SEARCH

SEARCH BY CITATION

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

The research explores the impact of socioeconomic and racial variables on accessibility to urban amenities and travel in compact urban built environments that have traditionally been viewed as improving access to daily destinations and promoting nonmotorized travel: urban environments characterized by high densities, mixed land uses, and high connectivity. The study focuses on six neighborhoods in the Detroit region. Two neighborhoods are within the city itself, and predominantly poor and Black, and four of the neighborhoods are in the region surrounding the city, and they are predominantly wealthy and White. This study at the neighborhood scale enables an analysis into how class and race affect accessibility and travel in neighborhoods experiencing urban disinvestment and decline. The research shows that the traditional relationship between high densities, mixed land uses, high connectivity, greater accessibility, and pedestrian activity is significantly weaker in declining inner cities.

More than any other elements, decentralization and dispersion define U.S. cities as we go into the twenty-first century. Much research into the environmental, economic, and social dimensions of the density of the built environment (high density vs. low density) has emerged, focusing in particular on the coupling of ongoing decentralization with the rapid decline of central cities. This combination of factors has followed clear lines of class and racial composition that have created poverty-stricken enclaves that are ethnically and racially entrenched. Such conditions of decline within cities are seen as facilitating perpetual cycles of disadvantage in shaping the “local burdens of place” (Vojnovic et al., 2013). Many U.S. cities are characterized by the coupling of extreme urban decline and inefficient suburbanization, including Youngstown and Dayton (Ohio), Buffalo (New York), Flint (Michigan), and others. However, with the release of 2010 Census data, Detroit emerges as the most dramatic example. The city of Detroit lost a quarter of its population during the decade 2000–2010—from a population of 951,270 in 2000 to 713,777 in 2010. This can be seen as a loss of some 24,000 people annually, over 63 people every day, or close to 3 people every hour. Since the 1950s, when Detroit had 1.85 million people, its population has been cut by more than 60%. As one might imagine, severe local stresses become evident when a city built to accommodate 1.85 million houses only some 700,000 within its boundaries. Chris Hansen (2010), on Dateline NBC, described the city's condition in this way:

They litter the landscape, thousands and thousands of abandoned homes. And just like these buildings, Detroit is a shell of its former self. One third of the people here live in poverty. Almost half the adults are illiterate, and about 75 percent of kids drop out of school. I could be describing some ravaged foreign nation, but this is the middle of America.

Metro Detroit is now generally perceived as the U.S. urban region most affected by the extreme urban decline associated with economic globalization, deindustrialization, neoliberal policies, and excessive suburbanization. This inefficient decentralization and dispersion, following clear class and racial dimensions, has come to define the U.S. model of white flight. In 1990, 75% of Detroit's population was Black; by 2010, the Black population had increased to 83% (U.S. Census Bureau, 2011). Decades of research on segregation and suburbanization in Detroit have shown that these development patterns did not occur by chance, but were controlled by apartment house managers, real estate brokers, and builders, supported by wealthyWhite suburban residents (Thomas, 1997).

The scale of suburbanization in the Detroit region is effectively demonstrated by the fact that the conversion of agricultural and natural lands to urban uses between 1960 and 1990 occurred at a rate that was 13 times greater than population growth in the area (Public Sector Consultants, 2001). Moreover, the scale of decentralization continues to escalate. Whereas the average residential density of housing developments in the Detroit region was 2.84 units per acre in 1990, throughout the decade of the 1990s new construction was built at an average density of only 1.26 housing units per acre, less than half the residential density of 1990 (SEMCOG, 2003a). Both the urban decentralization in the region and the population exodus from the city are reflected in Detroit's declining density, which fell from 13,249 people per square mile (ppsm) in 1950 to 5,170 ppsm in 2010 (U.S. Census Bureau, 2011). In terms of economic impact, Detroit's suburbanization is paralleled by the decentralization of the region's tax base and a polarization of fiscal capacity between the declining city and its wealthy suburbs. In 2000, the per capita taxable assessment in the city of Detroit was $7,573, while the per capita taxable assessment in Bloomfield Hills was $165,794 and $160,905 in Bingham Farms (SEMCOG, 2003b).

This polarization between the city and its suburbs is also reflected in basic socioeconomic profiles. In 2008, while the national poverty rate in the United States was 13.2%, and while the poverty rate in the Detroit-Warren-Livonia Metropolitan area stood at 9%, the poverty rate in the city of Detroit was 33.1%. In addition, while in 2008 the national per capita income was $27,589, and while in the Detroit-Warren-Livonia Metropolitan area the per capita income was $27,624, the city of Detroit maintained an average per capita income of $14,976 (U.S. Census Bureau, 2010a). Simply put, since the early 1960s, as the middle- and upper-income White population moved out, the city of Detroit has evolved into a place characterized by increasing concentrations of unemployment, poverty, visible minorities, and crime.

In studies of Detroit, a particular point of discussion in recent years has been how urban disinvestment affects access to urban amenities. For instance, with regard to public services, in 2010 the Detroit Police Department employed less than 3,000 officers to cover the same spatial area that in 1970 some 5,000 officers patrolled (Okrent & Gray, 2010). In 1970, the average drive for a police cruiser from its precinct station to a crime scene was about three miles; today, there are many homes that are more than 7.5 miles away from the nearest police station. Similar stresses in service provision confront the delivery of every other service within the City, including snow removal, road maintenance, trash collection, and street lighting. Perhaps the most significant local pressure in service provision is the effect on public education. In the 2003–2004 academic year, less than 25% of students graduated from the Detroit City School District's high schools, the lowest graduation rate in the country among the 50 largest U.S. cities (Swanson, 2008).

Following the residents, businesses have also left the city. In Andrew Grossman's (2009) Wall Street Journal article, “Retailers Head for Exits in Detroit,” he notes that even Starbucks, known for saturating U.S. cities with its coffee shops, has only four Detroit locations. Carrying on this theme, Roy Greenslade (2009) of London's Guardian describes Detroit as “a city where people pay $4 for a latte on one corner—if they can find it—and $10 for a rock of cocaine on the other.”

In this exodus of retail from the city, perhaps the most extensively covered topic by the media in recent years has been the loss of all major supermarket chains once the last two Farmer Jacks closed in 2007. As Grossman (2009, p. A3) notes in the Wall Street Journal:

No national grocery chain operates a store here. A lack of outlets that sell fresh produce and meat has led the United Food and Commercial Workers union and a community group to think about building a grocery store of its own.

In a similar spirit, Steve Hargreaves (2009), in a CNN piece called “Hunger Hits Detroit's Middle Class,” notes that “[f]ood has long been an issue in this city without a major supermarket. Now demand for assistance is rising, affecting a whole new set of people.” Hargreaves goes on to describe the new value placed on food in Detroit:

On a side street in an old industrial neighborhood, a delivery man stacks a dolly of goods outside a store. Ten feet away stands another man clad in military fatigues, combat boots and what appears to be a flak jacket. He looks straight out of Baghdad. But this isn't Iraq. It's southeast Detroit, and he's there to guard the groceries. “No pictures, put the camera down,” he yells.

In the context of Detroit's rapid urban exodus, this article explores how urban decline has affected access to amenities within the city. Using built environment objective data and qualitative survey data, we examine the impact of socioeconomic variables on access to urban amenities and travel in compact urban environments that have traditionally been viewed as improving accessibility to destinations and promoting nonmotorized travel. In other words, the study explores how access is affected by urban disinvestment and decline in neighborhoods characterized by higher urban densities, a concentrated land use mix, and high connectivity. Within the existing urban morphology literature, the relationships between class, race, and the urban built environment are recognized as underrepresented in research on community planning and design (USDHHS, 2000; Day, 2003, 2006; Vojnovic, 2006).

The study focuses on six neighborhoods in the Detroit area. Two neighborhoods are in the city, and predominantly poor and Black, and four neighborhoods are in the surrounding region, and predominantly wealthy and White. The neighborhood scale study enables an analysis into how class affects accessibility, travel, and public health. The research shows that traditional relationships between higher densities, mixed land uses, higher connectivity, and greater accessibility are not as strong—and can break down—in declining cores experiencing disinvestment. Simply put, disinvestment reduces access to amenities, including personal services, leisure, and healthy food options. The research reveals that socioeconomic variables can outweigh the importance of urban form in shaping access, travel, and health outcomes.

LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

Extensive research has focused on the impacts of the built environment on travel behavior, such as mode of travel, travel distance, and travel frequency. Three characteristics in urban form—density, land use mix, and connectivity—have been found to have considerable impacts on travel by affecting accessibility across in-networked space. While high densities, a fine-grained mix and concentration of land uses, and highly connected street systems facilitate nonmotorized travel, these environments do not allow motorists to realize the full speed capabilities of the car (Boarnet, 2011; Cao, 2010; Southworth, 2005; Sui, 2003). In contrast, built environments characterized by low densities, single-use zoning, and disconnected street networks, while less accommodating to pedestrians, are successful in accommodating high-speed automobiles. These differences in urban form are evident between older inner cities (as in Boston or New York) that were built prior to the diffusion of an affordable automobile, and post–World War II suburban landscapes or newer cities (such as Houston or Phoenix), which were built largely to accommodate high-speed automobile travel.

High densities, concentrated and mixed land uses, and high connectivity can reduce distances between destinations—all else being the same—encouraging nonmotorized travel, such as walking and bicycling (Boarnet, Joh, Siembab, Fulton, & Nguyen, 2011; Day, 2003, 2006; Moudon & Lee, 2003; Lee & Moudon, 2004; Vojnovic, 2000; Yang, Diez Roux, Auchincloss, Rodriguez, & Brown, 2011). The U.S. Department of Transportation (2005) has shown that keeping distances between destinations at less than one-third of a mile would ensure that some 45% of the U.S. population would be willing to walk. These findings are consistent with a number of other prior studies that have shown that maintaining short distances is a critical variable in promoting pedestrian activity (Ewing & Handy, 2009; Powell, Martin, & Pranesh, 2003; USDOT, 1994). As noted by the U.S. Department of Transportation (1994, p. 12), “distance is almost certainly the key factor limiting utilitarian trips” by nonmotorized travel.

Built environments that promote the use of public transit maintain similar characteristics as pedestrian-oriented built environments, in part due to cost-effective transit network requirements, which necessitate higher densities and concentrated and mixed land uses. In addition, built environments that accommodate mass transit must also have a pedestrian focus, because transit users are also pedestrians; walking to the transit line and walking once they arrive at their destination. As Newman and Kenworthy (2013, p. 236) maintain, to be successful “TODs [Transit Oriented Development] must also be PODs [Pedestrian Oriented Development].”

Urban Density

Urban researchers have shown that increasing residential and employment concentrations over any given area of land can reduce distances between destinations and promote walking. Peter Newman and Jeffrey Kenworthy (1989) completed one of the early and seminal studies on the relationship between population and employment densities and travel patterns, showing that increasing densities encouraged walking and public transit use. They replicated the study a decade later (Newman & Kenworthy, 1999), and similar relationships were evident, although the U.S. population had become more automobile-dependent, with fewer people walking.

Other researchers that have focused on the relationship between urban densities and travel have found similar outcomes, albeit to varying degrees (Cao, Mokhtarian, & Handy, 2007a; Mitchell, Hargreaves, Namdeo, & Echenique, 2011; Yang, French, Holt, & Zhang, 2012). For instance, Robert Cervero, who explored the relationship between urban environments and public transit use, argued that for every 10% increases in population and employment densities, transit ridership increased between 5% and 8% (Cervero, 1998, p. 72).

