Co-Principal Investigator: Stephens, Kristen; Henfield, Malik

Purpose: The goal of this project is to examine whether gifted education policies and programs in North Carolina contribute to beneficial academic and social-emotional outcomes. Through an in-depth examination of malleable factors that may moderate student success, the research team will examine what types of gifted education policies work, for whom, and under what conditions by asking: 1) Is gifted program participation associated with beneficial academic and social-emotional student outcomes?; 2) Which students benefit most from participation in gifted programs?; and 3) Which malleable factors mediate and/or moderate the effects of gifted education participation (and nonparticipation) on student outcomes? The research team will focus on the experiences of underserved (black, Hispanic/Latino, and economically disadvantaged) gifted students.

Project Activities: This project includes four distinct, yet inter-related studies. In Study 1, the research team will complete a content analysis of school district gifted education plans. This will inform the regression discontinuity (RD) design for Study 2, which will use public school data from the state to estimate the impact of being identified as gifted on students' test scores, test score growth, course selection, and high school graduation. In Study 3, the research team will use a state value-added data system to identify schools that demonstrate both academic growth among gifted students and an equitable representation of student subgroups participating in gifted education. In Study 4, researchers will collect qualitative data from students, teachers, and counselors at the schools identified in Study 3 to understand how they achieve both high academic growth and equity.

Products:The research team will provide preliminary evidence of what types of gifted education policies work, for whom, and under what conditions. Researchers will also produce peer-reviewed publication(s).

Structured Abstract

Setting: The research will take place in North Carolina, including rural, urban, and suburban settings.

Sample: Study participants include teachers, counselors, and gifted students at approximately 50 schools (elementary, middle, and high) across six school districts representing each of the organizational regions defined by the state's education department.

Intervention:Researchers will examine the association of gifted education policies and practices with student outcomes.

Research Design and Methods:The research methods vary across the four studies. Study 1 will complete a content analysis of district gifted plans. Study 2 will use a regression discontinuity (RD) and multi-level propensity score matching (PSM) design to estimate the impact of being identified for gifted services on a range of outcomes (see Key Measures below). Study 3 will use North Carolina's Education Value Added Assessment System (EVAAS) to create typologies to identify schools based on the efficacy of each school in supporting academic growth of gifted students, and equitable representation of students participating in gifted education. Study 4 will collect qualitative data using interviews, focus groups, surveys, and classroom observations at schools representing each of the typologies identified in Study 3.

Control Condition:In the RD design for study 2, the control group is comprised of students whose scores were just below the cutoff for identification as gifted. The other exploratory studies do not have control groups.

Key Measures: Measures will vary across the four studies. Researchers will develop codes for the content analysis for Study 1. For Study 2, student level measures in the state data include 1) test scores on a state-administered test; 2) test score growth on a state- administered test; 3) course selection; 4) and high school completion rates. School-level variables in the state data for Study 2 include SES (high, middle, low), gifted program ethnic minority representation, whole school ethnic minority representation, school size (large, medium, and small), and mean student achievement. Study 3 uses a composite index of growth for gifted sub-groups at each school. Study 4 uses structured interviews and focus groups with a purposive representative sample of teachers, counselors, and students; classroom observations of engagement and rigor to be developed by the researchers; and student surveys of ethnic identity and academic self-concept.

Data Analytic Strategy: The data analytic strategy varies across the four studies. The team will carry out Qualitative Content Analysis in Study 1. In Study 2, researchers will use an RDD (a quasi-experimental statistical technique to estimate a causal impact) using state administrative data and multi-level PSM. For Study 3, the team will create efficacy-equity typologies using value-added growth scores created by the state for each school and descriptive data about each school. Finally, in Study 4, the research team will import and analyze all data from individual interviews, focus groups, documents, student surveys, and observations into NVivo and will complete content analyses of that data.