«Contesting the streets Volume 18, number 1 • 2016 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
School Comparisons To construct the comparison between a nonpermissive and a permissive regulatory environment, a statistical analysis was performed to identify similar jurisdictions. Percentage of Spanish-speaking population was one criterion, based on local studies that connect sidewalk vending with the culture of Latin American immigrants (Rosales, 2013; Goetz and Wolstein, 2007). The second criterion was population living under the poverty level, taken from studies that show sidewalk vending is a fallback enterprise for many people who struggle to generate sufficient incomes within the formal economy (Austin, 1994; Morales, 2000). Three jurisdictions with more permissive vendor regulations were chosen for deeper analysis based on Spanish-speaking populations at least as high as in South Los Angeles and poverty rates that were more than 20 percent.
The statistical analysis was then extended to specific schools to identify the closest matches for specific environmental variables. Data were compiled reflecting the socioeconomic characteristics of residents living within a 1/2-mile proximity of the applicable school campus, as opposed to students attending that school or their households. In addition, mRFEI scores were collected for the applicable census tract of each school,13 as was the aggregate number of bicyclist and pedestrian collisions within 1/2 mile of each school from 2007 through 2009.14 Compton emerged as the ideal comparison environment because it was the only jurisdiction to offer close matches to South Los Angeles school neighborhoods for each of the environmental variables. Elementary schools from these jurisdictions were then matched based on close similarities for one of the previously mentioned environmental variables.
mRFEI data are accessible at http://www.cdc.gov/obesity/.
These data were compiled from the Safe Transportation Research and Education Center, Transportation Injury Mapping System, Safe Routes to School collision map viewer (2007 through 2009), accessed November 2013.
Field Data Collection School environments selected for the assessment were observed during March and April 2014 each for a 60-minute period beginning at the closing bell, which earlier studies documented as the time when sidewalk vendors make most of their sales near schools (Goetz and Wolstein, 2007). Observations took place only on midweek days (Tuesday through Thursday), but days with irregular bell schedules were avoided. In most cases, observations for matching schools took place on the same day or on consecutive days, so that conditions between the comparison schools would be most similar.
During each observation period, an assessor walked the periphery of the school campus and adjacent streets within a two-block radius to record the presence of any sidewalk food vendors and the general content of snacks and beverages offered. Two to four additional assessors were placed at key observation points, as needed, to tally the number of pedestrians and bicyclists present during the observation period.
Student surveys were administered from October through December 2014 at the four elementary schools in South Los Angeles where field observation had occurred previously. CHC was not able to coordinate with school personnel to administer surveys at comparison schools in Compton. Survey respondents (anonymous 4th and 5th grade students) indicated the general frequency (never, occasional, or frequent) for which they obtained snacks from eight types of access points within the wider nutrition environment and how often they obtained items from sidewalk vendors from five categories. Students also indicated how often they ate lunch or snacks provided at school.
Results Overall, results did not show noticeable differences between the two jurisdictions for sidewalk vendor presence. The quantity of observed vendors ranged from the highest (eight) to the lowest (zero) in South Los Angeles, while Compton had a more moderate range (exhibit 1). The permissiveness of the regulatory environment does not seem to clearly influence vendor presence near schools in these communities.
The analysis of socioeconomic factors, however, suggests that poverty and language might influence the presence of sidewalk vendors near schools. Elementary S1 and Elementary S3 had the highest number of vendors (eight and six, respectively) and also the highest percentage of people living in poverty (32.1 and 31.5 percent, respectively). By comparison, Elementary S2 and Elementary S4 had no vendor presence and the lowest population of Spanish speakers (42.6 and 46.8 percent, respectively). A similar pattern was not evident in Compton, where differences in poverty rates between schools were smaller and rates of Spanish speakers were higher.
