Einstein, Renoir, and Greeley: Some Thoughts about Evidence in Sociology: 1991 Presidential Address

Abstract

It is relatively easy to determine if evidence supports or does not support a given theory or conclusion, but a central problem in sociology is our inability to go beyond this and develop truly convincing evidence about a theory. Several changes in current practices are vital, including the cessation of nonproductive assaults between sociologists on the evidence used in the discipline. More fundamentally, theory should not be viewed as one pole of a continuum with research on the other end. A probabilistic view of theory is proposed such that a theory may be correct even if there is negative evidence. This leads to a different way of evaluating evidence, a different view of the role of examples, a distinction between explanations ofparticular events as opposed to evaluations of theories, skepticism about the possibility of developing theories that accountfor long complex chains of events, and a sharp differentiation between the evidence used in applied as opposed to basic research. A new sociological specialty should concern itself with epistemological issues that arise in determining the verifiable propositions that are the basic building blocks to sociological knowledge.