¹ It is neither in the logical structure of the propositions with which an investigation concludes, in their subject matter nor even in the aims of the investigator that the significant differences between applied social research and other types of social research can be found. The term “applied research” in the social sciences may be understood to refer to investigations performed for policy-makers who use or intend to use the resulting propositions as elements in their decisions. It is simply research, the results of which are to be applied in some way in practice by those who have in their charge the care of practical affairs. It is not applied research in the sense of the application of scientifically tested general principles obtained in “basic” or “pure” research, to the explanation of concrete and particular situations or to the management or construction of concrete and particular constellations of actions. Applied social research of the latter type might indeed develop in the course of time as the body of basic propositions, rigorously tested by systematic empirical procedures, grows; since there are practically no such propositions in social science today, this type of social applied research cannot exist for the time being. A first approximation to this type of applied research exists where concrete situations are explained and manipulated under the guidance of general hypotheses about behaviour more or less derived from psychoanalysis. Much of the research in industrial sociology practiced by Elton Mayo, the Chicago School and the Tavistock Institute is of this sort. Action-research in the field of race relations also falls more or less under this head.

To summarize the difference from applied research in the better established sciences consists (a) in the absence of rigorously tested general propositions and (b) in the absence of rigorous intellectual controls over the results of the manipulations introduced in accordance with those hypotheses.

² It is not a matter here of a sharp line which separates these two types of investigation; it is only a question of proportion.

³ These strictures do not imply that the counsel based on these improvised and certainly scientifically untested hypotheses was not valuable, intellectually and practically. Indeed, during the war, the advice based on such vaguely formulated and only clinically supported hypotheses was usually more pertinent to important decisions than that proferred by experts in the use of exact techniques of investigation, who had, however, no general theoretical orientation. The fruitful guesses of the anthropologists and psychiatrists about German behaviour during the war compares very favourably with the meagre guidance afforded by the polls conducted among prisoners-of-war.

⁴ It would be relevant for our present purposes to know just how much of the elegant and precise propositions of economic theory find application in the recommendations for policy and in the actual policies made by professional economists.

⁵ The Tavistock doctrine also requires the close association of the persons under study, the persons with formal policy-making executive powers and the “therapeutic” investigators.

⁶ Operational research merits mention in passing, not because it represents a genuine innovation in scientific procedure, in the choice of problems or indeed in any other respect, but rather because some of its proponents claim that it is all of these and more. As a matter of fact, it seems to be nothing else than the ordinary method of research applied to the problem of determining the relative efficiency of different techniques for achieving a given end. To this is added a certain political dictum regarding the superiority of applied, as over against pure science, the superior skill of natural scientists in the study of social processes and similar propositions which have little to do with the methodology or technique of scientific research.

⁷ The Department of Justice, etc.

⁸ This is not by any means always the case. Policy-makers with strong convictions about the correctness of their own perception power are not infrequently disposed to reject as wrong or irrelevant factual representations which contradict their own impressions or which bring to their attention facts which are unpleasant, i.e., which are incompatible with certain measures which they are using to attain a given end. This especially is likely to be the case with those investigations which consist more of the clinical insights of the investigator than of quantitatively handled data. The more amorphous the structure of the data, the less persuasive it is likely to be with a policy-maker with strong personal convictions. There is, also, of course, the policy-maker who orders an investigation to be conducted by social scientists not because he is interested in the results in order to instrument a policy but because it is expedient as a political tactic.

⁹ The general propositions about “human relations” in formal structures, especially in industry, are illustrative here. These practically oriented researches have never rigorously formulated or tested any hypothesis, yet they have to an increasing extent conveyed a general proposition into the minds of policy-makers.