Results also did not reveal a clear relationship between sidewalk vendor presence and the quality of the retail food environment or bicyclist and pedestrian safety (exhibit 1). Underdeveloped retail food environments might be expected to correlate with higher sidewalk vendor activity, but schools in this assessment with the highest local mRFEI scores (Elementary C2 and Elementary C4) had only moderate vendor presence (two and three observed, respectively), whereas schools with the lowest local mRFEI scores (Elementary S4 and Elementary S1) had highly differentiated vendor
Sources: Health Impact Assessment field assessments and student surveys; Missouri Census Data Center, Circular Area Profiles, American Community Survey version (2007 to 2011); Centers for Disease Control and Prevention, Children’s Food Environment State Indicator Report (2011); Safe Transportation Research and Education Center, Transportation Injury Mapping System, Safe Routes to School collision map viewer (2007 through 2009), accessed November 2013 presence (zero and eight observed, respectively). Another expectation could be that higher vendor presence leads to safer or less safe bicyclist and pedestrian environments, but schools with the lowest local collision rates (Elementary C2 and Elementary C1) had moderate vendor presence (two and four observed, respectively), and schools with the highest local collision rates (Elementary S1 and Elementary S2) had highly differentiated vendor presence (eight and zero observed, respectively).
South Los Angeles survey responses indicate that 62.8 percent of students at least occasionally obtain snacks and beverages from sidewalk vendors. Although this percentage represents a significant rate of patronage by students and caretakers, it is less than the rate with which they, not surprisingly, even more regularly patronize formal commercial sources within the wider nutrition environment (exhibit 2). By comparison, 84.1 percent of students patronize fast-food restaurants at least occasionally, and 76.7 percent patronize corner stores at least occasionally. These two access points are routinely highlighted as a source of high-calorie food and sugary beverages in underdeveloped food retail environments (Bassford et al., 2012; Borradaile et al., 2009). Although snack and beverage purchasing trends vary somewhat between schools, sidewalk vendors appear to compete for student and caretaker patronage on relatively close terms with mobile food trucks, which offer many of the same items, and with vending machines, which are healthier food sources in some cases because of tighter inventory regulations at schools and other public facilities. Survey responses also indicate that households remain a highly prevalent source of the snacks and beverages that students consume.
The frequency of snack and beverage purchases from sidewalk vendors, in general, is consistent with the degree of vendor presence observed in field observations. As vendor presence moved from a high of eight observed at Elementary S1 to zero observed at Elementary S2 and Elementary S4, the rate of at least occasional student patronage also fell from 84.5 to 40 percent. Indeed, the school with the highest vendor presence had a percentage of students patronizing vendors (84.5 percent) that was higher than the patronage of corner stores (77.3 percent) and near that of
fast-food restaurants (92.8 percent). Results of the student survey suggest, however, that Elementary S2 typically may have more sidewalk vendors than was observed in the assessment, because rates of vendor purchases there are very similar to those at Elementary S3 (exhibit 2).
Student responses also raise questions about the competition between home and school foods, which researchers have consistently found are healthier sources, and neighborhood food (Lachat et al., 2012). An inverse relationship was found between sidewalk vendor presence and participation in school meal programs and obtaining snacks and beverages from students’ households. Overall, only 42.9 percent of elementary school students reported eating school lunches at least occasionally. The rate of at least occasional school lunch patronage, however, rose within the school sample as sidewalk vendor patronage decreased. Elementary S4 had the highest rate of frequent snacks and beverages obtained from home (72.9 percent), and it also had the lowest rate of at least occasional sidewalk vendor patronage (40.0 percent) and the lowest rate of at least occasional fastfood restaurant patronage (78.6 percent). Elementary S1 conversely had the lowest rate of frequent snacks and beverages obtained from home (34 percent) and had, by far, the far highest rates of at least occasional sidewalk vendor patronage (84.5 percent) and occasional fast-food restaurant patronage (92.8 percent).