Land Use Mix

Studies have shown that a balanced mix and a concentration of different land uses (retail, commercial, residential) can reduce distances between daily activities and encourage nonmotorized travel (Boarnet, 2011; Ewing & Handy, 2009; Frank & Pivo, 1995; Saelens, Sallis, Black, & Chen, 2003; Southworth, 1997). There are two dimensions to land uses, their mix and concentration, which affect two different types of accessibility (Ewing, 1997). One aspect to the consideration of land uses is the relationship between where people live and their proximity to out-of-home destinations, such as work or shopping. With this type of accessibility (known as residential accessibility), the greater the mix of residential to nonresidential uses, the greater the likelihood of reducing distances between one's home and various destinations. Thus, improving residential accessibility (reducing distances between residential and nonresidential land uses) encourages walking from one's home to shopping, personal services, and other out-of-home activities.

In the current U.S. urban context, characterized by high levels of urban decentralization and dispersion, the importance of residential accessibility in shaping travel is evident in the jobs-housing balance and the commute to work. While some metropolitan areas have been successful in maintaining a balance between jobs and housing throughout the region, many metropolitan areas, due to excessive suburbanization and the residential exodus from the core, have evolved highly unbalanced land uses, forcing long worker commutes. For instance, in 1997, the city of Houston maintained some 137,000 jobs in its urban core, but only about 2,000 people lived within this district (Vojnovic, 2003). In the case of Detroit, for approximately every 16 jobs there was only one person that lived in the central business district (CBD), while in Los Angeles the CBD maintained some 18 jobs for every person that lived in the city's core (Newman & Kenworthy, 1999). This regional spatial structure—the “donut effect,” as residents flee the city to the suburbs—forces long work commutes as people living in the suburbs access employment in the city center. It should be noted, however, that despite the suburb-to-urban commute to work, the most significant commuting pressures in the United States are suburb-to-suburb.

The second dimension to the land use mix addresses accessibility between various out-of-home destinations. High concentrations of daily activities (such as work, shopping, and personal services) shortens distances between various daily destinations, improving a second type of accessibility, destination accessibility (Ewing, 1997). A concentration of destination activities (as evident in downtowns, shopping malls, or suburban activity nodes) shortens distances between out-of-home activities, promoting walking and the use of public transit. Even if a person drives into an activity center, they can park the car and complete several objectives by walking. This discussion should lead to an understanding of why high concentrations of residential and nonresidential land uses—such as in central New York City, San Francisco, or Boston—ensures both residential and destination accessibility, and facilitates high levels of nonmotorized travel.

Connectivity

Connectivity is determined by the extent to which different parts of a neighborhood, and also different neighborhoods, are linked to each other. At the most basic level, movement through a city is restricted by infrastructure networks (such as streets), houses, fences, and geographic features (including mountains and waterways) (Griffith, Vojnovic, & Messina, 2012). In order to reduce distances between destinations, continuous rights of way must be provided within and between neighborhoods that allow for minimal distances between various destinations within a city. However, the irregular street networks in automobile-oriented urban environments, characterized by curvilinear streets and cul-de-sacs, forces longer trip lengths between destinations and impedes pedestrian travel (Owens, 1993; Ewing & Cervero, 2010).

Street networks in modern suburbs are designed for low connectivity, in part to ensure privacy by keeping out through traffic and unwanted visitors (Vojnovic, 2006). The suburban discontinuous road network was designed to be traveled by car at speeds of 40 mph, as opposed to pedestrian travel at 3 mph. The block structure within these urban environments is characterized by the development of residential, retail, and commercial pods, all highly isolated from each other. While Euclidean distance (the so-called “bird's flight” distance) might be short, the street network distance, due to the lack of connectivity between the various pods, forces long travel distances. In contrast, street networks characterized by the grid, porous block structures with highly connected street networks, reduce distances between destinations and encourage walking (Baran, Rodriguez, & Khattak, 2008; Boarnet, 2011; Frank et al., 2005; Southworth, 1997, 2005).

Culture, Values, and Habits Shaping Travel Behavior

While urban form and accessibility play an important role in shaping travel, there are variables other than distance that are equally important, or even more so, in affecting pedestrian activity. These include personal variables (such as values, habits, and lifestyles), environmental variables (such as safety, climate, topography, and quality of place), and peer group acceptance (Rapoport, 1987; USDOT, 1994). Culture, in particular, has been viewed as critical in shaping behavior. Culture is a description of particular patterns in life and it is formed by the understanding of what activities are considered appropriate in particular settings. Many personal variables—including values, habits, and preferences—that affect travel are shaped by culture. As Vojnovic, Smith, Kotval-K, and Lee argue (2008, p. 100), “designing pedestrian-inviting streetscapes will have little impact on encouraging non-motorized travel if the population considers walking and cycling undesirable.” Similarly, Amos Rapoport (1987, p. 83) maintains that

Cultural variables are primary for any activity, including walking and others, occurring in streets. It is culture that structures behavior and helps explain the use or non-use of streets and other urban spaces—or of other settings. Thus, the use of streets by pedestrians is primarily culturally based, since physical environments do not determine behavior. Physical environments, however, can be supportive or inhibiting.

Culture is seen as influencing travel at many different levels. For example, culture is considered key in shaping shopping preferences, and hence travel outcomes. While the French have a preference for traditional, small-scale local retailers (bakers, butchers, and fishmongers)—an important element of the French food culture—the British have been moving away from these traditional store types, increasingly preferring to shop in larger supermarkets (Pettinger, Holdsworth, & Gerber, 2007). The store of choice, or of preference, can thus be more important than the actual distance or accessibility to a store in influencing travel patterns.

The influence of culture on travel behavior is also evident with neighborhood self-selection. Studies have shown that homebuyers who purchase housing in pedestrian-oriented neighborhoods do so, in part, because of specific values that place an importance on walking. For instance, Susan Handy's (1996) study of Austin, Texas shows that residents walked more in high-density, connected neighborhoods with mixed land use. However, her research also revealed that residents purchased housing in these pedestrian-oriented neighborhoods in part because of their preference for walking. Hence, personal values will influence neighborhood choice—whether to buy or not to buy a home in a walkable neighborhood—a decision that will reinforce whether specific urban environments are, or are not, pedestrian-oriented.

Research on neighborhood self-selection also shows, however, that while self-selection will influence travel, characteristics in the built environment maintain a separate and significant impact on travel and pedestrian activity (Cao, 2010; Cao, Mokhtarian, & Handy, 2007b; Chatman, 2009). Cao, Mokhtarian, and Handy's (2009) review of 38 travel behavior studies shows that, although most research reveals evidence of self-selection, yet almost all studies found that when controlling for self-selection there was a statistically significant influence of urban form on travel. Research continues to show that while self-selection accentuates the impact of the built environment on travel behavior, the built environment still maintains a separate influence on travel, an influence equal to or greater than the influence of neighborhood self-selection. Thus, while residents might buy a house in a neighborhood because it is pedestrian-oriented, this pedestrian-oriented neighborhood—because of its physical qualities—will also facilitate walking and cycling.

Urban Disinvestment, Accessing Urban Amenities, and Travel Behavior

While the existing research on urban form, accessibility, and travel has demonstrated that higher densities, mixed land uses, and increased connectivity encourage nonmotorized travel, this research has generally focused on urban environments with robust amenities. An important aspect of the research that has not been explored is the access to amenities and the nature of travel that develops in high-density, connected neighborhoods with mixed land use that are experiencing urban disinvestment and decline, as in Detroit. By exploring neighborhood structure and qualitative survey data, this research shows that the relationship between compact, connected neighborhoods with mixed land use and pedestrian activity is not as easily replicable in areas experiencing decline. The lack of access to amenities in neighborhoods experiencing disinvestment emerges as an important variable in shaping access and travel.

Some aspects of the relationship between access to urban amenities and urban decline have, to a certain degree, been covered by the food deserts literature. These studies, however, are narrower in scope, generally focusing on the location of food sources and not the wider relationships of travel to various destinations, including shopping, leisure, and personal services. Much of the food deserts literature also does not capture the travel component in shopping. This literature tends to explore the concentration of store types by neighborhood, not examining whether residents actually shop or eat at these locations. The dimension of store preference is simply not an element in many of these studies.

An array of research in the United States has shown that predominantly minority and low-income urban neighborhoods have limited access to affordable, nutritious, and culturally appropriate food sources (Howard & Fulfrost, 2007; Galvez et al., 2008). In comparison to more affluent and predominantly white urban and suburban neighborhoods, socially and economically disadvantaged urban neighborhoods have fewer large-scale retail supermarkets and an overabundance of convenience and liquor stores (Moore & Diez Roux, 2006; Powell, Slater, Mirtcheva, Bao, & Chaloupka, 2007a; Azuma, Gilliland, Vallianatos, & Gottlieb, 2010). Consequently, residents in these neighborhoods face higher food prices and a limited availability of nutritious food staples which when present tend to be of lower quality than those available to their suburban counterparts (Zenk et al., 2005, 2006; Franco, Diez Roux, Glass, Caballero, & Brancati, 2008; Smith et al., 2010).

Methodologically, many of these studies assume that the closest neighborhood shops are the stores of choice for residents. As a result, these studies fail to examine where people actually shop for their food provisions. Such an assumption ignores many of the insights garnered from the store choice and disadvantaged consumer literature which has shown that the shopping and travel patterns of economically deprived consumers are multifaceted and complex. Similar to more affluent shoppers, disadvantaged consumers tend to shop at retail supermarkets located outside their local neighborhood, but rely on smaller neighborhood corner stores to supplement their food budgets (Piacentini, Hibbert, & Al-Dajani, 2001; Clifton, 2004; Gittelsohn et al., 2007).

Therefore, a critique of the food deserts research is that many of these studies rely on geographic proximity to link individual health data to the retail food environment. While some studies have attempted to incorporate the shopping preferences and travel behaviors of residents living in a food desert (Inagami, Cohen, Finch, & Asch, 2006), much of the research assumes that residents shop at their neighborhood stores when linking individual health data with the neighborhood food environment. Such an assumption not only distorts on-the-ground reality but also denies the agency of residents to overcome their adverse urban food environment (LeDoux & Vojnovic, 2013).

CASE STUDIES AND METHODS

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

This study utilizes built environment objective data and qualitative survey data to analyze the relationships between urban form, accessibility, and pedestrian activity in neighborhoods experiencing disinvestment and decline. The research focuses on six four-square-mile neighborhoods in the Detroit area. Two neighborhoods are in the city, in east side Detroit, and four neighborhoods are in the Detroit region surrounding the city (Figure 1). The neighborhoods were selected to allow some control for built environment and demographic characteristics. Using census data, land parcel maps, and site surveys, six neighborhoods were selected based on income, density, connectivity, and land use mix. Four suburban neighborhoods in the Detroit region were selected with similarly high incomes, but two of these are of medium density, with a mix of land uses, and high connectivity (Ann Arbor and Birmingham), and two of the neighborhoods are typically automobile-oriented, that is, low-density and low-connectivity, with restrictive land uses (Bloomfield Hills and West Bloomfield). These four neighborhoods enable a comparison of different built environments among a similar socioeconomic subgrouping (Figures 2 and 3).

image

Figure 1. Map of Detroit Region Neighborhoods

Download figure to PowerPoint

image

Figure 2. Photos of Ann Arbor and Birmingham Neighborhoods

Download figure to PowerPoint

image

Figure 3. Photos of Bloomfield Hills and West Bloomfield Neighborhoods

Download figure to PowerPoint

Two Detroit urban neighborhoods (Figures 4 and 5) were selected that maintained medium densities, mixed land uses, and high connectivity (a built form comparable to neighborhoods in Ann Arbor and Birmingham) but that were experiencing disinvestment and decline (see Tables 1 and 2). These four neighborhoods allow a comparison of similar built environments (in terms of density, land use mix, and connectivity) but with two neighborhoods being of high socioeconomic status and two neighborhoods of low socioeconomic status. Information on land uses (residential, commercial, retail, and industrial) and building types (single family, duplexes, apartments, and factories) were also collected (see Table 1 and Figures 6 and 7). Given the high number of abandoned properties, the visible abandonment of housing was also documented.

