Part I
This introductory webinar on March 19 at 12 p.m. ET equips S-STEM PIs and evaluators with essential design thinking to move from descriptive outcomes to stronger, defensible evidence of program impact. The webinar introduces the foundations of quasi-experimental designs by explaining when and why they are used and the kinds of real-world problems they are meant to solve. It highlights common threats to validity in single- and multiple-group studies and focuses on the central challenge of selection bias—how to think carefully about what would have happened in the absence of a program or intervention. The session then walks through practical examples of comparison groups commonly used in education settings, emphasizing how thoughtful design choices strengthen conclusions and could be applied to the evaluation planning for S-STEM PIs and their evaluators.
Part II
The second webinar builds on the introductory session by guiding PIs and evaluators through the analytic techniques needed to strengthen quasi‑experimental designs when random assignment is not possible. This session provides a practical overview of propensity score methods, including matching, weighting, and model diagnostics, to help participants create more credible non‑equivalent comparison groups for S-STEM scholar program evaluations. Attendees will learn some of the technical aspects of these methods for reducing bias, improving balance, and enhancing the rigor of their impact findings.
