Final – Tamanna Rahman – Part 2

While the mapping results of the above factors demonstrate that there are higher concentrations of the low income, single-headed households with kids, and or limited English proficiency, it is difficult to discern whether this distribution is important. In order to determine whether these results are statistically significant, hotspot analysis (spatial analyst) was conducted to determine the significance of these distribution clusters. Doing this exercise will help determine locations where there are elevated levels of food insecurity. Before the data could be analyzed using Hotspot Analysis, the vector-based data had to be rasterized and then reclassified. Even though this process can be tedious, the modeling capability of ArcGIS helped to streamline the procedures. The results of the modeling process is provided below in Figure 11.

Figure 11a: Example of Modeling Used in Spatial Analyst – Converting feature class to raster

Figure 11b: Example of Modeling Used in Spatial Analyst – Reclassifying rasters

This process provides the following output (Figure 12) that shows the areas where food insecurity exists based on the factors inputted into the model. From this map at county level, we can see that there are high levels of food insecurity in the expected areas of South LA, East LA and Wilmington. There are also some major pockets in the San Fernando Valley. This data also highlighted the San Gabriel Valley, but that could be a result of the higher number of immigrant populations. Based on this map, the most urban, populated and low-income areas of Los Angeles County have high levels of food insecurity. The project area is also visible in this map, and it demonstrates that the clusters shown in the three separate risk factor maps were statistically significant, and confirms the conclusion that the East side of the project areas is more food insecure and thus will require a greater amount of resources. Having improved transit service can improve access to healthier food options. Although the mid-Wilshire area has a plethora of restaurants, sections of this area near the subway alignment are “food swamps” because of the relative lack of full service markets where residents can by fresh produce.

Figure 12: Output of Hotspot Analysis, applied to Los Angeles County


Economic factors such as poverty and low car ownership, as well as other personal factors such as level of education and English speaking ability can have a significant impact on what opportunities people have for healthy living. In the research arena, there has been a shift from a focus on individual and behavioral factors that influence food choices, to examining the physical or environmental factors and the geographical distribution of affordable healthy food (Morland et al., 2002; Sharkey et al., 2009).

Several approaches to increasing access to healthy, fresh foods have gained traction in recent years. In Pennsylvania, the Fresh Food Financing Initiative has invested almost $60 million in supermarket development in underserved areas. Other programs include the Fair Food Program, which supports local farmers in their production and marketing of fresh food; and Common Market, a non-profit that connects farmers and urban consumers. Other programs target development of community gardens in underserved areas in order to make healthy, safe, affordable food available in disadvantaged communities. Preliminary evidence suggests that involvement in community gardening may be associated with higher fruit and vegetable intake (Alaimo et al. 2008) as well as improve relations with neighbors (Teig et al. 2009).

Transportation and land use policies, programs and projects attuned to the communities food security needs can build bridges between local fresh produce, food retailers and consumers. Transportation programs and projects can make it easier for low-income families, the aged, and other with mobility challenges and particular nutrition needs to access supermarkets, farmers’ markets and other sources of affordable, healthful foods. Transportation improvements may include increasing bus routes to food retailers and supermarket-sponsored shuttle services.

Where poor access to healthy food options exists or is exacerbated by project activities, possible mitigation measures include allowing farmers markets on Metro property near stations, assuring good bus connectivity between stations and supermarkets, and coordinating efforts with local planning and redevelopment agencies that are working to improve the availability of healthy food options.

Improving transportation options to and from such food sources as supermarkets and farmers’ markets increases a community’s access to healthy foods. The area between Crenshaw and La Brea is particularly bereft of full service markets selling fresh fruit and vegetables. Through improved mobility for residents in food deserts such as this, and by making areas near the alignment more attractive to food retailers selling healthier foods, the proposed subway can play a valuable role in helping attain the goals set out in the Los Angeles Food Policy Task Force’s recently released “Good Food for All Agenda.”

Summary of Skills Used:

Required (3):

  • Modeling: used to create raster files of data and then to reclassify the raster files
  • Measurement/Analysis: created ½ mile buffer around transit stations
  • Original data: Food environment data was collected from various sources. Also used the CA Physical Fitness Test to determine students who are overweight.

