AI can generate code, fit competing models, produce visualizations, and summarize results at a speed that would have been difficult to imagine not long ago. For statisticians, this raises an uncomfortable but useful question: As AI does more analytics, what should statisticians do more of?
A statistician's view
From Model Accuracy to Data Reliability: A Practical Framework for ML Teams
Praveen Gupta Sanka discusses the Data Reliability Score, a framework developed with Vidya Sagar Minukuri. She says, “The idea is simple: As machine learning teams use quantitative metrics to decide whether a model is ready for deployment, they should also use quantitative metrics to decide whether the data is reliable enough to support that decision.”
Statistics in the Age of AI
Jae-Kwang Kim offers his thoughts about the potential evolution of university statistics departments as they face the environmental shifts of AI.
A Case Study in Publication Practices: Full House Modeling
Daniel J. Eck responds to David Banks’ earlier A Statistician’s View piece.
Statistician’s View: The Power of Collaboration in Statistics
I think Nick Beyler’s best sentence comes at the very end of his February 2025 STATtr@k column, says Barry Nussbaum. “Want to collaborate on something?” That’s the key to all our statistical work. To have impact, we use our talents as part of a team.
Statistician’s View: A Follow-Up to Gang Xie’s Practicing Statistician’s Plea
Jonathan Shuster offers a commendation and response to Professor Gang Xie for his editorial, “A Practicing Statistician’s Plea.”
Like College Students, ChatGPT Finds Statistics Difficult
Norman Matloff, computer science professor at the University of California, Davis discusses whether ChatGPT is up to the task of doing statistics at the level demanded of students.
A Statistician’s View: New Year Hopes for Andreas Georgiou as Court Decisions Loom
Interviewed recently for The Amherst Student, Georgiou said his legal troubles are “always a little bit tiring to the heart to think about.” But 2023 brings with it hope that justice, long denied, is still in reach.
A Call for More Graduate Programs in Statistics Education
Sam Behseta, Cherie Ichinose, and David Drew put forth the view that closing the gap in students’ appreciation of statistical thinking and decision-making under uncertainty can only be achieved if we train the next generation of academics with expertise in statistics education.
The Demand for Literacy Training on Quantitative Public Health Data
The global pandemic increased the public’s demand for information. Almost daily, media and policymakers present numbers on testing, deaths, hospitalization rates, and vaccine distribution. This information often includes data visualizations and tables viewers must interpret to better understand the pandemic’s scope and its impact on local communities and countries worldwide. As a result, there is a need for a new vocabulary around public health and data literacy.