The algebra-based introductory statistics course has seen tremendous growth in enrollments over the last two decades using a consensus curriculum and sequencing of topics. However, research has also shown students typically leave these courses with a shallow understanding of key inferential ideas. Recently, many statistics educators have proposed moving from this traditional curriculum to one centered on computer-intensive, randomization-based inference methods. Two advantages of this approach are: (1) randomization methods enable students to focus on the core logic of inference, and (2) efficiency in presentation allows students to gain experience in computer-intensive and multivariable methods that are being increasingly used by applied researchers. This project is providing instructors with a fully integrated set of curriculum materials with which to teach a substantially different curriculum that introduces statistical inference from the start. The materials are undergoing class-testing at numerous institutions and being disseminated through publication as a textbook, workshops, and presentations. The accompanying evaluation component is providing information about potential gains in student understanding of core concepts of inference and documentation of how students develop skills of inferential reasoning. These curricular materials and assessment findings have the potential for effecting a substantial change in the content and focus of introductory statistics courses across the country.

Project Report

The traditional approach to teaching introductory statistics includes material on probability and sampling distributions which is abstract and challenging for students. We have developed an alternative approach, using simulation and randomization methods, which makes use of tactile simulations and freely available javascript applets to model random chance in a way that is natural and intuitive for students. This approach is integrated in a new version of the commonly taught algebra-based introductory statistics course. The new curriculum also focuses students on the broader questions of the logic and scope of statistical inference (drawing conclusions from data) primarily utilizing published scientific research to motivate students’ conceptual learning. Finally, building on this stronger foundation, we’ve developed materials for a second course in statistics which bridges students to understanding multivariable statistical methods. The pedagogy for the curriculum involves an emphasis on conceptual understanding and active learning, with key concepts emphasized using a spiral approach which revisits concepts throughout the curriculum. To date our curriculum has been class tested with over 2600 students, at 18 institutions with 28 different instructors with the potential for many more as the text (Introduction to Statistical Investigations, by Tintle et al.) is published with an international publishing company (John Wiley and sons) in the upcoming year. Assessment data shows numerous areas of enhanced student outcomes compared to the traditional curriculum, with little to no areas where performance is worse. Descriptions of the curriculum and assessment results have been and will continue to be written up in scientific journal articles and presented at conferences worldwide. Furthermore, over the next few years we will offer numerous workshops training faculty on use of the materials, develop an online support/learning community for faculty using the materials and undergo a substantial, multi-institution assessment project as part of a recent, separate NSF-grant award.

National Science Foundation (NSF)
Division of Undergraduate Education (DUE)
Standard Grant (Standard)
Application #
Program Officer
John Haddock
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Dordt University, Incorporated
Sioux Center
United States
Zip Code