Optional requirements (7):

  • Hotspot Analysis/Spatial Analsyst: prepared and created raster dataset that was used to assess food insecurity.
  • Create Indices: created a food security index by combining several census data
  • Charts/Images: Used CHIS data to determine what type of transportation people use to get groceries, based on their ability to obtain food.
  • Aggregating attributable fields: Attribute fields were aggregated in ArcMap to create a new attribute, percent Grade 5, 7 and 9 students who were overweight [(students not in HFZ/total students taking test )x 100].
  • Boundary sub-set selection: created smaller shapefiles by selecting features that were located within a distance (0.5 miles) of the proposed route (study area), or those features that were contained within the study area (food establishments).
  • Geoprocessing: Used clipping tool to modify LA County layer.
  • Geocoding: Geocoded locations of supermartkets/grocery stores, farmers’ markets, convenience stores and fast food establishments.
  • Buffering: created ½ mile buffer around subway stations to denote walking distance.
  • Custom shapefile creation: Used Metro route information to create a shapefile of the proposed route and station locations.
  • Inset Map: created inset map showing LA County on California map, and project area on LA County map. Included both set of maps in two separate data frames to be displayed on the final layout.


Final – Tamanna Rahman – Part 1

Hear Ye! Hear Ye! Read all about it: Metro’s Westside Subway Extension Project moves forward

Although Los Angeles has a public transportation system that is currently lacking when compared with other major cities, Angelenos can become enthusiastic about potential change as strides are taken to push Los Angeles toward becoming a city more conducive to public transportation.

With the recent approval of the first phase of the long sought after “Subway-to-the-Sea,” Los Angeles is shifting towards a more integrated transit system that not only would add efficiency, but also a number of other benefits to riders and community members at multiple levels. The Los Angeles Metropolitan Transportation Agency, or Metro, recently released the proposed alignment that would extend the Purple Line subway from its current terminus at Wilshire and Western in mid-town Los Angeles approximately 9 miles westward along the Wilshire Corridor toward Beverly Hills, Century City, Westwood and West Los Angeles (See Figure 1). The proposed subway would not preclude other transit alternatives. Pedestrian and bicycle infrastructure, as well as rapid bus lanes, could be integrated into the project. Likely to alleviate some of the epic traffic congestion that occurs during rush hour along that corridor, Angelenos can look forward to 2022, the expected completion date at the earliest, to taking the subway as opposed to driving.

Figure 1: Metro’s proposed Westside Subway Extension Project

Source: Metro

Due to the subway project’s magnitude, it has the potential to transform surrounding neighborhoods and impact multiple, overlapping populations. These neighborhood changes could affect health in a number of ways, one of them being impacts on the food retail environment. Our transportation system plays a key role in how communities access fresh nutritious food. Millions of Americans, especially people with low-incomes, the elderly, disabled, and other transit-dependent populations, have difficulty accessing fresh, nutritious food. With diabetes, overweight and obesity rates reaching epidemic proportions, this project can set an example for how development can be done with health in mind.

Barriers to Healthy Food:

These health concerns illustrate the importance of better understanding the complexities of our food environment because there are a number of variables that come into play. Studies have proven that people with lower incomes are at greater risk to be obese than those in the middle and upper class. As contradictory as it may seem, it is a harsh reality that an increasing number of families and individuals are finding it difficult or impossible to maintain a healthy diet. Much research has addressed the fact that healthy food is often more expensive and difficult to obtain compared to a less healthy diet that tends to be high in calories and fat and low in quality (Furey et al., 2001). The result is that lower income groups do not get adequate nutrient intake when compared to people in higher income brackets (Clifton, 2004; Morland et al., 2008). Healthy food may also cost more for low income households because purchases are made in smaller quantities (Morland et al., 2002). Many healthy foods, such as fruits and vegetables, are often more expensive and are often viewed as luxury items by those with low-incomes. Morland et al. (2002) and Sharkey et al. (2009) report that the price of healthy food and the variation between prices is a major concern for those living on low incomes.

Transportation and Food Access:

Full service supermarkets and farmers’ markets are scarce in low-income areas. Residents of areas poorly served by food retail options are also more likely than the general public to be transit-dependent, so it can be difficult for them to travel to food markets located outside of their immediate neighborhoods. According to data from the federal government’s survey of personal transportation, a quarter of low-income households lack access to an automobile (reference). This percentage is higher in some urban areas, leaving many residents dependent on walking, cabs and transit for food shopping trips. For example in the Metro Service Planning Area of LA County, those who are not able to afford enough food were less likely to rely on their personal vehicles than walking/biking or using public transportation for their food shopping trips (See Figure 2).

