Outliers make for great stories and headlines, but they don’t do much for policy discussions—particularly school choice policy. Recently, there has been a flurry of headlines citing tales of “extreme sacrifice” by Detroit students in their efforts to commute great distances to the schools of their choice. The reality is that the majority of Detroit students, charter or traditional, don’t travel farther than four miles or ten minutes to school.
Using 2013–14 data from over one hundred thousand Detroit students’ homes and enrolling schools, Data Driven Detroit conducted a study with the Skillman Foundation and Excellent Schools Detroit to better understand the school commute. They grouped students’ residences into census tracts and measured the driving distance to schools, finding that the average K–8 charter school commute was 3.53 miles. The average high school commute was 4.92 miles.
With Google Maps’ new Direction and Distance APIs, we can estimate commute times and the most direct street routes with updated streetscape data. Using a sample of home addresses from 9,579 Detroit students enrolled in eighteen charter schools authorized by Grand Valley State University, we found that the typical student travels 3.5 miles and 8.9 minutes to school. This trend is consistent with national data from the 2008 National Household Transportation Survey, which indicates that the average school commute for students between the ages of six and twelve was 3.6 miles (it increased by student age).
There are outliers in the data. In our analysis, there was at least one student per grade level who travelled thirty or more miles to school. Predictably for a city large in geography and robust in choice, Detroit students’ commutes do not fall into a normal distribution. That is, the data indicates that the majority of students travel a very short distance (i.e., three miles or less), while few travel much farther to access the schools of their choice. The stories of extreme sacrifice to reach particular schools of choice aren’t inaccurate; they are just not representative of the majority of Detroit students.
The fact that Detroit parents are self-selecting schools that are in relatively close proximity to their homes in a high-choice market might mean that a.) transportation is a significant barrier to accessing the school choice market or b.) school siting procedures are working and that schools are being located in close proximity to student housing. The answer is probably a bit of both; however, given the significant proportion of students living relatively close to their chosen schools, it does seem reasonable that transportation alone is not a solution to improving school access.
Additionally, we found a direct relationship between students’ commutes and students’ achievement on the M-Step and ACT. This relationship suggests that access to transportation has some degree of impact on learning. While the change was consistent across various distances, it was very slight. Much more inquiry is required to understand whether and how commutes impact learning.
We do need to know more about Detroit families’ school choice decisions and the degree to which transportation may be a barrier to accessing schools. Stories about individual students are insightful because they can help inform further inquiry. But policy makers should be careful not to paint with too broad a brush.
Whether it’s a park, fire station, hockey stadium, or bus stop, public service planning works to ensure equitable access. The benefits associated with these community assets are expected to be consistent across a population, but some will always hold more value than others. Some love hockey and see great value in living close to the arena, while others find the traffic congestion a nuisance. No school is the same. Schools, like the teachers within them, are not widgets. In a choice environment, a school’s value is based on family preferences that inevitably vary. Understanding how a population best enjoys access to a service is straightforward if everyone values it equally. But the realities of changing politics, dynamics of population distribution, and changes in transportation technologies all mean that static school location strategies can only be optimal in the short run.
Grand Valley State University’s Charter Schools Office (Detroit’s largest charter school authorizer) uses a three phase review process for new school applicants. Proposed school locations are considered for their proximity to other schools, the performance and enrollment of nearby schools, and a myriad of community data indicators including crime, blight, and housing data. School siting processes in a choice market require that we distinguish between need and demand for school access. While all students deserve and need a high-quality school option, demand for different types of schools will naturally vary in a marketplace. The key is ensuring that everyone has equal access to schools.
Detroit’s transportation system was designed for a pre-choice system of schools, at a time when Detroit Public Schools enrolled the majority of students. The case for a robust, centralized transportation system is compelling if a large number of students are commuting significant distances. But the 2013–14 Data Driven Detroit analysis, along with our brief analysis, confirms that very few students across all grade levels are traveling more than ten miles to the schools of their choice. To be sure, most students are traveling less than three miles; however, three miles is much farther than the typical .0625 mile threshold for walking to school in the inner city. The tragic reality of Detroit’s 38 percent poverty rate means that transportation is a barrier keeping too many of the city’s residents from accessing the full marketplace of school choice options.
The idea of a centralized transportation system, complete with big yellow school buses running in grid patterns in the morning and afternoon, ignores the realities of Detroit’s governance and funding—to say nothing of the preferences of Detroit families. Students’ changing enrollment and housing patterns are fluid and difficult to constrain through policy. Fortunately, that nostalgic vision also ignores real (and inexpensive) technological advancements like geolocation and real-time data collection that can provide the foundation for an innovative, need-based transportation system to meet the needs of the most impoverished Detroiters. We have the opportunity to incentivize access to great schools and unleash the entrepreneurial spirit and soul of Detroit.
Rob Kimball is the deputy director of GVSU’s Charter Schools Office and a doctoral candidate at Michigan State University. He is also the leader of EdLabs@GVSUCSO, which facilitates research-based inquiry within GVSU charter schools to further their oversight and support.
