GENERATIVE AI IN STATISTICS EDUCATION: A BIBLIOMETRIC AND EMPIRICAL APPROACH

Mitra Jazayeri; Daniel Laurence; Andriy  Olenko; Andrew Buldt; Nadiya  Kosytsina; Xia  Li

Authors

Dr. Mitra Jazayeri La Trobe University https://orcid.org/0000-0002-8560-3243
Dan University of Melbourne https://orcid.org/0000-0002-4941-359X
Associate Professor Andriy Olenko La Trobe University https://orcid.org/0000-0002-0917-7000
Dr. Andrew Buldt La Trobe University https://orcid.org/0000-0002-2020-2780
Dr. Nadiya Kosytsina https://orcid.org/0009-0001-3879-8375
Dr. Xia Li La Trobe University https://orcid.org/0000-0003-1890-5548

Keywords:

Generative AI, Learning design, Statistics anxiety, statistics education, AI literacy

Abstract

Introduction The integration of generative AI into education has sparked growing interest, particularly in disciplines like statistics and data science, where many students — especially those from non-quantitative backgrounds — face learning challenges. While AI tools are increasingly adopted, the research landscape is fragmented, and guidance for educators and researchers remains limited. This study adopts an emerging design approach to explore and synthesise the evolving literature, aiming to inform both curriculum development and future research. Our aims are to:
(1) map the existing literature using bibliometric methods to identify and visualise the current intellectual structure of generative AI-related research in statistics education;
(2) complement this with findings from an empirical study investigating students’ perceptions of AI use in an assessment context, providing a foundation for educators and instructional designers to align AI capabilities with inclusive, learner-centred teaching strategies in statistics. Methods

The bibliometric component employs VOSviewer and the Bibliometrix package in R to systematically review and visualize the publications on Generative AI in statistics education. The analysis focuses on themes such as student motivation, support for non-quantitative learners, personalised feedback, AI-enabled engagement, and evolving role of generative AI in statistics teaching and learning. In parallel, the empirical component involves a student survey captures experiences of AI use in a tertiary statistics subject. Findings from both components aim to deepen understanding of how generative AI is shaping pedagogical practices and learner experiences in statistics education.

Conclusion

This emerging design study responds to the growing need for structured guidance in navigating AI's role in statistics education. By identifying influential studies, theoretical shifts, and research gaps, we aim to inform curriculum development and pedagogical strategies. The resulting conceptual framework and tool will support educators and researchers in understanding and applying AI effectively—particularly for diverse learners with limited mathematical confidence.

Author Biographies

Dr. Mitra Jazayeri, La Trobe University

I have extensive experience in teaching introductory and intermediate statistics subjects to undergraduate and postgraduate students across fields. In 2022, I received the UK FHEA (Fellowship in Higher Education Academy) and La Trobe University SCEMS Teaching Excellence awards.
My research focus is on factors such as AI and other technologies that impact student learning and assessment, with a prior interest in statistical techniques applied to medical and health data. I have served as a bio-statistician consultant for the NHMRC and NICS, resulting in a publication in Emergency Medicine Australasia in 2013. I also researched environmental and lifestyle factors that impact ED presentations for asthma in children and published a systematic review in the Journal of Allergy in 2017.
In 2018, I was seconded from my lecturer role to a teaching-focused lecturer role and have published two articles in the International Journal of Education in Mathematics, Science and Technology (IJMEST) regarding the effect of anxiety on psychology students' statistics learning. From 2021- the present, my research team and I have been involved in research into the effect of hybrid teaching on health science students' learning.
Dan, University of Melbourne

Dan Laurence is a senior learning designer, technologist, researcher and occasional teacher. In 2015 and again in 2017 he was awarded the Vice Chancellor’s Teaching Excellence award at Swinburne University, and again in 2019 the Vice Chancellor's Teaching Award at La Trobe University. His work was awarded the national 2017 AFR Educational Technology Award and a finalist in the 2022 ASCILITE Battle of the LMS.
An assemblage of pedagogical, design, usability and technology expertise continues to inspire the co-design quality student experiences.
Associate Professor Andriy Olenko, La Trobe University

Andriy's research interests cover a broad range of topics from the theory of stochastic processes, spatial statistics, Shannon and wavelets stochastic approximation, statistical inference, and statistical and data science applications, in particular, signals processing and cosmology. He is currently supervising research students on topics in spatial stochastic modelling and applications. Andriy actively collaborates with several research groups in France, UK, Japan, Italy, USA, Canada, Spain, Sweden and Ukraine. He is a coordinator of the La Trobe Master of Data Science.
Dr. Andrew Buldt, La Trobe University

Andrew is a lecturer in the Discipline of Podiatry in the School of Allied Health, Humans Services and Sport. His main academic role is the coordination of a large core first year subject for allied health and nursing students. He is passionate about curiculum design of large subjects and optimising the transition for first year students into university life. Additional teaching roles include the coordination of clinical podiatry subjects.

Andrew has research intesets in foot and ankle biomechanics and medical imaging for foot conditions. Andrew is currently involved in several research projects including the biomechanics evaluation of people with midfoot osteoarthritis and the analysis of foot morphology using 3-dimensional scanning. Andrew has published 25 peer-reviewed journal articles and currently supervises 2 PhD students.

Andrtew proudly acknowledges the Wurundjeri people of the Kulin nations as the Traditional Custodians of the land and its waterways on which he lives and works.
Dr. Xia Li, La Trobe University

Xia Li is a University Statistics Consultant in Statistics Consultancy Platform https://www.latrobe.edu.au/research-infrastructure/research-facilities/statistics-consultancy-platform.

Xia's principal research interests are applied statistics, biostatistics, especially longitudinal data analysis, survival data analysis, latent based model data mining methods and statistical learning methods with expertise in data analysis and statistics modelling. She is skilled in a variety of data analysis and statistics software packages (SAS, R, SPSS, WINBUGS, MPLUS, SQL, AMOS), very proficient in using many different R packages and writing functions, and coping with large data sets in R. As an early career researcher in the area of statistics, Dr Li has published more than one hundred research publications in peer reviewed journals. Xia also collaborates with the researchers from a variety of disciplines at or out of La Trobe University. She also provides statistics supports for the grant applications including NHMRC grants.

The job of a statistician involves all aspects of the study. From study design, implementation, study compliance, data analysis and interpretation, and manuscript preparation. Even during the very early stage of conceptualization, statisticians can make substantial contributions by helping the investigators formulate and refine their scientific questions.