The formation of a discipline: Mathematical statistics in the United States in the nineteenth and twentieth centuries

In 1839, five Boston physicians, lawyers, and businessmen founded the American Statistical Association. For most of the rest of the century, this organization's membership remained less than one hundred, and the majority of participants came from the Boston area. Their activities reflected the main concerns of statisticians of the time, which centered on the solution of social problems.;This situation prevailed, for the most part, until the end of the nineteenth century. Fifty years later, however, the landscape of professional statistics in the United States had altered dramatically. The national and international membership of the ASA exceeded 4,500, and the organization's quarterly Journal of the American Statistical Association consisted of more than five hundred pages of articles a year. A significant portion of these pages included up-to-date mathematical methods of statistical estimation and prediction. Most importantly for the present study, the statisticians in the ASA had spawned another disciplinary community whose members concentrated their investigations on those mathematical methods, creating and refining them, and studying their theoretical properties. The Institute of Mathematical Statistics, organized in 1935 under the presidency of Henry L. Rietz, coordinated this community's professional activities, including the production of the Annals of Mathematical Statistics, founded in 1930 by Harry C. Carver. Mathematical statisticians had jobs in colleges and universities, private industries, and bureaus of the state and federal governments. At the end of the 1940s, mathematical statistics, both as a profession and a discipline, occupied a well-defined place within the American research community, forming a sort of borderland between mathematics and the social and physical sciences.;In following the process by which mathematical statistics became an independent field of study, occupying a place between abstract mathematics and an area of its application (the social sciences), this dissertation demonstrates that the formation of the discipline resulted both from the creation of mathematical theorems and methods, as well as from the establishment of professional institutions and from a group of researchers' successful efforts to differentiate between their interests and those of the social scientists who had first begun making use of statistics.