The ASA will celebrate its 175th anniversary in 2014. In preparation, column “175”—written by members of the ASA’s 175th Anniversary Steering Committee and other ASA members—will chronicle the theme chosen for the celebration, status of preparations, activities to take place, and, best yet, how you can get involved in propelling the ASA toward its bicentennial.
Stephen M. Stigler taught at the University of Wisconsin-Madison before moving to The University of Chicago. He was editor of JASA: Theory and Methods from 1978–1981 and has published several books, including The History of Statistics and Statistics on the Table.
The American Statistical Association was founded in Boston in 1839. In one sense, it was a late arrival. There were some two dozen statistical societies begun before 1840, including the London (later Royal) Statistical Society (1835). Some of these had short lives; none of the first three (of four) appearances of the Societé de Statistique de Paris (1803, 1829, 1830) or the Glasgow and Clydesdale Statistical Society (1836) lasted even a decade. Others lingered on much longer before they succumbed; the Societé de Statistique de Marseille (1827) led a fairly active life until about 1900.
The secret behind the longevity of the ASA and RSS is not hard to deduce: Both societies adapted and changed with the times, while those that remained stagnant failed. All the early statistical societies began in response to the growing economic activity of the time and to curiosity about the facts of the state—the “statistics” of their respective nations. Most contented themselves with meetings and discussion; only the RSS began a serious publication program before 1840. The failed societies slowly or rapidly degenerated into social clubs for businessmen and politicians and disbanded when interest in dry statistics faded.
For most of its first half century, the ASA did little better than the others did. Before 1888, when it finally founded a journal, its publications were few, sporadic, and dull. But while the early ASA was a sedentary organization, with a small local membership meeting quarterly for discussion, its members’ interests maintained some contact with the exciting new nation. Perhaps the first work of real import that can be closely associated with the ASA was the Ninth Census of the United States in 1870, for which ASA member Francis Amasa Walker served as superintendent. The decennial American censuses date from 1790, but with the 1870 Census, their reports were lifted to a new level of quality in analysis and presentation of results. The latter included the great Statistical Atlas, published in 1874, which set a new standard for the graphical display of data. Nearly two feet in height, it towers as a landmark in American statistics, both intellectually and physically.
Walker also holds the dubious distinction of being the only ASA president to have lost his life in the line of duty. Elected in 1882, he served until his death and was instrumental in arranging the first ASA meeting to take place outside the Boston area in late December of 1896, in Washington, DC. Despite weak health, he traveled in winter weather to attend and died within five days of his return to Boston.
The mission of the ASA at its founding was “to collect, preserve, and diffuse statistical information in the different departments of human knowledge.” Since, at present, the ASA does none of these, at least as “statistical information” was understood in 1839, there has clearly been a major change. That change began during Walker’s presidency, with the founding of the Journal of the American Statistical Association (JASA) in 1888 and, consequently, a new role in publications. The discipline of a rigorous professional journal changed the character of the ASA and the focus of its meetings. It moved from being a regional discussion club to a national and then international producer of research.
The focus of the journal has changed, as well. Over the first decades of the 20th century, JASA was as much an economics journal as a statistics journal. But with influential works by Harold Hotelling and Holbrook Working (confidence bands for regression lines) in 1929 and Milton Friedman (rank tests for two-way ANOVA) in 1937, it moved toward its present status as a premier journal of applicable statistical theory and methodology. The 1958 paper by Edward Kaplan and Paul Meier on survival analysis remains the most-cited work on statistics in the scientific literature. The ASA has added other journals to its portfolio, and they too have grown in excellence.
After 175 years, the ASA still faces daunting challenges. Will we find a sustainable way forward in the digital age for publication? Will we find a solution to the problem of holding large meetings and maintaining a unifying focus despite the growing variation of topics under our umbrella? Can we continue to provide value to members beyond our core publications in a world that innovates at rapidly increasing speed? Other societies face the same challenges; none has as yet found solutions. But in view of the principled adaptability shown through our 175-year history, we can be optimistic of overcoming these and many similar challenges.
Leave a Reply