Author Date


Degree Name





Physical and Mathematical Sciences

Defense Date


Publication Date


First Faculty Advisor

Lynne Nielsen

First Faculty Reader

Ken Plummer

Honors Coordinator

Del Scott


introductory statistics education, statistical procedure selection


Researchers in multiple industries (biomedicine, engineering, etc.) cite the selection of an appropriate statistical test as a common problem. Experts draw on a framework of conceptual and procedural knowledge to navigate when to use statistical methods. Students also struggle determining the correct statistical method to use for a given research question. This is because they lack the opportunity to practice recognizing a host of features in each research question that provide clues for experts as to which method is most appropriate. “Decision Based Learning” (DBL) is a teaching method designed to help teachers and students address this struggle. In this study we create and implement a decision model for choosing which of the 14 statistical methods, taught in an introductory education statistics course, is most appropriate for the research questions. We compared the performance on method selection test questions of 1021 students using DBL and a comparison group of 930 students during the Winter semester of 2021 at Brigham Young University. The treatment and comparison groups were composed of randomly assigned sections of students. The performance on examinations lacks sufficient evidence to show that DBL significantly improved method selection capabilities. The recommendations in this study provide potential improvements in the presentation and delivery of decision-based learning for introductory education statistics.