As seen in previous studies (Tester, Yen, and Laraia, 2010), unhealthy snack food (for example, chips, cookies, candy, ice cream, elote) was, by far, the likeliest item sold by sidewalk vendors in this assessment (exhibit 3). Sugar-sweetened beverages (for example, soda, sports drinks, fruit punch, hot chocolate) were offered by all but two of the observed vendors found in South Los Angeles. Hot or prepared foods (for example, tamales) and healthy beverages (for example, 100 percent fruit juice, water) were offered by vendors at only one school (Elementary S3). Despite the popularity of fruteros throughout the city, no vendors at any schools observed in this assessment offered fruits or vegetables.
As expected, survey responses indicated that unhealthy snack food was purchased at least sometimes by 92.3 percent of occasional vendor patrons and purchased often by 60.9 percent of frequent vendor patrons (exhibit 3). Each of the previously mentioned product categories, however, was purchased at least sometimes by 72.8 to 92.3 percent of occasional vendor patrons
and purchased often by 39.1 to 60.9 percent of frequent vendor patrons, including products that were rarely or never observed during field assessments. A lack of clarity in the survey question or recall difficulty may have skewed survey response data toward a less accurate depiction of snack and beverage purchases from sidewalk vendors.
Field assessments indicated that one-half of the sidewalk vendors observed near schools caused sidewalk congestion among pedestrians (because of the crowding of waiting customers). In no instance did observers note that vendor-related congestion caused pedestrians to walk into the street or otherwise come into conflict with vehicles.
Observations indicated that a range of 22 to 82.5 percent of students at each elementary school walked home and a very small number of students bicycled home at the conclusion of the school day (see exhibit 4). During each observation period, 38 to 271 adult pedestrians (most of whom accompanied exiting students) and 1 to 9 adult bicyclists were noted in the vicinity of school campuses. Schools with higher sidewalk vendor presence in either jurisdiction were not likelier to have more or fewer pedestrians or bicyclists active in their vicinity. Schools with the lowest student pedestrian rates (Elementary S4 and Elementary S3) had highly differentiated vendor presence (0 and 6 observed, respectively), and schools with no vendor presence (Elementary S4 and
Elementary S2) had highly differentiated student pedestrian rates (22.0 and 82.5 percent of school enrollment, respectively). Schools with moderate vendor presence (Elementary C4 and Elementary C1) had highly differentiated student bicyclist rates (11 and 0 observed, respectively).
Discussion The evidence gathered for this assessment does not suggest that citywide prohibitions against sidewalk food vending are functioning as a protection of good-quality nutrition environments surrounding school campuses. Whether enforcement resources are inadequate to consistently enforce the prohibitions or the income-generation needs of low-income people are acute enough to justify the risks, school environments in South Los Angeles appear as likely to include an abundance of sidewalk vendors selling snacks and beverages as those in Compton, which has more permissive regulations. Assessment data appear to confirm other research suggesting a higher presence of sidewalk vendors in areas where residents have a more tenuous foothold in the formal economy and where the cultural context is more accustomed to informal enterprise in the public realm (Devlin, 2011; Dunn, 2014). As a result, reasonable, legitimized forms of sidewalk vending would be adaptable and socioeconomically beneficial in many sections of Los Angeles. Because the prohibition does not appear to discourage vendor activity near schools, permitted sidewalk vending would not likely amplify the hazards from nutrient-poor snacks and sugary beverages that already exist within the wider food ecology occupied by students.
Combined results of the observations and surveys reveal that more vendors generate sales from students by offering items that are less healthy than food and beverages offered by schools. If sidewalk vending were legalized, these results suggest that efforts should be made to limit the presence of vendors near schools or encourage those offering healthier items.
This assessment does not clarify whether sidewalk vendors have a positive effect on bicycle and pedestrian safety or the creation of defensible space. The study, however, does not support unsubstantiated views that the presence of sidewalk vendors contributes to unsafe and unkempt conditions in a neighborhood. More specific evidence is needed to assess the relationship of sidewalk vendors to a variety of environmental factors, such as crime prevention, food safety, and ancillary business activity.