Designing Statistics and Data Science Curriculum for Effective Teaching

Prince Afriyie Chair
University of Virginia
 
Tuesday, Aug 5: 8:35 AM - 10:20 AM
4089 
Contributed Papers 
Music City Center 
Room: CC-202C 

Main Sponsor

Section on Statistics and Data Science Education

Co Sponsors

Caucus for Women in Statistics
History of Statistics Interest Group

Presentations

A Design-Based Research Approach to Improving Statistics and Probability Teaching and Learning

The GAISE reports describe four phases of a statistical investigation: formulate a question(s) that can be answered with data, collect data, analyze data, and interpret the results of those analyses in the appropriate context. We have found that mathematics teachers almost always involve students in data analysis during statistics lessons. Yet, fewer than half of those lessons involve students in producing the data, and students are rarely involved in formulating questions or interpreting the results of data analyses.

The STEPSS program is a replacement unit designed to lessen the divide between the official curriculum and the enacted curriculum. Replacement units have two components: curriculum materials and teacher professional development. An experimental trial of STEPSS found that it improved statistics instruction and increased student understanding of statistics and probability. A replication study of STEPSS is underway.

Using a Design-Based Research methodology, the presentation will describe key components of the STEPSS program, provide the rationale for their inclusion, and discuss the program modifications that were made in preparation for the replication study. 

Keywords

statistics education

curriculum

randomized controlled trial

teacher education

design experiment

replacement unit 

First Author

Robert Schoen, Florida State University

Presenting Author

Robert Schoen, Florida State University

Beyond Data: Meet Researchers Featured in "Elementary Statistics: A Guide to Data Analysis Using R"

Introductory statistics courses often attract students out of necessity, leading to disengagement and low motivation. Since statistics is foundational to many scientific disciplines and fundamental to fields like artificial intelligence and machine learning, it is crucial for students to connect classroom concepts with real-world applications. "Elementary Statistics: A Guide to Data Analysis Using R" bridges this gap by presenting concepts like linear regression through tangible examples. In this session, attendees will meet the professors and researchers featured in this new textbook. These experts will share insights into their work and show how their research brings statistical concepts to life in an engaging way. By applying statistical ideas to real-world problems, this session aims to inspire students, educators, and researchers, highlighting how statistics is key to both academic and practical settings. The session will also highlight how the textbook creatively introduces R, a widely used tool for statistical computing. By integrating R with statistics, the textbook enhances learning, fosters critical thinking, and makes the study of statistics both relevant and engaging. 

Keywords

Statistics Education

R Programming

Machine Learning

Data Science

Elementary Statistics

Real-World Examples 

First Author

Nancy Glenn Griesinger, Huston-Tillotson University

Presenting Author

Nancy Glenn Griesinger, Huston-Tillotson University

Combining Statistics and Data Science in the New GAISE College Report

The GAISE College Report has been a cornerstone of undergraduate statistics education since its first edition was published in 2005. In this presentation, the two co-chairs of the revision steering committee will provide an update on the progress made in revising the statistics portion of the report. They will also discuss the data science section, outlining its intent and current status. 

Keywords

GAISE

statistics education 

Co-Author

Patti Lock, St. Lawrence University

First Author

Jamis Perrett, Brigham Young University

Presenting Author

Jamis Perrett, Brigham Young University

Combining Strengths: Turning strength in Statistics and Computing into a Data Science major

Many schools are looking for ways to create programs to attract students. In the Northeast US, this is often complicated
by enrollment challenges which have led to budget constraints and a lack of institutional willingness to pour resources into
the development of these programs without certainty of success. In this paper, we will explore how we met these realities
and were still able to develop a new data science major by building on institutional distinctives and strengths in statistics
and computing. This was a collaborative effort of faculty across disciplines, and led to a successful proposal for a new major
which we kicked off officially in the 2024-2025 academic year. We detail changes made to existing courses as well as what
new developments and costs were incurred. The hope is that our experience can be a model for others desiring to update and
modernize the offerings at their institutions while facing similar challenges. 

Keywords

Data Science

Curriculum

Statistics

Computer Science

Data Science Major

Data Science Minor 

Co-Author(s)

Nicholas Weaver, Messiah University
Jason Renn, Messiah University
David Owen, Messiah University
David Bibighaus, Messiah University
Devi Bhakta Suberi, Messiah University

First Author

Samuel Wilcock, Messiah University

Presenting Author

Samuel Wilcock, Messiah University

Integrating Teaching Applications in Undergraduate Introductory Statistics Courses

Robust teacher preparation requires attention to how preservice secondary mathematics teachers acquire statistical knowledge for teaching. For many preservice teachers, opportunities to learn about teaching statistics are limited to the introductory statistics course(s) that are required for their degree. We describe one model for how teacher preparation programs can integrate ideas about statistics and ideas about teaching statistics in undergraduate courses such as an introductory statistics course. This model makes use of teaching applications, mathematical or statistical tasks that attend to the dual goals of developing an understanding of mathematics or statistics content and practices and an understanding of some of the additional complexity involved when teaching mathematics or statistics to school students. We will provide examples of teaching applications and share research-based results from their use in introductory statistics courses across the United States. 

Keywords

Secondary teacher preparation

Statistics education

Introductory statistics course

Teaching applications 

Co-Author(s)

Elizabeth Burroughs
Bernard Yeboah, Montana State University

First Author

Elizabeth Arnold, Montana State University

Presenting Author

Elizabeth Burroughs

The Lady Tasting Tea Revisited: Insights on SUTVA Violations from a Canonical Example

Fisher's Lady Tasting Tea is often one of the first examples of an experiments considered when teaching a design-based causal inference course. However, we show that Lady Tasting Tea violates the stable unit treatment value assumption (SUTVA), a foundational assumption made by many commonly used causal inference models, and whose violations are typically associated with studies containing units that ``interact'' with each other (for example, those on social networks). We suppose that the Lady has knowledge that half of all cups receive tea first and the other half receive milk first, and that this assignment is completely randomized. For each cup, the Lady obtains a ``likelihood score'' of that cup receiving milk first that may vary depending on whether that cup is actually given milk first. After tasting tea from all cups, the Lady guesses that the cups with the largest likelihood scores are given milk first. We show that SUTVA may be violated under this model, even when the Lady's milk-first likelihood is always higher when that cup actually receives milk first. We then discuss how this may impact best practices in teaching an introductory causal inference class. 

Keywords

Lady Tasting Tea

Stable Unit Treatment Value Assumption

Causal Inference

Experimental Interference 

Co-Author

Michael Higgins, Kansas State University

First Author

Linus Addae, Boehringer-Ingelheim

Presenting Author

Michael Higgins, Kansas State University