- Publications
-
Research Programs
- Search Funded Research Grants and Contracts for NCER Career and Technical Education Civics Education and Social Studies Cognition and Student Learning Continuous Improvement Research in Education Early Learning Education Leadership Education Technology Effective Instruction English Learners Evaluation of State and Local Programs and Policies Improving Education Systems Literacy Methods Training for Education Research Pathways to the Education Sciences Research Training Postdoctoral Research Training
- Postsecondary and Adult Education Predoctoral Research Training Reading for Understanding Research Initiative Researcher-Practitioner Partnerships in Education Research Scalable Strategies to Support College Completion Science, Technology, Engineering, and Mathematics (STEM) Education Small Business Innovation Research Social/Behavioral Context for Academic Learning Statistical and Research Methodology Statistical and Research Methodology in Education - Early Career Supporting Early Learning Systematic Replication Training in Education Use and Practice
- Funding Opportunities
- News & Events
- About Us
Partially Nested Randomized Controlled Trials in Education Research: A Guide to Design and Analysis
In some tests of educational interventions, individual students are randomized directly to the treatment or control group, and both intervention and control protocols are administered in an individual setting. Such an experiment is an Individual-Level Randomized Controlled Trial (I-RCT). In other tests, clusters of students (e.g., classrooms) are randomized. This sort of experiment is called a Cluster Randomized Controlled Trial (C-RCT). However, in some designs, students in the treatment group are clustered like those in a C-RCT, but students in the control group are unclustered, like students in an I-RCT. This design is called a Partially Nested Randomized Controlled Trial (PN-RCT). It is partially nested because students in the treatment group are nested in some higher level unit, such as a tutoring group or class, but students in the control group are not nested as part of the experimental design.
This paper, commissioned by the National Center for Education Research, provides readers with an introduction to PN-RCTs and ways to design and analyze the results from them. This paper was written primarily for applied education researchers with introductory knowledge of quantitative impact evaluation methods. However, those with more advanced knowledge will also benefit from some of the technical examples and appendices.
- Chapters 1 and 2 define PN-RCTs and address design issues such as possibilities for random assignment, cluster formation, statistical power, and confounding factors that may mask the contribution of the intervention.
- Chapter 3 discusses basic statistical models that adjust for the clustering of treatment students within intervention clusters, associated computer code for estimation, and a step-by-step guide, using examples, on how to estimate the models and interpret the output.
- Chapter 4 and the technical appendixes discuss more advanced statistical topics pertaining to PN-RCTs.
