Interviews with Statistics, Data Science Educators
Return as Collections Expand

Juana Sanchez, JSDSE Editor-in-Chief


After a brief pause, the latest issue of the Journal of Statistics and Data Science Education brings back the Interviews with Statistics and Data Science Educators column, starting with part one of an interview with Jim Albert. In this segment, Albert shares his experience teaching the undergraduate introductory statistics course through the lens of Bayes’ rule. The interview has been added to one of the journal’s Interviews collections.

Interviews is not the only collection the journal offers. A newly launched collection is Generative AI in Statistics and Data Science Education, and we look forward to its rapid expansion. To kick things off, we have included the paper titled “Developing Students’ Statistical Expertise Through Writing in the Age of AI”  by Laura S. DeLuca and coauthors. The paper critically examines whether ChatGPT models can help students grow from novice to expert writers of statistical reports. Based on the models evaluated, the conclusion is they currently cannot.

This issue also includes two papers presenting original, classroom-tested data sets selected because they are contextualized and impactful for teaching statistics and data science concepts and modeling approaches. The first, “Identifying Biases and the Relevant Statistical Population: The Case of the Loch Ness Monster”  by Charles G. M. Paxton, Adrian J. Shine, and Valentin M. Popov, provides a data set of sightings of the Loch Ness Monster to support classroom discussions about the critical importance of clearly defining the population studied. The second, “Multiple Regression with Transformation and Variable Selection at the Industrial Scale”  by Gus Greivel and coauthors, introduces data resulting from steel mill operations. This data set is used to illustrate how multiple regression can be applied to predict the force necessary in a steel rolling mill operation based on a variety of predictor variables in the engineering physics-based model.

Authors might use data sets from prior scholarly work that align with specific pedagogical goals or leverage large institutional data sets to facilitate collaborative, community-centered learning. The paper titled “Healthcare Analytics Teaching Cases” by Concetta A. DePaolo and Milton R. Soto-Ferrari draws on data sets of nurses’ shifts, insurance claims, and COVID cases to support experiential learning focused on the impact of statistical inquiry in patient care and administration. The paper titled “Empowering Students to Assess the State of Diversity, Equity, and Inclusion on Campus” by Jonathan Auerbach and Christi Wilcox presents an approach that engages students in analyzing data from their own institution to better understand the student body while also supporting institutional efforts to derive meaning from its vast data resources.

Finally this issue features other helpful tips for designing courses and enhancing teaching practices. The paper titled “Illustrating a Framework for Evaluating Inclusivity in Teaching Through Student Data Analysis Projects” by Nicole M. Dalzell, Zoe L. Rehnberg, and Allison S. Theobold uses a framework (a set of questions) to decide whether the educational process (the student’s experience) designed to help students complete a data analysis project assignment is inclusive.

Lucy D’Agostino McGowan’s “Using Mathlink Cubes to Introduce Data Wrangling with Examples in R” shares a hands-on activity that facilitates visualization of data management tasks before students transition to coding in R.

“The Impact of Homework Deadline Times on College Student Performance and Stress: A Quasi-Experiment in Business Statistics” by Charlie Smith reports that when giving students the choice of 4:00 p.m. or 11:59 p.m. for turning in their homework, students chose 11:59 p.m., but learning outcomes were the same with both deadlines.

In the Statistics and Data Science in the Health Sciences category, the paper titled “Ethics in Clinical Research, E-Module Versus Traditional Online Lecture, a Randomized Study” by Lynette Smith and coauthors suggests using an e-learning module alone to train students taking an online clinical trials biostatistics class in the ethical aspects of conducting clinical trials may not result in satisfaction and learning outcomes that are as good as using the e-learning module complemented with traditional online lecture slides.

JSDSE and the July 2025 issue are open-access. Feedback about the journal and questions are welcome and can be emailed to Juana Sanchez.