A Century of Washington Statistics

Jonathan Auerbach, Washington Statistical Society President

The state of the Washington Statistical Society reflects the state of Washington statistics. Rapid changes in technology have deepened inequality, resulting in economic insecurity, anxiety, and resentment. In response, a populist movement is questioning the legitimacy of science as an independent and professional enterprise. The statistics profession—the discipline concerned with the mathematical principles of scientific inquiry—finds itself in the crosshairs.

It is under these conditions that we work to advance our mission of fostering statistical thinking and promoting unity among those concerned with statistical matters in the Washington metropolitan area.

It was also under these conditions that the Washington Statistical Society was born a century ago. The progressive era had largely ended, marked by the first Red Scare and new restrictions on speech. The anti-science movement reached a national flashpoint with the Scopes “Monkey” Trial just months before the society was established on January 19, 1926.

Since then, the state of the society has waxed and waned with the state of Washington statistics. After the New Deal, the society claimed nearly 1,000 members, representing a third of all members of the American Statistical Association. Following the Great Society three decades later, perhaps the peak of public trust in science, society membership grew to nearly 2,000 members, about an eighth of all members of the American Statistical Association.

Approximate Number of Members in the Washington Statistical Society and American Statistical Association

Sources: Allen (1988), Mason and McKenzie (2015)

Since its peak, however, membership has dropped back to New Deal–era levels. This decline is not due to a commensurate drop in local statisticians. Membership today is about two-thirds of what it was in the late ’90s, but the number of statisticians in the Washington metropolitan area is largely unchanged. The same is true for the number of statisticians in the federal government (as of March 2025).

If anything, the local statistics community has increased precipitously in recent years with the explosion of data science, a profession essentially equivalent to statistics, according to Jonathan Auerbach, David Kepplinger, and Nicholas Rios in their 2024 Significance article titled “What Is Data Science? A Closer Look at Science’s Latest Priority Dispute.” When data science is included, society membership constitutes roughly six percent of Washington-area statisticians today, down from a third in the late ’90s.

Approximate Number of Statisticians in the Washington, DC-Maryland-Virginia Metropolitan Statistical Area and Federal Government

Sources: BLS OEWS (Occupation Codes 15-2041, 15-2051, 25312 and OPM FedScope Occupation Codes 1529, 1530, 1531, and 1560. OEWS data is for the DC-VA-MD-WV MSA/PMSA. FedScope data is as of September except for 2025, which is as of March and preliminary.

Several factors explain the decline. Chief among them is the advent of the internet. Professional communities are no longer bound by geography, and the role the Washington Statistical Society once played in disseminating statistical knowledge to the local community is no longer necessary.

This shift is reflected in society activities. During the ’90s, the society helped organize more than 50 seminars and five short courses each year. A short course brought as many as 200 participants. Today, as in the early days of the society, seminars are held once or twice a month at most. While a seminar can attract more than 200 participants, attendance is remote or unpredictable. In addition, several short courses have been canceled since the pandemic due to lack of interest, leaving one to ask whether a regional association such as the Washington Statistical Society is obsolete.

I would argue the answer is no. While the role of the Washington Statistical Society has changed, its regional mission is, in fact, more relevant than ever. That is because the concerns of Washington-area statisticians—such as the quality of federal statistics and public information, the use of data as evidence for policymaking, and the promotion of statistics as necessary civics education—are of unprecedented national importance, shaping the infrastructure on which our modern data-dependent society is being built. These concerns are not limited to the Washington area, but our association with Washington—the seat of power in the US—provides the ultimate vantage point from which we can work together to advance them.

Our relevance is evident from recent programming, which have included subjects such as measuring income inequality (Connie Citro, 2024 Hansen Lecture), small area estimation and data integration (Partha Lahiri, 2025 Hansen Lecture, forthcoming), polling and poststratification (Andrew Gelman, 2024 seminar on special election with George Mason University and 2025 Joint Program in Survey Methodology Distinguished Lecture), political polarization and networks (Alex Volfovsky, 2024 Cox Award), and data linkage (John Eltinge, Amy O’Hara, and Martin Slawski, WSS President’s Invited Session at STATS Day), to name just a few.

But if we only reach 6% of those engaged in statistical work in the Washington metropolitan area, we are falling short of our mission of unity.

Our challenge is to demonstrate that statistical thinking is not optional in the age of artificial intelligence. Automation accelerates decision-making, but no algorithm can choose judiciously among the competing statistical principles that underlie design, measurement, and inference—just as no algorithm can decide where free speech ends and hate speech begins.

Looking back to 1926, we can be optimistic: Statistics soon entered a golden age. But we must also be cautious. That progress was earned only after confronting pseudoscience and learning from costly policy mistakes. Must the next golden age of statistics be equally hard won?

We can better position the Washington Statistical Society to fulfill its mission with the following three changes:

1. Bridge the generation gap. Currently, one quarter of members joined before 2000, while another quarter joined in 2023 or later. The society should seek events that appeal equally to members with both two and 20 years’ experience. (This problem is not new.)

2. Make volunteering easier. Many functions require coordination among multiple people, leading to back-and-forth email and fragile handoffs. Roles and transitions should be streamlined as the American Statistical Association did in 2003.

3. Make programming predictable. Seminars should be held on the same day each month to make turnout more reliable. Unreliable event dates and turnouts make it difficult to attract speakers, volunteers, and new members—particularly established statisticians with busy schedules.

I invite members to discuss these and other opportunities and challenges. In 1976, five past presidents of the American Statistical Association addressed members of the Washington Statistical Society at the annual dinner in celebration of the 50th anniversary. To celebrate the 100th anniversary, I have invited five past presidents of the Washington Statistical Society to speak at a proposed session of the Joint Statistical Meetings: Dan Kasprzyk (1986 –1987), Phil Kott (1996 –1997), John Czajka (2004 –2005), Michael P. Cohen (2007 –08), and Linda Young (2017 –2018). Hopefully, through this session and other activities planned throughout the year, we can chart a path for sustaining the Washington Statistical Society for the next 100 years.

Jonathan Auerbach

Washington Statistical Society President

Jonathan Auerbach is an assistant professor in the department of statistics at George Mason University and president of the Washington Statistical Society.