Exploring Small Sample Research Methods for Clustered STEM Data

There continues to be increasing demand for more comprehensive assessments of the impacts of STEM education programs, policies, and practices. One stumbling block to such research is the need for large sample sizes for studies to be able to make the intended claims. This research develops, and tests new estimation procedures for cross-classified structural equation models for the statistical analysis of small to moderate sample size studies of teaching and learning in mathematics. These new tools reduce the burden of large sample sizes. The team will develop new shareable software tools (in an open-source software application) and deliver training opportunities to researchers. Further, these estimation models and research tools will have broad impact as they are usable for similar study designs in all STEM content learning settings. This research bridges the gap between current research methods and the scale with which most STEM education studies are conducted. The team will develop new estimation methods for small to moderate sample cross-classified structural equation models that are tightly attuned to the features of teaching and learning research but broadly applicable across areas of STEM education. The research integrates four components. The research study (a) derives a new statistical framework and estimators, (b) develops empirical guidelines for study design with the proposed methods using a rich survey of teaching and learning programs and teacher development processes, (c) develops free and accessible software and tutorials, and (d) conducts highly accessible workshops to improve the capacity of the field. This research project is supported by NSF's STEM EDU Core Research (ECR) program. The ECR program emphasizes fundamental STEM education research that generates foundational knowledge in the field. Investments are made in critical areas that are essential, broad, and enduring: STEM learning and STEM learning environments, broadening participation in STEM, and STEM workforce development. The program supports the accumulation of robust evidence to inform efforts to understand, build theory to explain, and suggest intervention and innovations to address persistent challenges in STEM interest, education, learning and participation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.