Table 1. Urban Density, Land Uses, and Connectivity in the Detroit Region
Density
   Ann BloomfieldWest
Buildings per sq. mi.Detroit 1Detroit 2ArborBirminghamHillsBloomfield
  1. *Visibly abandoned buildings recorded during the land use surveys.

  2. Note: From the 2000 Census, which we used to define the case studies, the tracts that made up the two urban Detroit neighborhoods maintained a residential density of 6,314 people per square mile. It should be noted that a large segment of Detroit 2 is a Chrysler industrial plant, evident in figures 6 and 7. If the area of the plant is not included in the calculation of density, the actual density of the two Detroit neighborhoods stands at 6,914 people per square mile. In the year 2000, the tracts that made up the higher density suburban municipalities maintained a density of 4,696 people per square mile. In contrast, the tracts that made up the two lower density suburban neighborhoods maintained a density of 1,797 people per square mile (U.S. Census Bureau, 2000; MCGI, 2010).

  3. By 2010, this would be two years past our survey collection, there was a significant urban exodus. In 2010, the density of urban Detroit stood at 3,928 people per square mile, and if the Chrysler plant is removed, the figure becomes 4,227 people per square mile. The two higher density suburban neighborhoods in 2010 maintained a density of 4,723 people per square mile, while the two lower density suburban neighborhoods maintained a density of 1,711 people per square mile (U.S. Census Bureau, 2010b; MCGI, 2010).

  4. The survey was completed in 2008, so it could be assumed that the density of the two urban Detroit neighborhoods was not as low as the 3,928 people per square mile, which was recorded for the 2010 Census. The two neighborhoods in urban Detroit lost a population of 2,386 people per square mile between the last two census decades (from 2000 to 2010), so a loss of about 477.2 people per square mile every 2 years. It could be reasonably assumed that even without removing the land from the Chrysler plant, the density of urban Detroit was around 4,405 people per square mile when the survey was taken in 2008. At 4,405 people per square mile, the density of the two Detroit neighborhoods was comparable to the higher density suburban neighborhoods that averaged 4,723 people per square mile and much higher than the density of the two low density suburbs that averaged 1,711 people per square mile.

Single family-detached home1542.11024.71575.41416.3424.0716.3
Semi-detached0.02.06.70.00.019.0
Apartment2.05.642.919.619.48.4
Townhomes/ rowhouses3.014.28.518.825.130.2
Retail5.93.24.219.61.60.0
Service36.918.830.527.212.81.6
Public institution31.213.920.18.08.20.0
Industrial3.72.22.50.00.00.0
Abandoned*169.5111.90.00.00.00.0
Connectivity
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
Intersections per sq. mi.77.749.231.141.45.25.2
Table 2. Neighborhood Respondents by Race, Education, and Household Incomes (%)
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
Race (% White /% Non-White)
 White (%)5.0010.6493.6894.1891.6384.21
 Non-White (%)95.0089.366.325.828.3715.79
Educational Attainment (% of respondents in each category)
 No high school15.9720.920.000.001.923.06
 High school diploma52.1050.339.6210.778.6510.20
 2-yr assoc. degree13.4510.464.816.156.257.14
 4-yr college degree11.768.5031.2734.3635.1031.63
 Graduate or professional degree6.729.8053.2648.7248.0847.96
Household Incomes by Neighborhood (% of respondents in each category)
 Less than 20k45.3754.017.062.452.223.64
 20k to 40K31.4821.906.694.917.224.85
 40k to 60k10.1913.8715.249.208.339.70
 60k to 100k8.337.3030.4819.0225.0020.61
 100k to 150k4.632.9224.9126.9920.5632.73
 Greater than 150k0.000.0015.6137.4236.6728.48
image

Figure 4. Photos of Detroit East Side Neighborhoods

Download figure to PowerPoint

image

Figure 5. Photos of more stable areas of the Detroit East Side Neighborhoods

Download figure to PowerPoint

image

Figure 6. The Urban Forms of the Six Detroit Region Neighborhoods Modeled in 3D CAD

Download figure to PowerPoint

image

Figure 7. Land Uses in the Detroit Region Neighborhoods

Download figure to PowerPoint

A stratified random mail survey collected data on travel, exercise, diet, and other personal variables. A total of 1,191 surveys were collected (128 from Detroit 1, 158 from Detroit 2, 297 from Ann Arbor, 196 from Birmingham, 211 from Bloomfield Hills, and 201 from West Bloomfield). The eight-page survey included detailed questions on travel (frequency and purpose of trips, travel mode, and destinations) for a full array of trips (shopping, personal services, and leisure destinations). Respondents were asked to report travel behavior over a typical week while considering seasonal distinctions (i.e., winter versus summer travel). For physical activity measures, respondents were asked about both moderate (casual walking, gardening, vacuuming) and vigorous (running, shoveling snow) physical activity throughout the week (frequency and length). Some 8% of returned surveys were removed as outliers and because of insufficient data.

The research team was confronted with a rapid increase in foreclosures and vacancies in the Detroit region as the growing mortgage crises spread across the United States. In a period of three months—from the point at which addresses of occupied residences were obtained from the local postal offices to the point at which random samples were selected and the project introduction prompts, survey packages, and the two reminder prompts were sent—a total of 909 of the selected dwellings had been vacated. Over 75% of the vacated houses were in the Detroit neighborhoods.

The survey response rate was 20%. Given that the mail-out survey was administered to the general population, this response rate is considered good (Sommer & Sommer, 1997). Research has shown that lower return rates can be expected among racial minorities, individuals with fewer years of schooling, high-density urban areas, and high crime rate neighborhoods (Groves & Couper, 1998; Siegel, 2002; Zimowski et al., 1997). In addition, response rates for household travel surveys in the United States have been declining, with a number of travel surveys reporting response rates as low as 5% (Zimowski et al., 1997). Given the socioeconomic, racial, and neighborhood characteristics of the Detroit region, the response rates from this project are comparable to other similar U.S. mail surveys. However, it should be acknowledged that the 20% response rate and limitation to six neighborhoods may limit the research findings.

FINDINGS

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

Ethnic and Socioeconomic Profiles

In the two east side Detroit neighborhoods, about 92% of the respondents were non-White, while in the four suburban neighborhoods over 91% of the respondents were White (see Table 2). More specifically, in the Detroit neighborhoods, over 90% of the respondents were Black. A similar polarization between the urban and suburban respondents was also evident with socioeconomic profiles. In educational attainment, while about 70% of the Detroit urban respondents had either no high school degree or just a high school degree, over 82% of the suburban respondents had either a 4-year college degree or graduate degree (Table 2). The lower educational attainment in Detroit paralleled lower personal and household incomes.

Health Profiles

With self-reported values on height and weight, the body mass index (BMI) was calculated for respondents in each neighborhood (Table 3). In adults, a BMI value of 18.5 to 24.9 is normal, a value of 25.0 to 29.9 is overweight, and a value of 30.0 or higher is obese. The average BMI value for respondents in the Detroit neighborhoods was 29.9, an overweight value bordering on obese. In contrast, the BMI values for respondents in the higher density suburbs (Ann Arbor and Birmingham) averaged 24.6, while for respondents in the low-density suburbs (Bloomfield Hills and West Bloomfield) the average BMI was 24.8. As one would expect given the neighborhood profiles, there was a racial dimension to BMI. While average BMI across all six neighborhoods for Whites was 24.8, the average BMI for non-Whites was 28.6. The survey results also revealed that Black respondents maintained the highest average BMI value at 29.4. Hence, while wealthy suburban respondents were in the normal weight category—regardless of whether they lived in compact or sprawling neighborhoods—the urban Detroit respondents were overweight, despite living in high-density and high-connectivity walkable neighborhoods.

Table 3. Average BMI and Average Time Spent on Moderate and Vigorous Exercise per Week (mins.) by Neighborhood
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
  1. *In adults a BMI value of 18.5 to 24.9 is normal, a value of 25.0 to 29.9 is overweight, and a value of 30.0 or higher is obese.

Average BMI by neighborhood.
Average BMI*30.1629.4624.3925.0125.2424.26
Average time spent on moderate and vigorous exercise by neighborhood per week (mins.).
Average per week (in mins.)71.0678.4981.6880.5575.8282.79

With regard to engagement in physical activity (both moderate and vigorous), respondents in the higher density suburbs spent most time per week exercising (an average of 81.2 min), respondents in low-density suburbs were second at 79.3 min, while respondents in the Detroit urban neighborhoods engaged in the least physical activity, 75.0 min. Table 3 shows the breakdown of total physical activity per week by neighborhood. In addition, White respondents, at 79.7 min per week, engaged in slightly more physical activity than non-White respondents, at 76.8 min per week. The assessment of accessibility and neighborhood travel patterns that follows provides added insight into distinctions in BMI values across the neighborhoods.

Accessibility, Travel, and Perceived Versus Shortest Time Route Distances

Travel behavior within the six neighborhoods was examined with analyses on travel mode, travel frequency, and travel distance by neighborhood and activity function. Before the survey findings are explored, however, a brief review on the calculation of distance, as recorded by survey participants and as used in the analyses, will be helpful. While each participant was asked to record the perceived distances between their home, work, leisure, shopping, and services, they were also asked to include the exact addresses of the end point locations for all the different activities. All the locations were found through site surveys and mapping programs, and then recorded. Surveys with insufficient destination data were removed from the study. Since all the surveys had codes to household addresses, the starting and end points of all trips were known. This allowed all travel distances to be referenced through Google maps, over 15,000 trips, based on the algorithm that determines the distance of the shortest time route between destinations. By explicitly controlling starting and end points, this approach to calculating distance is expected to reduce errors associated with self-reported perceived distances.

Mode of Travel: Walking/Cycling, Public Transit, and the Car

In examining mode of travel, the degree of automobile dependence in both the urban and suburban settings is immediately evident. Only 17.4% of trips, across all six neighborhoods and across all activities, are by walking or cycling. The reliance on mass transit is even lower than by nonmotorized means, with only 4.2% of overall trips by public transit. In general, however, the assessment of travel mode by neighborhood does reveal that higher densities, mixed land uses, and well-connected neighborhoods do encourage both travel by nonmotorized means and by mass transit (Tables 4-6).

Table 4. Percentage of Walking/Cycling Trips in Aggregate by Neighborhood and Travel Activity, With Frequencies on a Weekly Basis Provided in Brackets
       By
 WorkSchoolShoppingRestaurantServiceLeisureneighborhood
  1. *Percentage of walking/cycling for all six neighborhoods and for all travel purposes.

Detroit 113.7%7.8%26.1%16.6%12.8%18.6%17.8%
 (37/270)(5/65)(103/395)(39/232)(36/283)(46/247)(266/1491)
Detroit 217.0%29.2%29.4%31.5%20.5%25.5%25.6%
 (48/282)(24/81)(178/606)(82/259)(64/310)(79/308)(473/1845)
Ann Arbor22.2%26.1%14.0%31.3%24.1%45.7%27.4%
 (236/1066)(37/142)(129/917)(131/420)(138/574)(457/998)(1128/4117)
Birmingham9.4%22.2%12.3%15.2%12.9%21.3%14.0%
 (53/567)(4/18)(92/744)(47/312)(61/472)(112/524)(367/2636)
Bloomfield Hills4.4%26.0%7.6%6.3%6.3%10.3%7.5%
 (24/550)(16/62)(56/732)(20/319)(33/530)(55/537)(204/2728)
West Bloomfield12.1%55.0%3.3%5.4%3.8%18.8%9.1%
 (75/618)(11/20)(22/659)(17/304)(19/484)(92/492)(234/2576)
By activity14.1%25.0%14.3%18.2%13.2%27.1%17.4%*
 (473/3352)(97/386)(579/4052)(335/1845)(351/2651)(840/3106)(2674/15392)
Table 5. Percentage of Public Transit Trips in Aggregate by Neighborhood and Travel Activity, With Frequencies on a Weekly Basis Provided in Brackets
       By
 WorkSchoolShoppingRestaurantServiceLeisureneighborhood
  1. *Percentage of walking/cycling for all six neighborhoods and for all travel purposes.

