Sociology and statistics in Britain, 1830-1990

Abstract

This thesis examines the historical relationship between British sociology and statistics in the nineteenth and twentieth centuries. It begins with an analysis of the role that the early development of statistics played in the history of social science, followed by an examination of other nineteenth century, non-quantitative, projects of social enquiry that were to have an influence on the later development of British sociology. The thesis then continues with an analysis of the contributions of the Sociological Society to the development of sociology in Britain and its role in sociology’s relationship with statistics. The last and most detailed part of the thesis is devoted to an examination of the trends in the development of academic sociology in Britain in the twentieth century. It analyses the major factors that had significant influence on the possible incorporation of quantitative methods, and a statistical and probabilistic worldview more generally, into British sociology. Most of the study is based on original archival research and uncovers previously unexamined aspects of events, movements and choices that have defined the character of British sociology since its academic beginnings. The argument is that the relationship between sociology and statistics in Britain has been characterised by a remarkable continuity and been subject to very little change over many years; that it has been distinguished by a negative obsession with statistics on the part of British sociologists who have made consistent efforts to try to prove statistics unsuitable for sociological research and excuse themselves from using them. The study concludes that the relationship between statistics and sociology in Britain has not been determined on the basis of pragmatic concerns but on the basis of uninformed preferences and deficiency in statistical knowledge. The divide that has existed between sociology and statistics was not inevitable but was the product of a particular set of circumstances and a particular set of choices made, both within and without British academic sociology. The aim of this thesis is to bring to the fore the interplay of these factors and show that the relationship between sociology and statistics matters and ought to be an area of growing concern to British sociologists. It explains not merely British sociology’s methodological choices but its relationship with the very thing that made both it and the society it studies – modern science. Ultimately, the relationship between sociology and statistics in Britain matters because discoveries in science in the last hundred years or so have shown that the World, all of existence, social or otherwise, is fundamentally probabilistic; and that statistics is the language best placed to describe how it works.