Wendy Chan is an expert on educational statistics. She seeks to improve statistical models and analysis so research can better inform policy and practice. Her work tries to unpack the intellectual and statistical challenges of generalizing the results of localized randomized trials to larger populations. She began her career in education as a member of Teach for America, where she taught sixth- and eighth-grade mathematics in a large middle school in New York City.

Professional Biography

Wendy Chan is an assistant professor of education in the Human Development and Quantitative Methods Division at Penn GSE. Dr. Chan’s research focuses on applied statistical methods to improve generalizations from small studies in education.

Dr. Chan received her Ph.D. in statistics from Northwestern University, where she was a graduate research assistant for the Institute for Policy Research. She began her career in education as a member of Teach for America, where she taught sixth- and eighth-grade mathematics in a large middle school in New York City. Her work has appeared in the Journal of Research on Educational Effectiveness, the Journal of Educational and Behavioral Statistics, Psychological Methods, and Evaluation Review.

Research Interests and Current Projects

Dr. Chan specializes in applied educational statistics, and her research projects and interests are at the leading edge of work on statistics methods in field contexts, including scaling up interventions. Her current projects assess the extent to which generalizations can be made from small studies and they consider the advantages and tradeoffs of different estimation approaches in the presence of limited data. In another project, Dr. Chan considers the impact of violations in model assumptions on the validity of estimates in generalization studies. In this work, she assesses the effectiveness of robust estimation approaches in improving the precision and bias of estimates when core model assumptions are violated. In addition to her work in generalization, Dr. Chan has also worked on research in experimental design and examined corrections to test statistics in the presence of clustering and treatment effect heterogeneity.

Publications

Selected Publications

Chan, W., & Oh, J. (2023). Robust small area estimation for generalization. Journal of Experimental Education, 91(3), 539–556.

Chan, W., & Hedges, L. V. (2022). Pooling interactions into error terms in multisite experiments. Journal of Educational and Behavioral Statistics, 47(6), 639–665.

Chan, W., Oh, J., & Luo, P. (2021). The implications of population changes on generalization and study design. Journal of Research on Educational Effectiveness, 14(2), 471–500.

Chan, W. (forthcoming, 2021). Analysis of variance. In R. Coe, M. Waring, L. V. Hedges, & L. D. Ashley (Eds.), Research methods & methodologies in education (3rd ed.). SAGE Publishing.

Chan, W. (2022). The role of distributional overlap on the precision gain of bounds for generalization. American Journal of Evaluation, 43(1), 90–107.

Chan, W., Hedges, L. V., & Hedberg, E. C. (2020). The impact of downstream clustering in longitudinal studies. Journal of Experimental Education.

Chan, W. (2019). An evaluation of bounding approaches for generalization. Journal of Experimental Education. Advanced online publication. https://doi.org/10.1080/00220973.2019.1697191

Chan, W. (2019). The relation among design parameters for statistical power between continuous and binomial outcomes in cluster randomized trials. Psychological Methods, 24(2), 179-195.

Chan, W. (2018). Applications of small area estimation to generalization with subclassification by propensity scores. Journal of Educational and Behavioral Statistics, 43(2), 182–224.

Chan, W. (2017). Partially identified treatment effects for generalizability. Journal of Research on Educational Effectiveness, 10(3), 646–669.

Tipton, E., Hallberg, K., Hedges, L. V., and Chan, W. (2016). Implications of small samples for generalization: Adjustments and rules of thumb. Evaluation Review (Special Issue: External Validity 2), 41(5), 472–505.

Kang, J., Chan, W., Kim, M. O., & Steiner, P. M. (2016). Practice of causal inference with the propensity of being zero or one: Assessing the effect of arbitrary cutoffs of propensity scores. Communications for Statistical Applications and Methods, 23(1), 1–20.

News & Media

Penn GSE News

Does this program work? Ask Wendy Chan

Assistant Professor Wendy Chan, an applied education statistician in Penn GSE’s Human Development and Quantitative Methods Program, helps educators and policy makers trust the findings generated by the tools and techniques of education research.

February 19, 2018

Awards & Honors

Wendy Chan was awarded $25,000 by the American Educational Research Association for the project “The Generalizability of Deeper Learning Using Small Area Estimation.” In an extension to the earlier Study of Deeper Learning by the American Institutes for Research, Dr. Chan will use hierarchical Bayesian models to estimate the generalized impact of the deeper learning approaches on the proportion of on-time graduation rates among high school students in California and New York. This study is the first to generalize the impacts of deeper learning and to apply small area estimation methods to generalization when the small areas are defined using a generalizability index.

Posted on November 25, 2019

Wendy Chan was awarded a grant from the Children’s Hospital of Philadelphia for $21,456 for the project “Reducing Disparities in Behavioral Health Treatment for Children in Primary Care.” This study assesses the impact of a behavioral therapy program on patient-centered outcomes for children with attention-deficit/hyperactivity disorder (ADHD). The study will focus primarily on low income households.

Posted on August 7, 2019

Wendy Chan has been named a recipient of the Association for Public Policy Analysis and Management's 40 for 40 Fellowship. The fellowship program provides funding for 40 outstanding early career professionals and supports scholars in growing their professional networks and sharing their research with the broader policy audience.

Posted on March 13, 2019

Wendy Chan’s project, "Assessing the Role of Covariates and Matching in Improving Bounds for Generalization and in Understanding Generalizations Over Time," was recently awarded $36,033.50 from the Spencer Foundation. The goal of this project is two-fold. First, the project investigates the impact of covariates and matching methods on improving the precision of bound estimates of population average treatment effects. Second, it explores the extent to which generalizations from experimental studies are made to populations defined at a future time point.

Posted on January 17, 2018

Journal Editorial Boards

Journal of Educational and Behavioral Statistics

Editorial Board