Detroit 112.4%34.5%7.1%9.1%11.5%10.9%11.0%
 (34/270)(22/65)(28/395)(21/232)(33/283)(27/247)(164/1491)
Detroit 216.8%28.6%12.7%4.2%6.1%13.8%11.9%
 (48/282)(23/81)(77/606)(11/259)(19/310)(43/308)(220/1845)
Ann Arbor11.7%32.2%1.9%2.1%3.1%1.5%5.6%
 (125/1066)(46/142)(18/917)(9/420)(18/574)(15/998)(229/4117)
Birmingham0.7%0.0%0.5%0.3%0.2%0.2%0.4%
 (4/567)(0/18)(4/744)(1/312)(1/472)(1/524)(11/2636)
Bloomfield Hills1.1%8.1%0.2%0.6%0.2%0.4%0.6%
 (6/550)(5/62)(2/732)(2/319)(1/530)(2/537)(18/2728)
West Bloomfield0.0%0.0%0.0%0.7%0.0%0.8%0.2%
 (0/618)(0/20)(0/659)(2/304)(0/484)(4/492)(6/2576)
By activity6.4%24.8%3.2%2.5%2.7%2.9%4.2%*
 (216/3352)(96/386)(128/4052)(46/1845)(72/2651)(91/3106)(648/15392)
Table 6. Percentage of Driving Trips in Aggregate by Neighborhood and Travel Activity, With Frequencies on a Weekly Basis Provided in Brackets
       By
 WorkSchoolShoppingRestaurantServiceLeisureneighborhood
  1. *Percentage of walking/cycling for all six neighborhoods and for all travel purposes.

Detroit 173.9%57.8%66.8%74.3%75.7%70.5%71.1%
 (200/270)(37/65)(264/395)(172/232)(214/283)(174/247)(1061/1491)
Detroit 266.1%42.2%57.9%64.3%73.4%60.7%62.4%
 (187/282)(34/81)(351/606)(167/259)(227/310)(187/308)(1152/1845)
Ann Arbor66.2%41.7%84.0%66.6%72.8%52.8%67.0%
 (705/1066)(59/142)(771/917)(279/420)(418/574)(527/998)(2759/4117)
Birmingham89.9%77.8%87.1%84.5%86.9%78.5%85.6%
 (510/567)(14/18)(648/744)(263/312)(410/472)(412/524)(2256/2636)
Bloomfield Hills94.5%65.9%92.2%93.1%93.5%89.4%91.9%
 (520/550)(41/62)(675/732)(297/319)(495/530)(480/537)(2506/2728)
West Bloomfield87.9%45.0%96.7%93.9%96.2%80.4%90.7%
 (543/618)(9/20)(637/659)(286/304)(465/484)(395/492)(2335/2576)
By activity79.5%50.2%82.6%79.4%84.1%70.0%78.4%*
 (2663/3352)(194/386)(3345/4052)(1464/1845)(2229/2651)(2174/3106)(12070/15392)

In the two Detroit urban neighborhoods, over 22.1% of all travel was by walking and cycling and over 11.5% of all travel was by public transit. Similarly, in the two more compact suburbs, over 22.1% of all trips was by walking and cycling, while transit use was considerably lower, making up only 3.5% of all trips. In contrast, in the two low-density suburbs, less than 8.3% of all trips were by walking and cycling and less than 0.5% of all trips were by public transit. This translates into 66.3% of overall trips in urban Detroit by car, 74.3% of all trips in the higher density suburbs (Ann Arbor and Birmingham) by car, and 91.3% of overall trips in the low-density suburbs (Bloomfield Hills and West Bloomfield) by car.

While lower levels of automobile reliance in urban Detroit are influenced by the built environment, the survey reveals that socio-economic conditions also play a role in shaping travel. Lower income levels in urban Detroit are partly responsible for contributing to lower levels of automobile ownership (Table 7). While the average number of operating vehicles per household in the Detroit neighborhoods was 0.67, the suburbs maintained some 2.5 times more operating vehicles per household, averaging 1.71 vehicles in the higher density suburbs and 1.85 vehicles per household in the low-density suburbs.

Table 7. Number of Vehicles in Operating Condition by Household in Each Neighborhood
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
Cars per household0.70.61.71.81.81.9

Accessing Amenities: Exploring Distance and Travel Frequency

In analyzing accessibility, four types of amenities will be assessed—restaurants, leisure, personal services, and food shopping—through the exploration of distance and travel frequency.

Restaurants

When distances to restaurants are observed by neighborhood type, predictable patterns emerge (Tables 8 and 9). On average, higher density, mixed land use, and connected neighborhoods have better access to restaurants than low-density, single-use, and disconnected neighborhoods. While average distances to all restaurants were the least in the urban and higher density suburban neighborhoods, residents of the low-density suburbs confronted an average distance almost two times greater than the distances in the other four higher density neighborhoods. In terms of monthly frequency to restaurants, urban respondents ate at restaurants an average of 6.4 times per month, low-density suburban respondents frequented restaurants an average of 5.6 times, while higher density suburban respondents averaged 5.2 visits to restaurants on a monthly basis.

Table 8. Mean/Median Distance and Monthly Frequency of Trips to Restaurants by Neighborhood Type*
 UrbanHigh-density suburbLow-density suburb
  1. *Because of rounding-off, some values might be 0 at the individual neighborhood level, but they will affect average distance and trip frequency values at the neighborhood type level, where the two neighborhoods in each classification and their travel characteristics are consolidated.

Fast food chains   
Mean minimal distance (mi)2.0 mi2.7 mi3.8 mi
Median minimal distance (mi)1.2 mi2.2 mi2.8 mi
Number of trips per month5.71.22.3
Coffee Shops & Bakeries   
Mean minimal distance (mi)01.7 mi4.0 mi
Median minimal distance (mi)01.2 mi3.2 mi
Number of trips per month00.40.4
Upscale restaurants   
Mean minimal distance (mi)02.1 mi5.4 mi
Median minimal distance (mi)01.5 mi4.5 mi
Number of trips per month00.60.6
Casual chains   
Mean minimal distance (mi)10.8 mi3.7 mi5.1 mi
Median minimal distance (mi)10.0 mi2.7 mi4.6 mi
Number of trips per month0.30.50.8
Ethnic restaurants   
Mean minimal distance (mi)2.8 mi2.2 mi4.1 mi
Median minimal distance (mi)1.0 mi1.3 mi3.0 mi
Number of trips per month0.10.70.7
Other   
Mean minimal distance (mi)4.3 mi2.1 mi4.2 mi
Median minimal distance (mi)2.1 mi1.4 mi3.8 mi
Number of trips per month0.31.71.0
ALL RESTAURANTS   
Mean minimal distance to2.5 mi2.4 mi4.2 mi
ALL RESTAURANTS (mi)   
Median minimal distance to1.3 mi1.6 mi3.2 mi
ALL RESTAURANTS (mi)   
Number of trips per month6.45.25.6
Percent of trips to ALL24.6%24.3%6.0%
RESTAURANTS by walking   
Percent of trips to ALL7.0%1.4%0.6%
RESTAURANTS by transit   
Table 9. Mean/Median Distance and Monthly Frequency of Trips to Restaurants by Neighborhood
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
Fast food chains      
Mean minimal distance (mi)2.4 mi1.6 mi2.3 mi3.0 mi3.5 mi4.0 mi
Median minimal distance (mi)1.6 mi1.0 mi2.2 mi2.3 mi2.7 mi3.0 mi
Number of trips per month6.15.41.01.62.32.3
Coffee shops & bakeries      
Mean minimal distance (mi)001.6 mi2.1 mi4.6 mi3.0 mi
Median minimal distance (mi)000.9 mi2.4 mi3.1 mi3.7 mi
Number of trips per month000.50.20.40.3
Upscale restaurants      
Mean minimal distance (mi)001.5 mi2.5 mi5.9 mi4.5 mi
Median minimal distance (mi)001.2 mi2.0 mi4.0 mi3.6 mi
Number of trips per month000.41.00.70.5
Casual chains      
Mean minimal distance (mi)12.29.1 mi4.2 mi2.8 mi6.8 mi4.2 mi
Median minimal distance (mi)11.259.4 mi3.1 mi1.5 mi6.0 mi3.5 mi
Number of trips per month0.40.30.50.50.51.1
Ethnic restaurants      
Mean minimal distance (mi)02.7 mi2.0 mi3.1 mi5.4 mi3.2 mi
Median minimal distance (mi)00.8 mi1.2 mi1.8 mi4.9 mi2.6 mi
Number of trips per month00.20.90.50.50.8
Other      
Mean minimal distance (mi)04.3 mi1.3 mi3.3 mi4.7 mi3.4 mi
Median minimal distance (mi)01.7 mi1.1 mi2.7 mi4.5 mi2.2 mi
Number of trips per month00.51.71.81.20.7
ALL RESTAURANTS      
Mean minimal distance to3.0 mi2.2 mi2.0 mi3.0 mi4.6 mi3.9 mi
ALL RESTAURANTS (mi)      
Median minimal distance to1.7 mi1.0 mi1.4 mi2.0 mi3.6 mi3.0 mi
ALL RESTAURANTS (mi)      
Number of trips per month6.66.35.05.55.65.7
Percent of trips to18.6%25.5%31.3%12.3%6.3%5.4%
ALL RESTAURANTS by walking      
Percent of trips to9.1%4.2%2.1%0.3%0.6%0.7%
ALL RESTAURANTS by transit      

However, once trips are disaggregated by restaurant type, some important distinctions emerge by neighborhood in the access and the frequency of visits. While major fast food chains (McDonald's, Burger King, and KFC) were located in the east side Detroit neighborhoods, many other restaurant types—healthier restaurant options—were located primarily in the vicinity of the wealthier suburbs, and were generally not frequented by the urban respondents. While urban residents traveled an average of 2.0 miles to fast food restaurants, higher density suburban residents traveled an average of 2.7 miles, and low-density suburban residents traveled an average of 3.8 miles to fast food establishments. Urban residents also ate at fast food chains most often—5.7 times per month versus 1.2 times per month for residents living in higher density suburbs, and 2.3 times per month for residents living in low-density suburbs. Some 90% of all dining out experiences by urban respondents were in fast food restaurants, compared to 24% by higher density suburban respondents and 41% for lower density suburban respondents.

Prior research has shown that higher obesity rates are prevalent among populations that have a higher reliance on fast food outlets, due to the higher fat content of food served in these restaurants (Block, Scribner, & DeSalvo, 2004; Maddock, 2004; Powell, Chaloupka, & Bao, 2007b). The Detroit research, however, shows that due to location decisions by restaurant owners, non–fast food restaurants are not locating close to the urban neighborhoods. Healthier food sources are simply not that accessible to the urban poor, despite the fact that they live in built environments that should improve access. Because of the greater distances Detroit residents have to travel to reach these destinations, they are not as reliant on healthy restaurant options as the suburban wealthy. In addition, the poor access to healthy restaurants is likely a variable that plays an important role in the higher BMI values among Detroit respondents, despite the built environment characteristics of their neighborhoods.

