Co-Principal Investigators: Ream, Robert; Santibanez, Lucrecia

Purpose: The purpose of this project is to explore the incidence and consequences of non-promotional student mobility and how it influences and interacts with student outcomes—achievement, behavior, social-emotional learning and other forms of development, such as language acquisition and disability status—for vulnerable populations in California. Non-promotional student mobility is defined as students changing schools either mid-year or from one year to the next in grades that do not correspond to the terminal grade of the school. Vulnerable populations are defined as economically disadvantaged students, English learners, students with disabilities, homeless youth, and foster youth. Student mobility, unlike residential mobility, occurs in the public realm, so that its potentially negative effects are more malleable and amenable to policy solutions.

Project Activities: The research team will explore a unique longitudinal student-level data set coming from the Project CORE initiative, a group of districts that have formed to collaborate on data collection and analysis for continuous improvement. These districts have embraced systematic measurement of social-emotional learning and school culture and climate in systems of school accountability. The research team will use the CORE data to conduct exploratory analyses of the incidence and consequences of student mobility for vulnerable student populations.

Products: The research team will share findings through annual CORE district briefings. They will develop a short policy brief for education practitioners and policy makers. They will give presentations at research and policy conferences and will also produce manuscripts for publication in peer-reviewed academic journals.

Structured Abstract

Setting: The study will use longitudinal data collected in the CORE districts (the California districts of Los Angeles, Oakland, Fresno, Long Beach, Garden Grove, and Santa Ana). The CORE districts include primarily urban schools but also contain some suburban and rural schools.

Sample: The sample comprises all K to 12 students (about 900,000) in six CORE districts in California. The longitudinal sample for this study includes several million student-level observations from the 2010-2011 school year through the 2016-2017 school year.

Malleable Factors: For this exploration study, malleable factors include school policies (discipline and school choice), school culture and climate, and students' social-emotional learning skills and behavior.

Research Design and Methods: The research team will use secondary data to conduct exploratory analyses of the incidence and consequences of student mobility for vulnerable student populations. They will use descriptive statistics to understand mobility patterns in this sample. They will track student outcomes before and after moves, distinguishing among moves of different types—structural, non-structural, voluntary, involuntary—comparing school changers to stayers. Since mobility could have positive effects in some situations and negative effects in others, the research team will account for the timing and reasons why students change schools and identify variation in the incidence and consequences of mobility across vulnerable and other demographic groups. They will also use information on the characteristics and climate of the schools students leave and enter and the degree to which mobile students "match" or fit in with their environments to determine the effect of school features on the outcomes of mobile students.

Control Condition: There is no control condition in this study.

Key Measures: The key measures for this study include information on students' school changes, the timing and frequency of the move, as well as information on students' demographics, achievement, social-emotional learning, behaviors (such as attendance and disciplinary actions), and English learner, homeless, foster youth, and disability designations. The CORE districts also collect data on the culture and climate of schools by grade.

Data Analytic Strategy: The research team will conduct descriptive analyses including means and frequencies of events, as well as inferential statistics such as t-tests and chi-square tests to assess whether any differences in mobility by subgroup are statistically significant. They will also fit discrete time hazard models to explore whether some groups of students are more likely to change schools and whether these moves are more likely to be involuntary.