The NSF CAREER project, Statistical Power Analysis and Optimal Sample Size Planning for Longitudinal Studies in STEM Education, develops statistical frameworks, software tools, design parameter estimates, examples, training materials, and workshops to help researchers plan longitudinal studies in STEM education.
Longitudinal studies are central to education research because many interventions unfold over time. Researchers may care about effects at a specific time point, average effects across time, or changes in effects over time. This project addresses a practical design challenge: how to plan longitudinal experimental and quasi-experimental studies with adequate statistical power under realistic design and budget constraints.
Within the lab, this is the central project for longitudinal design, power analysis, and optimal study planning. It also connects to applied Lastinger Center projects that provide realistic longitudinal education data for design-parameter estimation and methods illustration.
Project Information
- Funding agency: National Science Foundation
- Program: Faculty Early Career Development Program, EDU Core Research
- Full project title: CAREER: Statistical Power Analysis and Optimal Sample Size Planning for Longitudinal Studies in STEM Education
- Award ID: DRL-2339353
- Principal Investigator: Wei Li
- Institution: University of Florida
- Project period: August 1, 2024-July 31, 2029
- Award amount: approximately $1.25 million
Project Goals
- Develop statistical frameworks for power analysis and optimal sample size planning in longitudinal experimental and quasi-experimental studies.
- Compare design and analysis strategies so researchers can select appropriate longitudinal designs under practical constraints.
- Estimate design parameters using data from prior and ongoing STEM education studies.
- Develop software tools, documentation, examples, training materials, and workshops for broad use by education researchers.
Links
Related Publications
- Li, W., & Konstantopoulos, S. (2023). Power analysis for moderator effects in longitudinal cluster randomized designs. Educational and Psychological Measurement, 83(1), 116-145. doi:10.1177/00131644221077359
- Li, W., & Konstantopoulos, S. (2019). Power computations for polynomial change models in block-randomized designs. Journal of Experimental Education, 87(4), 575-595. doi:10.1080/00220973.2018.1496057
- Li, W., & Konstantopoulos, S. (2017). Power analysis for models of change in cluster randomized designs. Educational and Psychological Measurement, 77, 119-142. doi:10.1177/0013164416641460
- Li, W., Xie, Y., Pham, D., Dong, N., Spybrook, J., & Kelcey, B. (2024). Design and analysis of cluster randomized trials. Asia Pacific Education Review, 25(3), 685-701. doi:10.1007/s12564-024-09984-z
Related Presentations
- Li, W., Huang, J., Zhang, Q., Konstantopoulos, S., & Strickland, K. J. (2025). Using real-data simulation methods to estimate statistical power for longitudinal experimental studies. SREE Annual Meeting.
- Li, W., Konstantopoulos, S., & Shen, Z. (2024). Design and analysis of longitudinal multisite randomized trials: Estimation, statistical power, and optimal sample size. SREE Annual Meeting.
- Li, W., & Kotsiopoulos, S. (2024). Power analysis for main and moderator effects in multisite longitudinal experiments with site fixed effects. AERA Annual Meeting.
- Li, W. (2022). Design parameters for planning longitudinal experiments in education. SREE Annual Meeting.
Related Projects
- Enhancing Longitudinal Program Evaluations
- Propensity Score-Based Quasi-Experimental Study Planning
- Causal Discovery for Online Learning Platforms
Software and Training Materials
Software tools and training materials from the CAREER project are being developed for public use. Planned outputs include tools for power analysis and sample size planning in longitudinal studies, including R-based and interactive resources.
See also the lab’s resources page.
NSF Acknowledgment and Disclaimer
This material is based upon work supported by the National Science Foundation under Award No. DRL-2339353. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.