This analysis of access to healthier restaurant options shows that characteristics in the urban built environment represent only one element in defining accessibility. High densities, mixed land uses, and connectivity are important built environment attributes to shortening distances between destinations and encouraging nonmotorized travel, but the equitable spatial investment in urban amenities is also vital in shaping access. If amenities are not located in a neighborhood, built environment characteristics can do little in reducing distances between destinations. Urban disinvestment, resulting in the absence of amenities, thus emerges as a variable that shapes both travel behavior (by limiting available destinations) and diet (by limiting access to food options). This is clearly evident in east side Detroit, where the residents, despite living in neighborhoods characterized by higher densities, mixed land uses, and high connectivity, have to travel the greatest distances to access healthy restaurant options.

Leisure and Personal Services

In comparing the wealthier suburbs, consistent with the traditional literature on urban form and travel, when it comes to accessing leisure activities (theaters, parks, skating rinks, etc.) and personal services (doctors, banks, dry cleaners, etc.), it is evident that the higher density, higher connectivity, mixed land use neighborhoods confront lower distances to these destinations than the low-density, low-connectivity, single-use neighborhoods (Tables 10 and 11). Ann Arbor in particular, with its robust activity center, maintains a high concentration of leisure activities and personal services surrounded by a high concentration of residential dwellings, ensuring high accessibility. Ann Arbor residents maintained the shortest distances to leisure and personal services and it is reflected in some of the highest rates of walking/biking to these activities. Some 46% of all trips to leisure destinations in Ann Arbor were by walking and biking, and about 38% of all trips to personal services and leisure activities combined were by nonmotorized means. In contrast, in the low-density suburbs, distances to leisure and personal services increase, as does automobile dependence. Average distances to leisure and personal services in the low-density suburbs are approximately double the distances in Ann Arbor, and it results in less than 10% of all trips to these destinations being by nonmotorized means.

Table 10. Mean/Median distance and Monthly Frequency of Trips to Leisure Activities and Personal Services by Neighborhood Type
 UrbanHigh-density suburbLow-density suburb
Leisure activities   
Mean minimal distance (mi)6.4 mi5.4 mi11.6 mi
Median minimal distance (mi)5.0 mi1.5 mi5.0 mi
Number of trips per month4.37.06.9
Percent of trips to leisure activities by walking (%)22.5%37.4%14.3%
Percent of trips to leisure activities by transit (%)12.6%1.1%0.1%
Personal services   
Mean minimal distance (mi)3.4 mi2.1 mi3.5 mi
Median minimal distance (mi)2.0 mi1.4 mi2.1 mi
Number of trips per month6.75.78.7
Percent of trips to personal services by walking (%)16.9%19.0%5.1%
Percent of trips to personal services by transit (%)8.7%1.8%0.1%
Table 11. Mean/Median Distance and Monthly Frequency of Trips to Leisure Activities and Personal Services by Neighborhood
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
Leisure activities      
Mean minimal distance (mi)5.8 mi6.9 mi4.6 mi6.6 mi12.7 mi10.4 mi
Median minimal distance (mi)4.9 mi5.0 mi1.4 mi1.7 mi4.6 mi6.4 mi
Number of trips per month3.94.77.56.27.06.8
Percent of trips to leisure activities by walking18.6%25.5%45.7%21.3%10.3%18.8%
Percent of trips to leisure activities by transit10.9%13.8%1.5%0.2%0.4%0.8%
Personal services      
Mean minimal distance (mi)4.2 mi2.7 mi1.9 mi2.3 mi3.6 mi3.5 mi
Median minimal distance (mi)2.8 mi1.4 mi1.3 mi1.7 mi1.8 mi2.4 mi
Number of trips per month6.37.13.87.18.68.8
Percent of trips to personal services by walking (%)12.8%20.5%24.1%12.9%6.3%3.8%
Percent of trips to personal services by transit (%)11.5%6.1%3.1%0.2%0.2%0%

In the case of leisure and personal services, urban Detroit neighborhoods again illustrate the importance of socioeconomic variables in defining accessibility. Tables 10 and 11 show that access to leisure and personal services in the Detroit neighborhoods falls somewhere between the high-density and low-density suburbs, despite the higher densities, mixed land uses, and increased connectivity in urban Detroit. In reaching leisure activities, while urban mean distances for residents of the Detroit neighborhoods approach the high density suburbs, the median distances are similar to those in the low-density suburbs. For personal services, urban Detroit neighborhoods maintain accessibility characteristics, in terms of mean and median distances, that are similar to those in the low-density suburbs, despite the higher densities, mixed land uses, and high-connectivity characteristics of the urban neighborhoods. This lack of access to leisure and personal services in east side Detroit—including doctors, pharmacists, dentists, banks, theaters, and parks—is not the result of built environment characteristics, but rather the result of urban disinvestment and the resulting lack of amenities within the Detroit neighborhoods, demonstrating again that socioeconomic conditions can thwart the advantages of urban form.

Grocery Outlets

As recognized in the national media coverage, accessing food in Detroit is an issue that has been dramatically affected by urban disinvestment. However, there is an important dimension of shopping behavior in Detroit that contradicts the basic assumptions behind travel behavior and much of the food desert literature. For east side residents, despite living in urban environments characterized by higher densities, mixed land uses, and high connectivity, access to food shopping (when examining all store types) is far poorer than the access enjoyed by high-density suburban residents, who live in similar built environments (see Table 12). Lower-income urban respondents are particularly burdened with great distances to national/regional supermarket chains (such as Meijer and Kroger), which existing literature has already shown to have the greatest healthy food options at the highest quality, and at the lowest prices. Detroit neighborhood respondents travel more than 5.5 miles (mean distance) to reach major supermarket chains. This is more than twice the average distance travelled by respondents living in the four suburban neighborhoods, whose average distance to a major supermarket chain is about 2.6 miles. This greater distance resulted in a much lower reliance on national/regional supermarket chains by east side Detroit respondents when compared to respondents living in the higher and lower density suburbs (Table 13). As a result, low-density suburban respondents visited major supermarket chains more than twice as many times per month as the Detroit respondents.

Table 12. Mean/Median Distance and Monthly Frequency of Trips to Grocery Outlets by Neighborhood Type
 UrbanHigh-density suburbLow-density suburb
National/regional supermarkets   
Mean minimal distance (mi)5.52.23.1
Median minimal distance (mi)4.61.72.7
Number of trips per month5.07.610.6
Convenience stores   
Mean minimal distance (mi)1.700
Median minimal distance (mi)1.000
Number of trips per month1.200
Boutique grocery stores   
Mean minimal distance (mi)01.94.9
Median minimal distance (mi)01.74.4
Number of trips per month03.82.0
Pharmacies   
Mean minimal distance (mi)2.101.8
Median minimal distance (mi)1.401.1
Number of trips per month0.600.1
Farmer's markets   
Mean minimal distance (mi)3.61.30
Median minimal distance (mi)2.71.30
Number of trips per month0.30.80
Independent supermarkets   
Mean minimal distance (mi)1.92.03.4
Median minimal distance (mi)1.41.73.4
Number of trips per month5.50.70.6
Other stores   
Mean minimal distance (mi)1.82.90
Median minimal distance (mi)1.31.80
Number of trips per month0.80.10
ALL STORES   
Mean minimal distance (mi)3.32.03.4
Median minimal distance (mi)1.61.62.9
Number of trips per month13.513.013.3
Percent of trips by walking28.1%13.3%5.6%
Percent of trips by transit10.5%1.3%0%
Table 13. Mean/Median Distance and Monthly Frequency of Trips to Grocery Outlets by Neighborhood
   Ann BloomfieldWest
 Detroit 1Detroit 2ArborBirminghamHillsBloomfield
National/regional supermarkets      
Mean minimal distance (mi)7.74.52.41.93.13.0
Median minimal distance (mi)6.43.51.91.32.62.8
Number of trips per month3.76.16.78.910.310.8
Convenience stores      
Mean minimal distance (mi)1.51.800.700
Median minimal distance (mi)0.91.100.700
Number of trips per month0.71.700.100
Boutique grocery stores      
Mean minimal distance (mi)001.92.05.34.2
Median minimal distance (mi)001.41.98.83.5
Number of trips per month003.54.42.31.7
Pharmacies      
Mean minimal distance (mi)2.11.9000.82.5
Median minimal distance (mi)1.41.1000.72.7
Number of trips per month1.20.1000.10.1
Farmer's markets      
Mean minimal distance (mi)2.85.21.3000
Median minimal distance (mi)2.75.21.3000
Number of trips per month0.40.21.3000
Independent supermarkets      
Mean minimal distance (mi)2.41.71.22.13.43.7
Median minimal distance (mi)1.41.41.41.83.44.5
Number of trips per month4.56.30.21.61.10.1
Other stores      
Mean minimal distance (mi)1.72.21.06.000
Median minimal distance (mi)1.22.50.66.000
Number of trips per month1.30.40.10.100
ALL STORES      
Mean minimal distance (mi)3.92.92.12.03.53.2
Median minimal distance (mi)2.01.51.71.52.83.0
Number of trips per month11.814.911.814.913.912.7
Percent of trips by walking26.1%29.4%14.0%12.3%7.6%3.3%
Percent of trips by transit7.1%12.7%1.9%0.5%0.2%0%

Another measure of neighborhood accessibility involves path distance calculations of minimal in-network distances to the closest major supermarket chain, involving every house in each neighborhood. Major discount supermarkets were included in this analysis, such as Aldi's, which is located in the Detroit 2 neighborhood. The average in-network distance for all households by neighborhood to the closest major supermarket illustrates again the lower accessibility to supermarkets among Detroit residents (see Figure 8). While the shortest average mean distance to a major supermarket for the Detroit residents was 2.03 miles, the shortest average distance was only 1.55 miles for low-density suburban residents (Bloomfield Hills and West Bloomfield), and 1.09 miles for high-density suburban residents (Ann Arbor and Birmingham). The Detroit residents, despite living in built environments where higher quality food sources are traditionally thought to be more accessible (because of density, land use mix, and connectivity), actually lived furthest from major supermarket chains and were less reliant on these stores for shopping than even respondents living in low-density, disconnected suburbs. Existing literature has shown that major supermarket chains tend to follow wealthy residents into suburban locations. In addition to following purchasing power, the suburbanization of supermarkets is encouraged by perceptions and realities of crime—and lower insurance premiums—in the suburbs compared to what are perceived as riskier inner city locations (Bromley & Thomas, 1993; Pothukuchi, 2005; Pothukuchi, Mohamed, & Gebben, 2008; Teaford, 2006).

image

Figure 8. Average In-Network Distance to Closest Supermarket Chain by Neighborhood

Download figure to PowerPoint

As much of the food deserts literature recognizes, while the Detroit respondents are most distant from major supermarket chains, they are surrounded by and maintain high accessibility to convenience and party stores. As a result, urban respondents shop at convenience stores more frequently than suburban respondents, who essentially have no reliance on convenience stores for food (Table 12). However, unlike what is assumed by most of the food deserts literature, Detroit residents are not very reliant on convenience stores for their shopping either. East side Detroit respondents shopped at convenience stores for food an average of only 1.2 times per month. Farmer's markets were also largely irrelevant as a food source in east side Detroit, with only some 0.3 trips per respondent per month being made to a farmer's market to purchase food.

While the media effectively captured the loss of Detroit's major supermarket chains, it is important to recognize that there has been an exaggeration of the loss of food sources in the city. Independent supermarkets in Detroit have been almost exclusively overlooked in the national and international media coverage. In fact, for Detroit urban residents, the most important stores for grocery shopping are local independent supermarkets (such as Public Foods and Food Town Supermarket; see Figure 9). While Detroit urban respondents shopped at smaller independent supermarkets an average of over 5.5 times per month, for suburban respondents independent supermarkets were largely irrelevant for shopping, accounting for less than one visit per month.

image

Figure 9. Photos of Independent Supermarkets in East Side Detroit Neighborhoods

Download figure to PowerPoint

For Detroit respondents, 41% of all grocery shopping trips were to independent supermarkets, while only some 5% of shopping trips were to independent supermarkets for suburban respondents. In contrast, while some 37% of all shopping trips for groceries were to major supermarket chains for Detroit respondents, about 70% of all shopping trips were to major supermarket chains for suburban respondents. The second most important shopping destination in the suburbs was boutique grocery stores, such as upscale delis and health food stores.

