Grant Open

Multilevel Regression Discontinuity Design With Latent Variables

NCER

Program: Statistical and Research Methodology in Education

Program topic(s): Core

Award amount: $883,198

Principal investigator: Ji Seung Yang

Awardee:

University of Maryland, College Park

Year: 2022

Award period: 4 years (06/01/2022 - 05/31/2026)

Project type:

Methodological Innovation

Award number: R305D220030

Purpose

A regression discontinuity (RD) design is often employed to provide causal evidence when a randomized control trial is practically infeasible or unethical. Conventional RD models assume that all running variables, covariates, and outcomes are observed variables. The purpose of this grant is to extend the modeling framework to augment the structural RD model with multilevel latent variable measurement models for any or all of the variables.

Project Activities

The research team will first develop the models and their causal estimands. They will then develop software for estimating model parameters and conducting diagnostic tests of the assumptions made by the models. After the software is ready, the researchers will run a real-data demonstration of the RD model with multilevel latent variable measurement models, using data from the Oregon English Language Learners Program and the Baltimore City Public Schools Gifted and Advanced Learning Program.

People and institutions involved

IES program contact(s)

Charles Laurin

Project contributors

Yang Liu

Co-principal investigator

Products and publications

The research team will also develop a user-friendly version of the software in R, with supporting instructional materials. Other products from the grant will include workshops for using the software and peer-reviewed journal manuscripts in methodological and applied journals.

Publications:

Han, Y., Yang, J. S., & Liu, Y. (2024). Assessing Item Fit Using Expected Score Curve Under Restricted Recalibration. Journal of Educational and Behavioral Statistics, 10769986241268604.