The Multiple Dimensions of Student Mobility and Implications for Academic Performance: Evidence from New York City Elementary and Middle School Students

Abstract

The purpose of this report is threefold: First to develop measures of alternative types of student mobility; second to document the magnitudes of each type of mobility in aggregate and by student income, race/ethnicity, and immigrant status; and third to analyze how mobility of different types affects student academic performance. Although mobility is an oft discussed phenomenon, inadequate attention has been paid to the alternative ways that it can be defined, how the alternatives differ and which alternatives appear to be of sufficient size to be of consequence to policy and practice. Previous research on mobility often focuses on the impact of changing schools on an individual’s academic achievement. As an example, Hanushek, Kain and Rivkin (2004) find that switching schools is harmful to student growth in performance in the year following a switch, even when switching is required because students reach the top grade offered at their school. Other researchers find that academic performance is lower among students who have changed schools in previous years, whatever the reason, (Alexander et al, 1996; Rumberger and Larson, 1998; Swanson and Schneider, 1999). Relatively little attention has been paid in the existing quantitative analyses to distinguishing between different types of mobility i.e., midyear vs. between year; annual vs. cumulative. By providing district level statistics on alternative types of student mobility, this report may help policymakers decide which types of student mobility are important for districts to report regularly. Defining alternative measures may help both policymakers and researchers identify the types of mobility that are the most harmful to student performance and effectively design and target interventions. The main findings of this study are that there is considerable mobility into grades 2 through 8 from outside the New York City school district, across schools, across years for students staying in the district, and some mobility even across schools within academic years. Furthermore, over time, between 6% and 7% enter into each grade of a cohort, and students move several times over their schooling history in the city district. In addition, the entrants and frequent movers have characteristics that are generally associated with harder to educate children. Finally, student mobility has a consistently negative effect, ceteris paribus, on 8th grade reading scores, although the statistical significance of the effect is sensitive to the specification used in the analysis. The report is organized as follows. In the second section, we describe the sources of data. In this section, as throughout the report, we present more detailed information in a separate box in the text. The third section introduces alternative measures of student performance, the fourth section presents magnitudes of annual mobility, and the fifth section does a cumulative mobility analysis. The sixth section analyzes moves in terms of the characteristics of new schools and moves coincident with significant moves in student residence, as measured by zip code changes. The seventh section analyzes the effects of alternative measures of mobility on student performance for New York City (NYC) elementary and middle school (hereafter primary schools) students and the last section concludes.