¹ In our projected inquiry, the materials will be further delimited and confined to a few substantive fields of application, of which the following are merely illustrative: Diverse problems of community planning, practices in personnel selection, propaganda for diverse purposes, advertising, democratic participation in political or organizational life, reduction of race hostilities, planned migration, inter-cultural relations. Limiting the inquiry provisionally to the study of applied research in a relatively small number of diverse fields should provide a broad range of experience and yet permit a fairly intensive study within each field.

^1a This general observation now takes on added force since the subsequent public reactions to the election forecasts on November 2, 1948 by the major polling organizations. It remains to be seen if the reaction against empiristic polling forecasts is generalized to the discredit of social science.

² The importance of this was long since recognized by Walter Lippmann, in his perceptive chapters on the potential role of the social science expert. Public Opinion (New York: Macmillan, 1922), Chapters 25, 26.

³ Further observations on this point are presented by S. A. Stouffer, “The Strategy of the Social Sciences,” address before the Harvard Graduate Forum, April 20, 1948 and by R. K. Merton, “Discussion of ‘The Position of Sociological Theory,’” American Sociological Review, April 1948, 13, 164–168.

⁴ Since I have provisionally discussed this in a paper presented to the American Sociological Society in 1946, I shall not comment on it further at this time. “The Bearing of Empirical Research upon the Development of Social Theory,” American Sociological Review, October 1948, 13, 505–515. More study of actual cases is required to outline the conditions under which theoretic derive from applied research in social science.

^4a Since this was written, the 1948 election forecasts have provided the most dramatic recent instance of the danger of operating with wholly empirical and theoretically ungrounded uniformities. In previous national elections it had been found that virtually no net shift in vote-intentions occurred during the last weeks of the political campaign. The extrapolation of this empirical pattern to the 1948 campaign by the polling organizations led to now familiar and unfortunate consequences.

⁵ Since this was first written, James Bryant Conant has set forth apposite remarks (in his presidential address before the A.A.A.S., December 28, 1947): “To my mind we need to analyze the present situation, not by attempting to classify the various sciences and their subdivisions into pure and applied science, but by examining closely each separate undertaking. I have suggested in a paper on ‘Science and the Practical Arts’ that we need to inquire as to the degree of empiricism now present in any branch of science. The cases I quoted as examples were classical optics and chemotherapy. In the former the conceptual scheme employed has wide validity, the degree of empiricism is very low. In the latter the concepts are few and of limited application, progress toward a new drug is still very much of a ‘cut and dry’ affair, the degree of empiricism is high. … I should like to suggest that unless progress is made in reducing the degree of empiricism in any area, the rate of advance of the practical arts connected with that area will be relatively slow and highly capricious.” (underscoring supplied). There remainds for us the question of the factors which make for the retention of a high degree of empiricism in much of applied social science, and it is in connection with this question that we make the suggestions found in the text at this point.