Figure 2: Type of transportation used to access grocery stores by food security (see figure in uploaded file)

A poor transportation system cuts off access to many food outlets—especially for those who do not own a car or have access to reliable and affordable public transportation. Residents of lower income and minority neighborhoods in most urban areas face a double burden that severely limits their access to fresh, healthy food. Those food markets that are located in low-income neighborhoods are often smaller, with less selection in general, and less and lower quality produce. With fewer supermarkets and farmers’ markets, residents of low-income neighborhoods rely heavily on convenience stores, liquor stores, and small ethnic grocers for groceries. These establishments stock packaged and processed food items, but few, if any, fruits and vegetables. Fast food chains, serving high-fat items are often the only source of prepared foods

Assessment of the Retail Food Environment:

In an effort to better understand the current food retail environment and the factors that may limit access, I decided to explore the area around the proposed stations of the Westside Subway Extension because this project serves as an opportunity to develop with the goal of designing healthy communities.

This analysis examines the coverage of food stores and key demographic patterns in the defined study area. Analyzing the current distribution will help to understand what type of food the community has access to and where there are clear gaps and a need for improving healthy food access.

With alignment information from Metro and census tract data from UCLA Mapshare, ArcGIS was used to create a layer shapefile of the Purple Line extension. Census tracts within a ½ mile of the proposed route were selected to define the study area (see Figure 3). The Los Angeles County GIS Portal provided a city layer that was overlaid on the study area for better geographical orientation. For this assessment, census tracts were chosen as the unit of analysis. The demographics of this area will be detailed later.

Figure 3: Defined study area

Data Source: Metro (route), UCLA Mapshare (census tract and local freeways), LA County GIS Portal (cities)

In order to assess the food environment from the perspective of transit-users, a ½ mile buffer radius was created around each subway station (Figure 4). Several research studies have used either ¼ or ½ miles radii as walkable distances, but 1/2 mile radius was used in this analysis to get a greater sampling of the food options.

Figure 4: Study area with ½ mile buffer around proposed subway stations

Food store data:

Several levels of the food environment have been identified, with one being the “community environment.” This defines the place where food can be obtained, including grocery stores, convenience stores, specialty stores, restaurants, farmers’ markets that are generally open to the public. This analysis focused on supermarkets/grocery stores farmers’ markets, convenience stores and fast food establishments. The North American Industry Classification System (NAICS) codes were used to select supermarkets/grocery stores and produce stores that were geocoded and mapped. Farmers’ markets locations were gathered from the California Farmers Market Association website and geocoded and spatially joined. Figures 5 and 6 display the spatial distribution of the healthy and unhealthy food environment, respectively, within the project area. While there are some clear clusters of supermarkets and grocery stores on the west side, there are evident gaps in the availability of healthy food sources near certain locations in the project area. For example, Figures 5a and 6a displays the same three stations, but it is clear to see that more fast food establishments and convenience stores are readily available compared to supermarkets/small chain grocery stores and farmers’ markets. Table 1 denotes the number of food establishments, by type of business, available within ½ mile radius of each Metro subway station. Clearly, fast food establishments and convenience stores have a majority over establishments that offer nutritious, healthful foods at affordable prices.

Figure 5: Healthy food Environment – Supermarkets/grocery stores and Farmers’ Markets

Data Source: Network for a Healthy California, California Farmers’ Market Association, UCLA Mapshare, Metro

Figure 5a: Healthy Food Environment – looking at Wilshire/La Brea, Wilshire/Fairfax and Wilshire/La Cienega stations

Data Source: Network for a Healthy California, California Farmers’ Market Association, UCLA Mapshare, Metro

Figure 6: Unhealthy food environment – Convenience stores and fast food

Data Source: Network for a Healthy California, UCLA Mapshare, Metro

Figure 6a: Unhealthy food environment: looking at Wilshire/La Brea, Wilshire/Fairfax and Wilshire/La Cienega stations


Data Source: Network for a Healthy California, UCLA Mapshare, Metro

Table 1: Distribution of Supermarkets, Farmers’ Markets, Convenience stores and fast food establishments within ½ mile of the proposed subway stations (Please see attached document for table)

Demographics of Study Area:

As shown below (Table 2), the demographic composition of neighborhoods along the Wilshire Corridor are highly varied, generally following a gradient from predominantly low-income, Latino and Asian residents in Koreatown/Wilshire Center on the eastern edge of the corridor to predominantly high-income, majority white neighborhoods to the West.  County-wide the population is expected to be significantly older with fewer Whites and more Latinos by 2035.  Similar demographic trends could be expected for most of the project area. Except where noted, data are from Metro’s 2010 Westside Subway Extension DEIR/EIS, “Technical Report 08 – Community and Neighborhood.” For data not in the DEIR/EIS, data from the Los Angeles Times’ “Mapping L.A. Project” were used.  Neighborhood names and boundaries varied somewhat between the two data sources.