Much of the food deserts literature does not acknowledge this distinction between where residents shop and the location of the closest store. There are few studies that analyze where people actually shop for food, and, as this research has shown, the closest store is not necessarily the store of choice.

The importance of preference is also seen in the shopping behavior at major supermarkets and it is in part reflected in the average distances shown in Tables 12 and 13 and Figure 8. From Figure 8, it is evident that while the shortest average distance to a major supermarket chain for the Detroit east side residents was about 2.0 miles, the east side residents actually travelled some 5.5 miles on average to access their supermarket of preference.

This analysis illustrates that neighborhoods with built environments traditionally viewed as ensuring increased accessibility—neighborhoods with higher densities, mixed land use, and greater connectivity—do not guarantee shorter distances. An equitable spatial investment in urban amenities will also be influential in shaping access. Without an equitable spatial investment in amenities, as evident with major national/regional supermarket chains in Detroit, certain destinations will remain distant from some population subgroups.

In addition, it is not just that Detroit urban respondents traveled a greater average distance to reach major national supermarkets when compared to the low-density suburban respondents. Respondents from the higher density urban neighborhoods and respondents from the low-density suburbs traveled similar average distances to reach “all shopping destinations” when comparing mean values. This illustrates again the significance of disinvestment and the resulting lack of access to amenities in declining inner cities in shaping travel.

This analysis also shows that, merely because stores are accessible to residents, it does not mean that these residents will shop at the closest location. The east side Detroit respondents bypassed closer convenience stores and completed over 78% of their shopping trips to supermarkets, both independent and major chains. The urban residents did confront greater costs in this process, both monetary and temporal, and this reveals the ongoing “burdens of place” of neighborhoods experiencing urban disinvestment. Nevertheless, east side Detroit residents were largely shopping in stores that offered healthy food options.

COMMENTARY AND CONCLUSIONS

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

This neighborhood-scale Detroit area study has advanced a number of important theoretical contributions. One contribution is enabled by the focus of this research specifically on declining neighborhoods. With regard to the built environment and its accessibility characteristics, while higher densities, mixed land uses, and greater connectivity do improve access in general, the research reveals that this relationship is stronger in communities that are wealthy. The relationship between compact, mixed-use developments and improved access is not as strong—and can break down completely—in neighborhoods experiencing disinvestment. In declining neighborhoods, the exodus of amenities limits residents’ proximity to daily destinations, despite the fact that these neighborhoods maintain a built environment that should improve access. Class is an important factor in defining location for retail and other commercial activities. It can be argued that, in the Detroit region, class plays an even more important role than the built environment in shaping access and travel.

Thus a fundamental contribution of this research has been a challenge to the traditional understanding of the built environment and accessibility. Traditional urban form, access, and travel behavior relationships are not necessarily replicable in poor and declining neighborhoods. Without an equitable spatial investment in amenities, such as healthy restaurants or personal services, certain goods and services will remain distant from particular populations, regardless of urban form. More broadly, this should raise concern that despite our extensive understanding of the built environment, analysts are lacking the basic knowledge of particular types of urban form, and specifically ones associated with communities in decline. Without a more comprehensive knowledge of the condition of the physical structure of disadvantaged communities, urban planners will not be in a position to address many, if not most, of today's critical urban stresses.

The importance of devoting more attention to research on the disadvantaged is also accentuated by this study's findings that the built environment and human behaviors have a separate and distinct set of parameters that are influenced by not only class but also culture. In order to fully comprehend the condition of the city, planners and designers need to have a better grasp of local cultures and resulting behaviors—an area of interest that has been losing ground in recent decades. This is particularly important in the context of the structure of communities in decline. There is a critical need to focus on marginalized communities and the outcomes associated with the physical form of their environments across a range of U.S. cities.

Indeed, it is this lack of research that has led to the growing call to pay more attention to the condition of lower income populations and their neighborhoods in urban planning and design (Day, 2003, 2006; USDHHS, 2000; Vojnovic, 2006; Vojnovic, Jackson-Elmoore, Summers, & Bruch, 2006). While the number of disadvantaged has been on the rise, planners and designers have been increasingly uninterested in the condition of marginalized neighborhoods. Concern for the poor and minorities was a crucial aspect of urban research and interest during the 1960s and 1970s, but the focus on the disadvantaged has been diminishing since the 1980s (Slater, 2006; Podagrosi & Vojnovic, 2008; Podagrosi, Vojnovic, & Pigozzi, 2011).

The distinction made in this study between declining higher density urban neighborhoods and wealthy suburban counterparts—low density vs. high density—reveals findings that challenge another line of existing literature. The Detroit results contradict studies that have drawn associations between low urban densities and high BMI values (Ewing et al., 2003; Kelly-Schwartz et al., 2004; Lopez, 2004). The Detroit research shows that urban respondents maintain the highest BMI values, bordering on obese, despite living in higher density neighborhoods. The Bloomfield Hills and West Bloomfield respondents, living in low-density, single-use, low-connectivity neighborhoods, fell into the normal weight category.

Most of the research on urban form and obesity covers large geographic areas, including counties or larger spatial scales. Given their area of coverage, these studies average out unique distinctions evident among different socioeconomic and ethnic populations. This Detroit region study, by concentrating on neighborhood-scale comparisons between declining and prospering neighborhoods, is able to recognize very different outcomes from patterns evident across larger scale data. However, while contradicting some of the literature on the relationship between densities and BMI values, the findings of this study are consistent with two other streams of research. Our findings correspond to existing U.S. research that shows the highest prevalence of obesity among the poor (Morland et al., 2006; Wang et al., 2007). In addition, the Detroit research is also consistent with health studies that have focused specifically on minority populations and the poor in high-density urban neighborhoods, where health outcomes (including higher BMI values) are closely linked to poverty and minority populations, regardless of built environment and design characteristics (Krieger, 2000; Scott et al., 2009; Williams, 2005; Vojnovic et al., 2013). The scale of study and distinguishing by population subgroups thus emerge as critical in linking BMI values with characteristics of the urban built environment.

Ultimately, this discussion of the relationship between obesity and urban form is closely linked to the main thesis of this study. Given the consistent relationship between obesity and poverty across the United States, the urban form and obesity question illustrates, once again, that socioeconomic variables can outweigh the importance of the built environment in shaping outcomes. It is not only with regard to accessibility that class will be more relevant than the built environment, but when it comes to public health, as evident with BMI across the Detroit region, socioeconomic variables will likely be more influential than urban form in shaping outcomes.

The importance of socioeconomic and cultural influences on public health is also evident in values placed on health. Despite living in automobile-oriented neighborhoods, low-density suburban respondents maintained normal BMI values, marginally higher than the BMI values of respondents living in high-density suburbs. In contrast, despite living in urban environments that promoted walking, Detroit urban respondents were in the overweight category, bordering on obese. It is likely that upper-income groups are more health-conscious and are more aware of their dietary intake, an issue linked to class and cultural variables such as income, education, social norms, and peer pressure. The distinction in frequenting fast food restaurants between Detroit urban and suburban respondents also is likely, in part, a reflection of class differences.

In order to get a sense of the value placed on health by neighborhood, information was collected on smoking. Consistent with prior research, the highest prevalence of smoking was in lower-income neighborhoods (Jha, Ranson, Nguyen, & Yach, 2002; Avendano, Glymour, Banks, & Mackenbach, 2009). In urban Detroit, over 35% of the respondents smoke, while less than 8% of the suburban respondents smoke. There is also a racial dimension to smoking, with only 8% of White subjects smoking compared to 27% of non-White subjects. These distinctions in smoking and frequenting fast food restaurants between the urban and suburban respondents do, in part, reflect the different values placed on health across the Detroit urban region, and also, in part, reveal the influence of socioeconomic and cultural variables on health outcomes in marginalized communities.

Another important finding in this study is the role of preferences in influencing shopping, a variable seldom considered in the food deserts literature. It is generally assumed that residents who live in food deserts shop in food deserts. The recent media coverage on food supply in Detroit has further contributed to this misapprehension. In fact Detroit residents, despite being surrounded by a concentration of convenience stores, rarely shopped for food at these locations, averaging only about one shopping trip per month. Some 78% of food shopping by east side residents was at independent and major supermarkets. However, one should not necessarily expect that this shopping behavior will be replicated in other U.S., or even Michigan cities. In a recent Lansing, Michigan study, Vojnovic and colleagues (2013) showed that lower-income Lansing respondents relied on convenience stores for over 35% of food shopping trips, over 3.5 times more on average than the Detroit respondents. This emphasizes, again, the need for more neighborhood-scale access and travel behavior analyses in communities experiencing disinvestment and decline across the United States.

As for the ability of east side Detroit residents to access major supermarket chains, urban respondents traveled the longer necessary distances to reach the national/regional chains, some 5.5 miles on average. Unsurprisingly, the Detroit respondents that tended to shop at these stores maintained higher relative incomes and owned a car. However, informal car-pooling services to these major supermarket chains were regularly reported within the east side neighborhoods.

The research also shows that what emerges as a relevant dietary-related health risk in Detroit is the frequenting of fast food restaurants, with some 90% of dining out by the east side Detroit respondents being in fast-food establishments. As prior studies have shown, populations that rely on the cheap but highly processed and high–fat content meals at fast food restaurants maintain higher obesity rates (Maddock, 2004; Powell et al., 2007b). It is also apparent in the Detroit context that healthier restaurant options—as in the case of major supermarket chains—are not that accessible.

By analyzing places shopped as opposed to simply the concentration of store types within neighborhoods, this study allows for an assessment of actual shopping behavior. The results not only problematize recent media reports on food access in Detroit but also much of the food deserts literature. East side residents regularly shop at major suburban supermarket chains. Independent Detroit supermarkets also emerge as important nutritious food suppliers for urban residents. This research, in fact, exposes the critical role of independent outlets in declining neighborhoods as suppliers of healthy food. Studies that fail to take account of independent supermarkets in their research will likely overestimate the accessibility constraints imposed on residents living in neighborhoods experiencing disinvestment.

This is not to say that access to urban amenities is not an issue in Detroit. The burdens faced by Detroit residents to reach basic amenities are severe. East side Detroit residents confront much greater costs—monetary and temporal—in reaching major supermarket chains and other amenities when compared to wealthier suburban residents, despite the built environment that they live in, which should promote access. There is a clear burden of place associated with living in east side Detroit. Unlike what is generally assumed, however, Detroit respondents in this study were conscientious about making smarter shopping decisions, and they actively and regularly incurred the higher costs of shopping at the healthier (and in many cases more distant) store locations. It is also clear, however, that the same cannot be said for the frequency of dining out in fast food chains among Detroit residents. An important contribution of this research has thus been to provide a detailed illustration of how a disadvantaged community has tried to adjust to its extreme decline and the loss of urban amenities in its surrounding environment. In the process, the analysis has altered our traditional theoretical understanding of urban form, accessibility, travel, and shopping behavior.

With regard to the impact of these research findings on policy, recent studies showing higher obesity rates among residents of low-density urban environments have renewed interest among policymakers on developing/redeveloping the built environment to improve accessibility and encourage nonmotorized travel. For certain populations and neighborhoods, such initiatives can contribute to healthier lifestyles. The limited access to urban amenities in Detroit's declining core, however, shows that characteristics in the built environment are only one element in defining access, travel, and public health among the disadvantaged. An equitable spatial investment in urban amenities will also be vital in shaping accessibility, travel, and health. If necessary amenities are not located in a neighborhood, built environment characteristics can do little to promote access to healthy food options or other needs. In neighborhoods characterized by disinvestment and decline, providing incentives to bring back specific amenities should be considered critical in addressing the obesity epidemic.

