Project Activities
In Year 1 (2022–23), the research team will finalize district and school recruitment, establish district data sharing agreements, and collect baseline. In Year 2 (2023–24), the research team will set the first cohort sample of 7th graders, survey the students, and conduct random assignment. Students will complete self-affirmation or comparison group writing exercises. Teachers will hand out exercises to all students in their 7th grade ELA classrooms, with both teachers and students blind to condition. After each exercise, the research team will collect short fidelity surveys from each teacher, and site liaisons will collect student exercises, who will ship them to the research team. In Year 3 (2024–25), the team will follow the same procedures for a second cohort of 7th graders. In Year 4 (2025–26), participating schools will implement the SA intervention under normal conditions without researcher support and collect implementation data. Each summer and fall except Year 5 (2026–27), the research team will collect districtwide student administrative records data. Implementation and impact analyses will begin in Years 2 and 3, with dissemination of results of the implementation and impact analyses for each year from fall of 2024 to June of 2027. They will conduct a cost study in Years 2–4 of the project, as well as for the entire project.
Structured Abstract
Setting
The study will take place in eighty middle schools from approximately 13 districts from California, Florida, North Carolina, New York, Nevada, and Virginia across four racial density strata.
Sample
Participating students will include approximately 16,164 Black and Latinx 7th grade students.
The self-affirmation intervention consists of four 15-minute writing exercises implemented throughout the year that prompt students to write about their values. They begin close to the start of school and are timed to precede potentially high stress events like mid-terms or end-of-course exams.
Research design and methods
The researchers will conduct four concurrent RCTs—one for each racial density stratum of schools in a given range of racial composition. Following previous research, the primary academic outcome will be GPA, and the primary disciplinary outcome will be number of suspensions. Each RCT is designed to meet WWC standards with a design that allows for internally valid causal inferences as well as low attrition rates. It applies a blocked (within school) student level random assignment for 7th graders to generate intent-to-treat (ITT) impact estimates across two successive cohorts of 7th graders. It relies primarily on outcomes from administrative student records. It also includes a student survey collected within classrooms at the beginning and end of the school year of implementation for each cohort. Data collection is consistent and simultaneous for treatment and comparison groups, and consistent within classrooms and across schools. The research team will use district wide administrative data collection to follow students across years even if they change schools (within the same district). Therefore, the team expects low levels of attrition and lower levels of differential attrition, even across school years. They will use baseline data collected prior to random assignment to ensure there are no systematic differences between treatment groups.
Control condition
Control students receive similar exercises at the same time in the same classrooms, prompting them instead to write about values that might be important to others.
Key measures
SA implementation measures include the number of exercises completed by student and the extent to which students actually self-affirm in their written response. Outcome measures include, GPA, numbers of Fs and D, z-scores of standardized reading and math grades, referrals, suspensions, and social-emotional outcomes including, social belonging, school trust evaluation anxiety, and stereotype activation.
Data analytic strategy
Intent to treat impact estimates estimated in a multi-level model with students nested within schools, controlling for student and school level baseline characteristics. Moderator and mediator analyses using interactions and structural equation modelling
Cost analysis strategy
The research team will identify cost per participant using the "ingredients method" and will include all expenditures on things such as personnel, facilities, equipment, materials, and training, and apply a market rate to each ingredient. The research team will estimate cost-effectiveness by comparing the cost per participant to the average impact per participant on each measure.
People and institutions involved
IES program contact(s)
Project contributors
Products and publications
Products: The research team will produce conference papers and presentations for research and practitioner conferences, including those with an emphasis on rigorous causal studies and those with a focus on racial equity. The project team will produce implementation and impact reports for results in grades 7, 8 and 9 for broad audiences, and for submission to peer reviewed journals. They will also produce practitioner friendly research briefs and conduct webinars designed to reach practitioners and policy makers with decision-making authority.
Related projects
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