Table 2: Demographic Characteristics of Neighborhoods along the Wilshire Corridor (displayed East to West)  (Please see attached document for table)

There are also a number of schools in the project area.  Schools located within 0.6 miles of the proposed subway alignment include 27 primary schools with enrollment of 12,499 students and 8 secondary schools with an enrollment of 11,217 students. While not all are residents or walk to school, a large number can potentially walk to many of the fast food establishments or convenience stores located nearby.

Each year, California students take part in physical fitness test that evaluate a child’s performance according to established benchmarks.  Students are placed into categories based on their performance and those who do not meet the standard are said to fall outside the Healthy Fitness Zone (HFZ). These students are considered overweight and at risk for diseases that result from sedentary living and unhealthy diet. For this study, I characterized childhood overweight issues by analyzing the Body Composition test results available through the California Department of Education’s Dataquest website. After determining the schools located within the project area, the fitness results were aggregated for students in grades 5, 7, and 9 at all active public schools located within each zip code. Attribute fields were aggregated in ArcMap to create a new attribute, percent Grade 5, 7 and 9 students who were overweight [(students not in HFZ/total students taking test )x 100]. These results were then mapped and are available in Figure 7. While zip code level data may be too large to show neighborhood level effects, there has been research demonstrating that health status of an individual may be linked to their zip code, indicating that one’s environment may be related to health status. There is clearly more overweight students on the east end of the project area compared to the west. Income is also a strong predictor of student weight status. Median household income data (obtained from Social Explorer) showed a slight relationship between the student weight distribution and income levels in the area. However, there was no method of assuming that the children who attended these schools also lived in the area, therefore, the results were not substantive enough to include ,.

Figure 7: Percent of Students (Grade 5, 7 and 9) Not in the Healthy Fitness Zone for Body Composition (Overweight), by school zip code, 2010-2011

Data source: CA Dept of Education – Physical Fitness test results, CA Dept of Education – School Directory, Metro, UCLA Mapshare (zip code polygon and highways)

Creating a Food Insecurity Index:

Many researchers have studied how food environments vary across communities. Several characteristics that food desert areas tend to have are: larger proportions of residents without high school degree, higher poverty rates, lower median family incomes, greater fractions of families living in rural areas, a larger elderly population, higher amounts of small grocery stores and convenience stores per capita, and a lack of access to fresh food (Larson, Story and Nelson, 2009; Morton and Blanchard, 2007). Considering these factors and understanding the demographics, I wondered whether there was a measure that could help target areas where resources need to be focused. Recent research has shown that food insecurity – the limited or uncertain availability of nutritionally sound, safe food – is common in the recent recession and is positively associated with being overweight (Adams, Grammar-Strawn, and Chavez 2003). Food-insecure youths may be particularly vulnerable to inhospitable food environments, as they turn to fast food more frequently (Widome et al., 2009). Families that are limited by income and time also rely on more “convenient” meals that are not always nutritious.

In order to understand where food insecurity can occur in the project area, 3 factors were analyzed:

  • The percentage of single female-headed households with kids (Figure 8)
  • Households living in poverty (income below $20,000) (Figure 9)
  • Populations poorly speaking English (Figure 10)

Female-headed households with children tend to have limited income. These mothers are likely to work more hours to provide a living for their families. With more hours away from the home, they may not always have the time to provide a home-cooked meal, and therefore may rely on fast-food or restaurants. The next variable considered those living in poverty, which according to the Health and Human Services Poverty Guidelines, is below $20,000 for a household of four. Household with lower incomes are at a limited capacity to afford necessary daily needs. Higher cost of living and transportation may outweigh the luxury of fresh fruits and produce, forcing people to choose between nutrition and cheaper alternatives. Another factor that tends to indicate immigrant populations and lower socioeconomic status is the ability to speak English. Research indicates that people with limited English proficiency may have greater language and cultural barriers that are likely to contribute to poorer health outcomes.

Figure 8: Distribution of single female-headed households with kids

Data Source: Census, UCLA Mapshare, Metro

Figure 9: Distribution of Percent of Households with Incomes below $20,000

Data source: US Census, UCLA Mapshare, Metro
Figure 10: Distribution of Poor English-Speaking Population in the Study Area by census tract