently than residents of new subdivisions? This also leads to a third question of how and when people’s attitudes toward travel behavior and community life form. Finally, to follow up on the link between perceptions of the local walking environment and neighbor interactions even when controlling for pedestrian behavior, we need to learn more about how individuals define "friendly" streetscapes, if and how these definitions vary across populations, and what role these streets play in the life of a community (in addition to providing a place to walk). And the list could certainly go on. Some of the questions raised here can be addressed in another, more focused, household survey; others will require more in-depth case studies. They are all questions, however, that we need to continue asking—in a variety of formats and across a variety of social and physical environments—in order to make informed decisions about how we develop and change our neighborhoods. BIBLIOGRAPHY Abu-Ghazzeh, Tawfiq M. 1999. Housing layout, social interaction, and the place of contact in Abu-Nuseir, Jordan. Journal of Environmental Psychology 19:41-73. Appleyard, Donald, and Mark Lintell. 1972. The environmental quality of city streets: The residents' viewpoint. Journal of the American Institute of Planners March:84-101. Bosselmann, Peter, Elizabeth Macdonald, and Thomas Kronemeyer. 1999. Livable streets revisited. Journal of the American Planning Association Spring:168-180. Bothwell, Stephanie E., Raymond Gindroz, and Robert E. Lang. 1998. Restoring community through traditional neighborhood design: A case study of Diggs Town public housing. Housing Policy Debate 9 (1):89-114. Brown, Barbara B., John R. Burton, and Anne L. Sweaney. 1998. Neighbors, households, and front porches: New urbanist community tool or mere nostalgia? Environment and Behavior 30 (5):579-600. Ewing, Reid, Padma Haliyur, and G. William Page. 1995. Getting around a traditional city, a suburban planned unit development, and everything in between. Transportation Research Record 1466:53-62. Frank, Lawrence D., and Gary Pivo. 1995. Impacts of mixed use and density on utilization of three modes of travel: single-occupant vehicle, transit, and walking. Transportation Research Record 1466:44-52. Gordon, Stephen P., and John B. Peers. 1993. Designing a community for transportation demand management: The Laguna West pedestrian pocket. Transportation Research Record 1321:138-145. Handy, Susan L. 1992. Regional versus local accessibility: Neo-traditional development and its implications for non-work travel. Built Environment 18 (4):253-267. Handy, Susan L. 1996. Understanding the link between urban form and nonwork travel behavior. Journal of Planning Education and Research 15:183-198. Handy, Susan L. 1996. Urban form and pedestrian choices: Study of Austin neighborhoods. Transportation Research Record 1552:135-144. Kitamura, Ryuichi, Patricia L. Mokhtarian, and Laura Laidet. 1997. A micro-analysis of land use and travel in five neighborhoods in the San Francisco Bay Area. Transportation 24:125-158. Kuo, Frances E., William C. Sullivan, Rebekah Levine Coley, and Liesette Brunson. 1998. Fertile ground for community: Inner-city neighborhood common spaces. American Journal of Community Psychology 26 (6):823-852. Langdon, Philip. 1997. Can design make community? The Responsive Community Spring:25-37. Moudon, Anne Vernez, Paul M. Hess, Mary Catherine Snyder, and Kiril Stanilov. 1996. Effects of site design on pedestrian travel in mixed-use, medium-density environments. Transportation Research Record 1578:48-55. Plas, Jeanne M., and Susan E. Lewis. 1996. Environmental factors and sense of community in a planned town. American Journal of Community Psychology 24 (1):109-143. Rutherford, G. Scott, Edward McCormack, and Martina Wilkinson. 1998. Travel impacts of urban form: Implications from an analysis of two Seattle area travel diaries. Washington, D.C.: Bureau of Transportation Statistics, U.S. Department of Transportation. Shriver, Katherine. 1996. Influence of environmental design in pedestrian travel behavior in four Austin neighborhoods. Transportation Research Record 1578:64-75. Skjaeveland, Oddvar, Tommy Garling, and John Gunnar Maeland. 1996. A multidimensional measure of neighboring. American Journal of Community Psychology 24 (3):413-435. Skjaeveland, Oddvar, and Tommy Garling. 1997. Effects of interactional space on neighbouring. Journal of Environmental Psychology 17:181-198. Steiner, Ruth Lorraine. 1998. Traditional neighborhood shopping districts: Patterns of use and modes of access. Journal of the American Planning Association.
Author and Copyright InformationCopyright 2001 by Author Hollie Lund Person is an Urban Studies doctoral candidate at Portland State University. Her primary interests lie in the social, travel and environmental implications of neighborhood environments, with particular emphasis on the land use, transportation and urban design attributes of neighborhoods. The author can be contacted at mhperson@earthlink.net. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||