Uploaded October 2026 | Updated November 2026, 2 days ago
Recording Date: October 22, 2026, 12 p.m. ET
Transcript of Learning Curves | Episode 4: Teaching Reproducibility in the Age of AI (tandfonline.com/journals/ujse21)
Learning Curves | Episode 4 Audio-Only Format: drive.google.com/file/d/19RNw8l4TNEuga_rYwkbebaUxaIkAR71G/view?usp=drive_link
Participants in Episode 4:
Nicholas Horton (host), Statistics and Data Science, Amherst College
Sara Stoudt (host), Mathematics and Statistics Department, Bucknell University
Tiffany Timbers, Department of Statistics, University of British Columbia
Elizabeth Upton, Department of Mathematics and Statistics, Williams College
Owen Ozier, Department of Economics, Williams College
Synopsis of Episode 4
In this podcast, authors of papers published in the Journal of Statistics and Data Science Education (JSDSE) engage in conversation about their experience teaching reproducibility. The latter was the theme of Issue 3 of Volume 30 of the JSDSE, published in 2022 (tandfonline.com/toc/ujse21/30/3).
After discussions and deliberations that prompted that Issue and extensive efforts by the ASA’s Journal of the American Statistical Association, the JSDSE augmented the requirements for authors to facilitate the reproducibility of their results by readers. This entails sharing code and de-identified data. Teaching reproducibility, however, sometimes presents challenges, as Ostblom and Timbers (2022) explain (doi.org/10.1080/26939169.2022.2074922).
In 2024, an editorial written by Nicholas Horton and Sara Stoudt reviewed the reproducibility requirements, summarized ongoing trends in open science, and updated the options for data and code sharing (see doi.org/10.1080/26939169.2024.2364737 ). Reproducibility is not a static topic, and it is here to stay, but 2026 has arrived and once more we are confronted with sweeping changes that force us to revisit the requirements for reproducibility and how we teach students about them. What does reproducibility mean in the age of AI?
When thinking about reproducibility, many of us in the statistics education community think of R, and the suite of applications it has to facilitate reproducibility. But Python has become more prevalent recently, and the traditional statistical software such as SAS, SPSS, STATA and others continue to be used as well, as Upton et al. (2026) (doi.org/10.1080/26939169.2026.2725967 ) illustrates. What do users of those need to do to help others reproduce their results? Are there applications for those that make it easy?
Acknowledgments
This podcast would not be possible without the advice and help of Kim Gilliam (ASA) and Eric Sampson (ASA), and the support of the JSDSE Editorial Team and the podcast participants.
References
Horton, N. J., and Stoudt, S. (2024), “Editorial: Guidelines and Best Practices to Share Deidentified Data and Code,” Journal of Statistics and Data Science Education, 32(3), 227–231. doi.org/10.1080/26939169.2024.2364737
Horton, N. J., Alexander, R., Parker, M., Piekut, A., and Rundel, C. (2022), “The Growing Importance of Reproducibility and Responsible Workflow in the Data Science and Statistics Curriculum,” Journal of Statistics and Data Science Education, 30(3), 207–208. doi.org/10.1080/26939169.2022.2141001
Ostblom, J., and Timbers, T. (2022), “Opinionated Practices for Teaching Reproducibility: Motivation, Guided Instruction and Practice,” Journal of Statistics and Data Science Education, 30(3), 241–250. doi.org/10.1080/26939169.2022.2074922
Upton, E., Cai, X., Jakiela, P., Ozier, O., and Raman, S. (2026), “The Software Behind the Stats: A Student Exploration of Software Trends in Economics, Political Science, and Statistics,” Journal of Statistics and Data Science Education, 1–19. doi.org/10.1080/26939169.2026.2725967
Special Issue of the JSDSE: Vol 30(3), 2022, tandfonline.com/toc/ujse21/30/3?nav=tocList
Recording Date: October 22, 2026, 12 p.m. ET
Transcript of Learning Curves | Episode 4: Teaching Reproducibility in the Age of AI (tandfonline.com/journals/ujse21)
Learning Curves | Episode 4 Audio-Only Format: drive.google.com/file/d/19RNw8l4TNEuga_rYwkbebaUxaIkAR71G/view?usp=drive_link
Participants in Episode 4:
Nicholas Horton (host), Statistics and Data Science, Amherst College
Sara Stoudt (host), Mathematics and Statistics Department, Bucknell University
Tiffany Timbers, Department of Statistics, University of British Columbia
Elizabeth Upton, Department of Mathematics and Statistics, Williams College
Owen Ozier, Department of Economics, Williams College
Synopsis of Episode 4
In this podcast, authors of papers published in the Journal of Statistics and Data Science Education (JSDSE) engage in conversation about their experience teaching reproducibility. The latter was the theme of Issue 3 of Volume 30 of the JSDSE, published in 2022 (tandfonline.com/toc/ujse21/30/3).
After discussions and deliberations that prompted that Issue and extensive efforts by the ASA’s Journal of the American Statistical Association, the JSDSE augmented the requirements for authors to facilitate the reproducibility of their results by readers. This entails sharing code and de-identified data. Teaching reproducibility, however, sometimes presents challenges, as Ostblom and Timbers (2022) explain (doi.org/10.1080/26939169.2022.2074922).
In 2024, an editorial written by Nicholas Horton and Sara Stoudt reviewed the reproducibility requirements, summarized ongoing trends in open science, and updated the options for data and code sharing (see doi.org/10.1080/26939169.2024.2364737 ). Reproducibility is not a static topic, and it is here to stay, but 2026 has arrived and once more we are confronted with sweeping changes that force us to revisit the requirements for reproducibility and how we teach students about them. What does reproducibility mean in the age of AI?
When thinking about reproducibility, many of us in the statistics education community think of R, and the suite of applications it has to facilitate reproducibility. But Python has become more prevalent recently, and the traditional statistical software such as SAS, SPSS, STATA and others continue to be used as well, as Upton et al. (2026) (doi.org/10.1080/26939169.2026.2725967 ) illustrates. What do users of those need to do to help others reproduce their results? Are there applications for those that make it easy?
Acknowledgments
This podcast would not be possible without the advice and help of Kim Gilliam (ASA) and Eric Sampson (ASA), and the support of the JSDSE Editorial Team and the podcast participants.
References
Horton, N. J., and Stoudt, S. (2024), “Editorial: Guidelines and Best Practices to Share Deidentified Data and Code,” Journal of Statistics and Data Science Education, 32(3), 227–231. doi.org/10.1080/26939169.2024.2364737
Horton, N. J., Alexander, R., Parker, M., Piekut, A., and Rundel, C. (2022), “The Growing Importance of Reproducibility and Responsible Workflow in the Data Science and Statistics Curriculum,” Journal of Statistics and Data Science Education, 30(3), 207–208. doi.org/10.1080/26939169.2022.2141001
Ostblom, J., and Timbers, T. (2022), “Opinionated Practices for Teaching Reproducibility: Motivation, Guided Instruction and Practice,” Journal of Statistics and Data Science Education, 30(3), 241–250. doi.org/10.1080/26939169.2022.2074922
Upton, E., Cai, X., Jakiela, P., Ozier, O., and Raman, S. (2026), “The Software Behind the Stats: A Student Exploration of Software Trends in Economics, Political Science, and Statistics,” Journal of Statistics and Data Science Education, 1–19. doi.org/10.1080/26939169.2026.2725967
Special Issue of the JSDSE: Vol 30(3), 2022, tandfonline.com/toc/ujse21/30/3?nav=tocList









