Preferences, Proximity, and Controlled Choice: Examining Families' School Choices and Enrollment Decisions in Louisville, Kentucky

• In Event: Tensions Between School Choice and Neighborhood Schools in the Changing City

Abstract

School choice has grown rapidly in the last several decades, as many local districts have responded to federal and state incentives along with parental pressure to offer some type of intradistrict choice options. Concurrently, federal courts have limited what types of policies districts could voluntarily implement to try to maintain diverse schools. Research confirms that access to diverse schools is an important educational resource for all students, but especially for students of color (see Linn & Welner, 2007; Mickelson & Nkomo 2012).

This paper explores competing tensions of choice and proximity in Jefferson County Public Schools (JCPS) with a new generation of controlled choice policies post-Parents Involved:

1) Are there differences by student characteristics in who gets their school choices under the new student assignment policies?
2) How does receiving a top school choice predict who enrolls?

About half of JCPS’s nearly 100,000 students are white; black students comprise the largest minority group at 37%. 60% of students receive free or reduced price lunch.

I analyze de-identified student level data provided by JCPS from students’ applications and subsequent enrollment. These data are supplemented by NCES Common Core of Data to provide information on school demographics. I calculate descriptive statistics and test for significance to examine the percentage of subgroups who selected their nearest school as their first-choice, were assigned to their first-choice school, that enroll, and in the types of school choices. I estimate logistic regression models predicting (1) whether students were granted their first choice or not and (2) whether individual characteristics or receipt of first choice school make students likelier to enroll. Independent variables include student characteristics such as race, free/reduced lunch status, and diversity code (Jefferson County’s block group diversity metric).

Slightly over half of JCPS applicants selected their closest school as their first choice in 2011, although this varied by student characteristics with white and middle-class students more likely to select their proximity school. White and/or advantaged area students were also more likely to get their first choice school generally and if they chose their closest school; the district has deep residential segregation (Author, 2017) meaning that these choices may, within district guidelines, reflect different student body composition.

While 97% of applicants subsequently enrolled in JCPS, descriptive statistics suggest differential uptake of school assignments. 90% of White students or those living in the most advantaged neighborhoods subsequently enrolled. By contrast, 99.6% of applicants in the most disadvantaged neighborhoods enrolled.

Logistic regression confirmed these descriptive statistics, finding that students of color (except Asians) and low-income students were significantly less likely than white students and middle-class students, respectively, to receive their first choice school. Moreover, receipt of first-choice school was a significant predictor of whether students who applied ultimately enrolled in the district.

This study helps us to understand how a choice-based integration policy with a weak geographic preference may still advantage those who choose their nearest schools as well as white students and those living in more advantaged areas.

Author

• Erica Frankenberg, The Pennsylvania State University