Ultimately, the wealthy, living in low-density suburbs, took advantage of their many opportunities (for instance, by ensuring a healthier diet and exercise) in countering the disadvantages of their suburban environment, and this was reflected in an overall healthy body weight. In contrast, Detroit residents, while living in a built environment that encourages walking, were dramatically affected by the inequitable spatial investment in amenities. Research that continues to show that obesity in the United States is most prevalent among the poor further necessitates policy focused on disadvantaged communities. It is therefore essential to reverse the trend, mentioned earlier, of growing disinterest in the condition of the poor.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies

The authors are very grateful to Dr. Laura Reese and the anonymous reviewers for their comments and criticisms that helped improve the quality of the article. We would like to thank the U.S. National Science Foundation that has funded this research under the Human and Social Dynamics program grant SES 0624263. We would also like to thank Michigan State University's Obesity Interest Group for providing supplemental funding for this study. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF or MSU's Obesity Interest Group. We are extremely grateful to the assistance of the Governor's Council on physical fitness, health and sports in Michigan, and particularly Ms. Marilyn Lieber (President and CEO) and Ms. Wilkerson (project officer with Active Living By Design of North Carolina's Gillings School of Global Public Health). We would also like to acknowledge our community partners, the Michigan Suburbs Alliance, U-SNAP-BAC and Messiah Housing Corporation for their support of this project. Finally, the support of Ms. Julie Brixie (currently Meridian Township treasurer) is also greatly appreciated. Ms. Brixie's ongoing interest and backing of projects carried out by faculty and students in the Department of Geography at MSU are indispensable to our programs.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies
  • Avendano, M., Glymour, M. M., Banks, J., & Mackenbach, J. P. (2009). Health disadvantage in US adults aged 50 to 74 years: A comparison of the health of rich and poor Americans with that of Europeans. American Journal of Public Health, 99, 540548.
  • Azuma, A. M., Gilliland, S., Vallianatos, M., & Gottlieb, R. (2010). Food access, availability, and affordability in 3 Los Angeles communities, Project CAFE, 2004–2006. Preventing Chronic Disease, 7, 19.
  • Baran, P. K., Rodríguez, D. A., & Khattak, A. J. (2008). Space syntax and walking in a New urbanist and suburban neighborhoods. Journal of Urban Design, 13(1), 528.
  • Block, J. P., Scribner, R. A., & DeSalvo, K. B. (2004). Fast food, race/ethnicity, and income: A geographic analysis. American Journal of Preventative Medicine, 27, 211217.
  • Boarnet, M. (2011). A broader context for land use and travel behavior, and a research agenda. Journal of the American Planning Association, 77(3), 197213.
  • Boarnet, M., Joh, K., Siembab, W., Fulton, W., & Nguyen, M.T. (2011). Retrofitting the suburbs to increase walking: Evidence from a land-use-travel study. Urban Studies, 48, 129159.
  • Bromley, R. D. F., & Thomas, C. J. (1993). The retail revolution, the careless shopper and disadvantage. Transactions of the Institute of British Geographers, 18, 222236.
  • Cao, X. (2010). Exploring causal effects of neighborhood type on walking behavior using stratification on the propensity score. Environment and Planning A, 42, 487504.
  • Cao, X., Mokhtarian, P. L., & Handy, S. L. (2007a). Cross-sectional and quasi-panel explorations of the connection between the built environment and auto ownership. Environment and Planning A, 39, 830847.
  • Cao, X., Mokhtarian, P. L., & Handy, S. L. (2007b). Do changes in neighborhood characteristics lead to changes in travel behavior? A structural equations modeling approach. Transportation, 34, 535556.
  • Cao X., Mokhtarian, P. L., & Handy, S. L. (2009). Examining the impacts of residential self-selection on travel behavior: A focus on empirical findings. Transport Reviews, 29, 359395.
  • Cervero, R. (1998). The transit metropolis. Washington, DC: Island Press.
  • Chatman, D. G. (2009). Residential choice, the built environment, and nonwork travel: Evidence using new data and methods. Environment and Planning A, 41, 10721089.
  • Clifton, K. J. (2004). Mobility strategies and food shopping for low-income families. Journal of Planning Education and Research, 23, 402413.
  • Day, K. (2003). New urbanism and the challenges of designing for diversity. Journal of Planning Education and Research, 23, 8395.
  • Day, K. (2006). Active living and social justice: Planning for physical activity in low-income Black and Latino communities. Journal of the American Planning Association, 72, 8899.
  • Ewing, R. (1997). Transportation and land use innovations. Chicago: American Planners Association.
  • Ewing, R., Schmid, T., Killingsworth, R., Zlot, A., & Raudenbush, S. (2003). Relationship between urban sprawl and physical activity, obesity, and morbidty. American Journal of Health Promotion, 18, 4757.
  • Ewing, R., & Handy, S. (2009). Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban Design, 14, 6584.
  • Ewing, R., & Cervero, R. (2010). Travel and the built environment. Journal of the American Planning Association, 76, 265294.
  • Franco, M., Diez Roux, A. V., Glass, T. A., Caballero, B., & Brancati, F. L. (2008). Neighborhood characteristics and availability of healthy foods in Baltimore. American Journal of Preventive Medicine, 35, 561567.
  • Frank, L., & Pivo, G. (1995). Impacts of mixed use and density on utilization of three modes of travel: Single-occupant vehicle, transit, and walking. Transportation Research Record, 1466, 4452.
  • Frank, L., Schmid, T. L., Sallis, J. F., Chapman, J. E., & Saelens, B. E. (2005). Linking objective physical activity data with objective measures of urban form. American Journal of Preventive Medicine, 28, 117125.
  • Galvez, M. P., Morland, K., Raines, C., Kobil, J., Siskind, J., Godbold, J., & Brenner, B. (2008). Race and food store availability in an inner-city neighbourhood. Public Health Nutrition, 11, 624631.
  • Gittelsohn, J., Franceschini, M. C. T., Rasooly, I. R., Ries, A. V., Ho, L. S., Pavlovich, W., Santos, V. T., Jennings, S. M., & Frick, K. D. (2007). Understanding the food environment in a low-income urban setting: Implications for food store interventions. Journal of Hunger and Environmental Nutrition, 2, 3350.
  • Greenslade, R. (2009). Time magazine sets up home in Detroit to report on depressed city. The Guardian, September 27. http://www.guardian.co.uk/media/greenslade/2009/sep/22/time-magazine-useconomicgrowth
  • Griffith, D., Vojnovic, I., & Messina, J. (2012). Distances in residential space: Implications from estimated metric functions for minimum path distances. GIScience & Remote Sensing, 48, 130.
  • Grossman, A. (2009). Retailers head for exits in Detroit. Wall Street Journal, June 16, A3.
  • Groves, R. M., & Couper, M. P. (1998). Nonresponse in Household Interview Surveys. New York: John Wiley & Sons, Inc.
  • Handy, S. (1996). Urban form and pedestrian choices: Study of Austin neighborhoods. Transportation Research Record, 1552, 135144.
  • Hansen, C. (2010). America now: City of heartbreak and hope. Dateline NBC. April 20. http://www.msnbc.msn.com/id/36665950/ns/dateline_nbc-the_hansen_files_with_chris_hansen/
  • Hargreaves, S. (2009). Hunger hits Detroit's middle class. CNN, August 13. http://money.cnn.com/2009/08/06/news/economy/detroit_food/
  • Howard, P. H., & Fulfrost, B. (2007). The density of retail food outlets in the central coast region of California: Associations with income and Latino ethnic composition. Journal of Hunger and Environmental Nutrition, 2, 318.
  • Inagami, S., Cohen, D. A., Finch, B. K., & Asch, S. M. (2006). You are where you shop: Grocery store locations, weight, and neighborhoods. American Journal of Preventive Medicine, 31, 1017.
  • Jha, P., Ranson, M. K., Nguyen, S. N., & Yach, D. (2002). Estimates of global and regional smoking prevalence in 1995, by age and sex. American Journal of Public Health, 92, 10021006.
  • Kelly-Schwartz, A. C., Stockard, J., Doyle, S., & Schlossberg, M. (2004). Is sprawl unhealthy? A multilevel analysis of the relationship of metropolitan sprawl to the health of individuals. Journal of Planning Education and Research, 24, 184196.
  • Krieger, N. (2000). Discrimination and health. In Berkman L. & Kawachi, I. (Eds). Social Epidemiology (pp. 3675). Oxford: Oxford University Press.
  • LeDoux, T., & Vojnovic, I. (2013). Going outside the neighborhood: The shopping patterns and adaptations of disadvantaged consumers living in the lower east side neighborhoods of Detroit, Michigan. Health & Place, 19, 114.
  • Lee, C., & Moudon, A. V. (2004). Physical activity and environment research in the health field: Implications for urban and transportation planning practice and research. Journal of Planning Literature, 19, 147181.
  • Lopez, R. (2004). Urban sprawl and risk for being overweight or obese. American Journal of Public Health, 94, 15741579.
  • Maddock, J. E. (2004). The relationship between obesity and the prevalence of fast food restaurants: State-level analysis. American Journal of Health Promotion, 19, 137142.
  • MCGI, (Michigan Center for Geographic Information). (2010). Michigan Geographic Data Library. Retrieved on April 29, 2013, from http://www.mcgi.state.mi.us/mgdl/?action=ext.
  • Mitchell, G., Hargreaves, A., Namdeo, A., & Echenique, M. (2011). Land use, transport, and carbon futures: The impact of spatial form strategies in three UK urban regions. Environment and Planning A, 43, 21432163.
  • Moore, L. V., & Diez Roux, A. V. (2006). Association of neighborhood characteristics with the location and type of food stores. American Journal of Public Health, 96, 325331.
  • Morland, K., Diez Roux, A., & Wing, S. (2006). Supermarkets, other food stores, and obesity: The atherosclerosis risk in communities study. American Journal of Preventative Medicine, 30, 333339.
  • Moudon, A. V., & Lee, C. (2003). Walking and biking: An evaluation of environmental audit instruments. American Journal of Health Promotion, 18, 2137.
  • Newman, P., & Kenworthy, J. (1989). Gasoline consumption and cities: A comparison of U.S. cities with a global survey. Journal of the American Planning Association, 55, 2437.
  • Newman, P., & Kenworthy, J. (1999). Sustainability and cities: Overcoming automobile dependence. Washington, DC: Island Press.
  • Newman, P., & Kenworthy, J. (2013). Urban sustainability and automobile dependence in an Australian context. In I. Vojnovic (Ed.), Urban sustainability: A global perspective (pp. 227–253). East Lansing, MI: Michigan State University Press.
  • Okrent, D., & Gray, S. (2010). The future of Detroit: How to shrink a city”. Time, November 11. http://www.time.com/time/nation/article/0,8599,2030761–1,00.html
  • Owens, P. (1993). Neighborhood form and pedestrian life: Taking a closer look. Landscape and Urban Planning, 26, 115135.
  • Pettinger, C., Holdsworth, M., & Gerber, M. (2007). “All under one roof?” Differences in food availability and shopping patterns in Southern France and Central England. European Journal of Public Health, 18, 109114.
  • Piacentini, M., Hibbert, S., & Al-Dajani, H. (2001). Diversity in deprivation: Exploring the grocery shopping behavior of disadvantaged consumers. International Review of Retail, Distribution and Consumer Research, 11, 141158.
  • Podagrosi, A., & Vojnovic, I. (2008). Tearing down Freedmen's Town and African American displacement in Houston: The good, the bad, and the ugly of urban revival. Urban Geography, 29, 371401.
  • Podagrosi, A., Vojnovic, I., & Pigozzi, B. (2011). The diversity of gentrification in Houston's urban renaissance: From cleansing the urban poor to supergentrification, Environment and Planning A, 43, 19101929.
  • Pothukuchi, K. (2005). Attracting supermarkets to inner-city neighborhoods: Economic development outside the box. Economic Development Quarterly, 19, 232244.
  • Pothukuchi, K., Mohamed, R., & Gebben, D.A. (2008). Explaining disparities in food safety compliance by food stores: Does community matter? Agriculture and Human Values, 25, 319332.
  • Powell, K., Martin, L., & Pranesh, C. (2003). Places to walk: Convenience and regular physical activity. American Journal of Public Health, 93, 15191521.
  • Powell, L. M., Slater, S., Mirtcheva, D., Bao, Y., & Chaloupka, F. J. (2007a). Food store availability and neighborhood characteristics in the United States. Preventive Medicine, 44, 189195.
  • Powell, L. M., Chaloupka, F. J., & Bao, Y. (2007b). The availability of fast food and full-service restaurants in the United States: Associations with neighborhood characteristics. American Journal of Preventative Medicine, 33, S240S245.
  • Public Sector Consultants Inc. (2001). Michigan Land Resource Project. Lansing: Public Sector Consultants.
  • Rapoport, A. (1987). Pedestrian street use: Culture and perception. In A. V. Moudon (Ed.), Public street for public use (pp. 8094). New York: Van Nostrand Reinhold.
  • Saelens, B. E., Sallis, J. F., Black, J. B., & Chen, D. (2003). Neighborhood-based differences in physical activity: An environmental scale evaluation. American Journal of Public Health, 93, 15521558.
  • Scott, M. M., Dubowitz, T., & Cohen, D. A. (2009). Regional differences in walking frequency and BMI: What role does the built environment play for blacks and whites? Health & Place, 15, 897902.
  • SEMCOG. (2003a). Land Use Change in Southeast Michigan: Causes and Consequences. Detroit: Michigan Council of Governments.
  • SEMCOG. (2003b). Fiscal Capacity of Southeast Michigan Communities: Taxable Value and its Implications. Detroit: Michigan Council of Governments.
  • Siegel, J. S. (2002). Applied Demography: Applications to Business, Government, Law, and Public Policy. San Diego: Academic Press.
  • Slater, T. (2006). The eviction of critical perspectives from gentrification research. International Journal of Urban and Regional Research, 30, 737757.
  • Smith, D. M., Cummins, S., Taylor, M., Dawson, J., Marshall, D., Sparks, L., & Anderson, A. S. (2010). Neighbourhood food environment and area deprivation: Spatial accessibility to grocery stores selling fresh fruit and vegetables in urban and rural settings. International Journal of Epidemiology, 39, 277284.
  • Sommer, B. B., & Sommer, R. (1997). A Practical Guide to Behavioral Research: Tools and Techniques. New York: Oxford University Press.
  • Southworth, M. (1997). Walkable suburbs? An evaluation of neotraditional communities at the urban edge. Journal of the American Planning Association, 63, 2844.
  • Southworth, M. (2005). Designing the walkable city. Journal of Urban Planning and Development, 131, 246257.
  • Sui, D. (2003). Musings on the fat city: Are obesity and urban form linked? Urban Geography, 24, 7584.
  • Swanson, B. C. (2008). Cities in crisis: A special analytical report on high school graduation. Bethesda, MD: Education Research Center.
  • Teaford, J. C. (2006). The metropolitan revolution. New York: Columbia University Press.
  • Thomas, J. M. (1997). Redevelopment and Race: Planning a Finer City in Postwar Detroit. Baltimore: The Johns Hopkins University Press.
  • U.S. Census Bureau. (2000). Census, 2000. Summary File 1. Washington, DC: Government Printing Office.
  • U.S. Census Bureau. (2010a). American Community Survey, 2008. Washington, DC: U.S. Government Printing Office.
  • U.S. Census Bureau. (2010b). Census, 2010. Summary File 1. Washington, DC: Government Printing Office.
  • U.S. Census Bureau. (2011). 2010 Census shows Black population has highest concentration in the South. CB11-CN.185. Washington, DC: U.S. Government Printing Office.
  • U.S. Department of Health and Human Services (USDHHS). (2000). Healthy people 2010 Vol. 2, 2nd ed. Washington, DC: U.S. Government Printing Office.
  • U.S. Department of Transportation (USDOT). (1994). National bicycle and walking study. Case No. 1: Reasons why bicycling and walking are and are not being used more extensively as travel modes. Washington, DC: U.S.
  • U.S. Department of Transportation (USDOT). (2005). 2001 National Household Travel Survey. Washington, DC: U.S. Department of Transportation.
  • Vojnovic, I. (2000). Shaping metropolitan Toronto: A study of linear infrastructure subsidies. Environment and Planning B: Planning and Design, 27, 197230.
  • Vojnovic, I. (2003). Governance in Houston: Growth theories and urban pressures. Journal of Urban Affairs, 25, 589624.
  • Vojnovic, I. (2006). Building communities to promote physical activity: A multi-scale geographical analysis. Geografiska Annaler Series B, 88, 124.
  • Vojnovic, I., Jackson-Elmoore, C., Summers, J., & Bruch, S. (2006). Principles, practices, and tools in Michigan for promoting increased physical activity and non-motorized travel. Cities, 23, 6790.
  • Vojnovic, I., Smith S., Kotval-K, Z., & Lee, J. (2008). Built environments: Planning cities to encourage physical activity. In D. Davies (Ed.), Obesity in childhood and adolescence: Medical, biological, and social issues (Vol. 1, pp. 93129). Westport, CT: Praeger.
  • Vojnovic, I., Lee, J., Kotval-K, Z., Podagrosi, A., Varnakovida, P., LeDoux, T., & Messina, J. (2013). The burdens of place: A socio-economic and ethnic/racial exploration into urban form, accessibility, and travel behavior in the Lansing capital region, Michigan. Journal of Urban Design. 1, 135.
  • Wang, M. C., MacLeod, K. E., Steadman, C., Williams, L., Bowie, S. L., Herd, D., Luluquisen, M., & Woo, M. (2007). Is the opening of a neighborhood full-service grocery store followed by a change in the food behavior of residents? Journal of Hunger and Environmental Nutrition, 2, 318.
  • Williams, D. R. (2005). Patterns and causes of disparities in health. In Robert Wood Johnson Foundation (Ed.), Policy Challenges in Modern Health Care (pp. 115134). Piscataway, NJ: Rutgers University Press.
  • Yang, J., French, S., Holt, J., & Zhang, X. (2012). Measuring the structure of U.S. metropolitan areas, 1970–2000. Journal of the American Planning Association, 78, 197209.
  • Yang, Y., Diez Roux, A. V., Auchincloss, A. H., Rodriguez, D. A., & Brown, D. G. (2011). A spatial agent based model for the simulation of adults’ daily walking within a city. American Journal of Preventative Medicine, 40, 353361.
  • Zenk, S. N., Schulz, A. J., Israel, B. A., James, S. A., Bao, S., & Wilson, M. L. (2005). Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in metropolitan Detroit. American Journal of Public Health, 95, 660667.
  • Zenk, S. N., Schulz, A. J., Israel, B. A., James, S. A., Bao, S., & Wilson, M. L. (2006). Fruit and vegetable access differs by community racial composition and socioeconomic position in Detroit, Michigan. Ethnicity and Disease, 16, 275280.
  • Zimowski, M., Tourangeau, R., Ghadialy, R., & Pedlow, S. (1997). Nonresponse in Household Travel Surveys. Prepared for Federal Highway Administration. Chicago: NORC.

