Public Transit and School Choice in Philadelphia: Exploring Spatial Equity and Social Exclusion

• In Event: 47.061 - Geographic School Choice Research Across the Local-Global Scale

Abstract

This paper makes the case that in urban areas, mode of transportation matters for calculating distance and commute. Some research measures commute in terms of miles (e.g., Hamlin, 2017), and other studies use drive time (e.g., Glazerman & Dotter, 2017). However, transportation scholars indicate that parents choose different transportation modes based on school type and ethnic background (Wilson, Marshall, Wilson & Krizek, 2010). Using Philadelphia as a case study, we explore potential differences in the options available between driving and using public transportation. We then examine if spatial inequity exists in the Philadelphia choice sets.

This study is framed in the tenets of spatial equity. Space is vital in the construction of opportunity (Gulson & Symes, 2007). Furthermore, urban planning scholars maintain that commute is linked to social capital. Lucas (2011) argues that transportation systems are inherently inequitable. This leads to a process of social exclusion, mechanisms used to institutionally control access (Madanipour, 1998/2016). Even in large cities, not everyone is networked to public transportation (Lucas, 2011). Thus, we maintain that if Philadelphia students do not have equitable access to public transportation, then their school options will be limited.

We calculate the travel time between a zoned neighborhood high school and every public school serving students in grades 9-12 in Philadelphia for both driving and using public transit. Given that we do not have individual student addresses, we cannot construct individual choice sets and thus utilize neighborhood school boundaries to make comparisons.

Our first question considered if there was a statistically significant difference between driving and public transportation. We find that there is a significant difference in commute time between public transit (μ=46.86, σ=20.37) and car (μ=24.82, σ=11.05); t(1721)=70.61, p < 0.001. Given that most secondary students use public transit to commute (Redfern, 2013), we argue that this is the most appropriate analysis to determine choice sets.

The second question asks if there is spatial equality and/or equity between school choice sets. To do this, we model the differences between choice sets. The null hypothesis of the two equality models is that there is no statistical difference in the quality of schools between catchment service areas. Similarly, the null hypothesis of the equity models is that catchment area poverty for students in a given boundary is not significant in determining choice set quality.

We fail to reject the null hypothesis of all four models. Because they each approach an abstract conception of choice sets, we then calculated the marginal predicted score for each catchment area. The variances of the marginal predictions are reduced by about three points when controlling for catchment area poverty, suggesting that sets are more equitable than equal. Thus, the argument that spatial equity exists for Philadelphia students appears valid. Yet, the overall school quality for each catchment area still may function as a mechanism of social exclusion. The mean marginal scores of quality (as calculated by the Philadelphia School District) for the four models range between 27.52% and 32.81%.

Authors

• Michael R. Scott, The University of Texas at Austin

• David T. Marshall, Auburn University