My optimism is rooted in the belief that there has never been a more exciting or more consequential time to be a statistician. I don’t say this to be pithy, but as a considered conviction shaped by watching our field navigate one of its most complex periods. Yes, there are pressures, and yes, we sometimes struggle to articulate our own value. These are signs of a field that matters, that is being tested, and that is rising to meet the challenge. I am optimistic about the field of statistics, and I want to explain why.
One source of my optimism is captured in an observation from the National Academies report Frontiers of Statistics in Science and Engineering: 2035 and Beyond. That observation is that familiar concepts in statistical thinking and theory emerge “in response to new assumptions, new availability of broad classes of data, new linkages between separate data sets, and new questions of interest enabled by new data.” This is not a field standing still. It is a field being continually advanced by the problems the world brings to it. The merger of administrative records with survey data, the linking of electronic health records across institutions, the integration of satellite imagery with census data—each of these creates analytical possibilities that demand not just computational power but the careful reasoning about measurement, bias, and inference that is the core of statistical practice. Data work today means negotiating the gap between what was recorded and what we want to know. That negotiation has always been, at its heart, a statistical one.
When Silicon Valley began declaring that “data scientist” was the sexiest job of the 21st century, it sometimes seemed to sideline the statisticians who had been doing that work for decades. The emergence of data science has created an enormous demand for people who can do what statisticians do best: think carefully about data collection; quantify uncertainty; design sound studies; and resist the seductive pull of spurious correlation. The growing community of data practitioners who are coming from computer science, economics, biology, and beyond all need statistical foundations. Programs like the ASA’s Data Science Professional (DSP) certification reflect exactly this kind of forward-looking engagement, creating recognized pathways that embed statistical thinking at the heart of the data profession.
Kathy Ensor,
Frontiers of Statistics in Science and Engineering: 2035 and Beyond Committee Chair
The Frontiers report recognized that “expanded interdisciplinary team science, in which statistics plays a key role, is vital to personalized medicine, public health, ecology, aerospace, applied physics, and climate studies, among others” and points to something I find genuinely exciting. Statisticians are essential collaborators in the most consequential scientific work of our time. When a clinical trial team is determining whether a new therapy works for a specific patient population, when ecologists are modeling species response to a changing climate, when aerospace engineers are quantifying uncertainty in a system where failure is not an option, it is in these settings that the statistical contribution is not peripheral. It is the thing that makes the conclusion trustworthy. Team science has expanded both the reach and the responsibility of our field, placing statisticians at the table where discoveries are made and decisions are taken. This is a reason for optimism.
When you read this column, I will be retired, so my optimism comes from watching what young statisticians are doing, and it is extraordinary. They are developing methods for causal inference from observational data that would have seemed impossibly ambitious 20 years ago. They are building interpretable machine learning frameworks that bring statistical rigor to artificial intelligence. They are working at the intersection of statistics and social justice, developing tools for detecting discrimination in algorithms, and measuring inequity in healthcare delivery. They are doing statistics in R and Python and Julia and sharing their code openly, building a culture of reproducibility that is transforming scientific practice.
The students entering our field today are diverse in background and ambition in ways that enrich the discipline enormously. The growth of statistics education at the undergraduate level is producing graduates who see statistical thinking not as a technical specialty but as a fundamental mode of reasoning about the world.
Perhaps the most frequent source of anxiety I hear among statisticians is the rise of artificial intelligence. The concern is understandable. When large language models can answer data questions in plain English and automated pipelines can fit models without human intervention, what becomes of the statistician? My answer is the statistician becomes more important, not less.
Lance Waller,
Frontiers of Statistics in Science and Engineering: 2035 and Beyond Committee Vice Chair
AI systems are, at their core, statistical systems. They are trained on data, they make probabilistic predictions, and they fail in characteristically statistical ways. Understanding those failure modes requires statistical thinking. Evaluating whether an AI system is fair, accurate, and calibrated requires statistical methodology. Deciding what data to collect to train a model, how to handle missing data, and how to quantify uncertainty in model outputs are statistical problems. The rise of AI has made statistics urgent.
Finally, I am optimistic because the way we communicate statistics is changing. For much of its history, statistical knowledge lived in journal articles and textbooks accessible only to specialists. That is no longer true. Podcasts, data journalism, open-access publishing, and interactive visualization have opened statistical ideas to broad audiences in ways that are genuinely exciting. Statisticians are increasingly present on social media, in policy conversations, and in public life. This is a version of our field I think our founders would have celebrated.
I am optimistic about the field of statistics because it is needed, because it is honest enough to reform itself, because its next generation is brilliant and broad-minded, because it has a central role to play in the most important technological development of our era, and because we are learning to tell our own story. The challenges are real, but they are the challenges of a field at the center of things. That, to me, is cause for genuine and enduring optimism.
This article does not include the author’s name.