Biographies

  1. Top of page
  2. ABSTRACT
  3. LITERATURE REVIEW: THE BUILT ENVIRONMENT AND TRAVEL BEHAVIOR
  4. CASE STUDIES AND METHODS
  5. FINDINGS
  6. COMMENTARY AND CONCLUSIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Biographies
  • Igor Vojnovic is an associate professor in the Department of Geography at Michigan State University. He holds cross-appointments and affiliations with the Global Urban Studies Program, Environmental Science and Policy, the Center for Global Change and Earth Observation, and Urban and Regional Planning. His areas of research include metropolitan environments, urban form, urban design, transportation and local governance. His work has been published in journals such as Environmental Conservation, Environment and Planning A, Environment and Planning B, Urban Geography, GIScience & Remote Sensing, Cities, Journal of Urban Design, Health & Place, Journal of Urban Affairs, GeoJournal, and Geografiska Annaler Series B. He is also an associate editor for the Journal of Urban Affairs.

  • Zeenat Kotval-K is a doctoral candidate in Urban Geography at Michigan State University. She holds MS degrees in Urban and Regional Planning (from Michigan State University) and Hospitality and Tourism Management (from the University of Massachusetts at Amherst). Her research interests include sustainable development, urban built environments and transportation. Her dissertation focuses on the Impacts of the Urban Built Environment on Travel Behavior, Gasoline Consumption, and Vehicle Emissions in the Detroit Region. Kotval-K has published several book chapters and articles in journals, including the Journal of Urban Design, Journal of Planning Practice and Research, Town Planning Review, and Local Economy.

  • Jieun Lee is a Visiting Assistant Professor of Geography/ Environmental Studies at New College of Florida. She is also a PhD. candidate in the Geography Department at Michigan State University. Lee specializes in the areas of urban and health geography, with a focus on urban built environments, marginalized communities, and gender and public health. Lee has several academic publications, including in the Journal of Urban Design. Previously she was a researcher at Seoul Development Institute in Seoul, Korea, where she contributed to several monographs on urban development, and was named Researcher of the Year in 2005.

  • Minting Ye is a PhD student in Department of Geography at Michigan State University. Her main research areas are in urban-economic development and redevelopment, gentrification processes, and suburbanization, with a focus on Hong Kong. She has published more than ten articles, including in Economic Development Quarterly, International Journal of Remote Sensing, Environment, Environmental Monitoring and Assessment, Progress in Geography (In Chinese), Resource Science (In Chinese), and Acta Ecologica Sinica (In Chinese).

  • Timothy LeDoux is a Ph.D. candidate in Geography at Michigan State University and an Assistant Professor/GIS Coordinator in the department of Geography and Regional Planning at Westfield State University. His research interests explore the intersection between urban food environments, poverty, hunger and racial segregation in industrial and post-industrial cities across the United States. His current work examines the evolution of the urban food environment in Detroit, Michigan in response to massive economic restructuring, broadening socioeconomic neighborhood inequality and suburbanization. This research has been published in journals such as Health & Place and the Journal of Urban Design.

  • Pariwate Varnakovida (Ph.D. Michigan State University 2010) is an assistant professor at Prince of Songkla University in Phuket, Thailand. He holds a BSc in Mathematics and a MS in Technology of Information System Management both from Mahidol University. His academic interests include GIScience, urban and environmental modeling, urban forms, environmental impact assessment, sustainable urban development, and urban and natural resources management, as well as using photogrammetric applications, Remote Sensing, GIS, quantitative methods, and Cellular Automata.

  • Joseph P. Messina (Ph.D – University of North Carolina at Chapel Hill) is a Professor of Geography, Member of the Center for Global Change and Earth Observations, Member of AgBioResearch, Member of the Center for Latin American and Caribbean Studies, Member of the African Studies Center, Member of the Ecology, Evolutionary Biology, and Behavior Program at Michigan State University. He was awarded research honors from NASA through the New Investigator Program, the National Institutes of Health Roadmap Program, and the Sigma Xi / MSU Young Scholar of the Year. Professor Messina explores spatial drivers of health, disease, and land use change in many countries around the world.