Since propaganda deals by definition with controversial matters, its ultimate audiences, whether reached by direct or indirect media, can be ranged into three groups: (1)those who are initially predisposed to react as the propagandist wishes, (2) those who are neutral or indifferent, and (3) those who are antagonistic. It is advisable in many cases to include symbolism and to select media appropriate to many positions along this continuum.
Recent research tends to indicate that the most dependable result ofpropaganda is likely to be an increase in the resolve or the efforts of those who are already the propagandist’s friends. No matter how great the volume or symbolic intensity of propaganda, neutrals and opponents are likely to be little’affected unless the propaganda is reinforced by relevant nonpropaganda transactions or other events. For example, propaganda that aims to induce loyalty to a given regime among a landlord-ridden population that is experiencing the modern “revolution of expectations” may have to be reinforced by delivering as well as promising land and tax reform and bona fide physical protection.
Measurement of propaganda effects.
The problem of measurement is almost as broad as the field of behavioral research methodology. The modern world is seething with rival propaganda campaigns and counterpropaganda and with countless other symbolic transactions, The problem of disentangling the effects of one’s own propaganda from the effects of the other transactions is often insoluble. ‘Yet it is occasionally possible to conduct research whose results can be viewed with moderate confidence.
Reasonably dependable quantitative evidence as to the contents of propaganda can be obtained by the method known as “content analysis.” The numbers of column inches of printed space or seconds of radio or television time that propaganda occupied can be tabulated. The symbols and themes it contained can be categorized, as already indicated, in terms of expressed or implied demands for actions of various types and in a number of other psychologically or socially significant ways.
Fairly objective evidence as to the intensity and semantic significance of propaganda (i.e.; the cognitive and affective associations it evokes in given reactors) can be gathered by extended intensive interviews (of a psychoanalytic or psychiatric type) with small, carefully drawn samples of the intended audience. If this audience is in a place where freedom of such inquiry is restricted (i.e., most of the world), the next best method is to interview any presumably informed persons who can be reached–e.g., refugees, expellees, or scholars concerned with the area .
Sometimes participant observers can be sent to the relevant places. Voting statistics, press reports, or the speeches and other actions of affected leaders can also give clues. Evidence on the size and composition of the intermediate (including “relay point”) audiences and the ultimate audiences can be obtained from extensive sample surveys, press repoPts, and leaders’ reactions. Where printed or telecommunications media are used, their readership or listenership figures can perhaps be obtained. If public meetings or demonstrations are involved, there may be observers’ reports.
Experiments and panel interviews.
Evidence that ensuing behavior of the audience–for instance, its vote for candidate X or its buying of product Y–is due in whole or in part to the propaganda and not to something else remains far from conclusive, however, except in the rare situations where something like an experiment is possible. In some cases, matched groups can be compared–one of them exposed to the propaganda and the other not, or one of them exposed to version A of the propaganda and another to version B, and so on. In some cases, the propaganda reaching one group can be abruptly stopped or intensified and some of the presumably consequent reactions may be observed. However, there is always the possibility that it was not one’s own propaganda that brought about the changes, but someone else’s, or that the changes were caused by some unknown third factor or set of factors. There is also the problem of “sleeper effects”–long-delayed reactions that may not become visible until the propaganda has worked its way through or around resistances that it may encounter deep down in the reactor’s unconscious or until obstacles to expression reactions (e.g., political policemen or suspected informers) have left the reactor’s environment. And there is the possibility that the propaganda may have “boomerang effects”-effects the opposite of those intended–or combinations of boomerang and desired effects. Research design that does not allow for all these possibilities is of doubtful evidential value.
In view of the extreme difficulty of tracing effects of propaganda upon reactors in their native habitats, a great deal of . effort has been spent in recent years on strictly controlled experiments and repeated semi-intensive interviews (“panel interviews”) with small matched groups, with a view to establishing general principles of propaganda and persuasion. Among the many factors examined have been the relative credibility and acceptability,. to given audiences, of different sources of infonnation, advice, and opinion; the uses of different propaganda contents aimed at the same results; and the effects of different ways of arranging and presenting the same contents. However, reliance on such findings is notably limited by the fact that the behaviour of reactors available for such testing mayor may not be representative of the behaviour of those actual audiences in whom the propagandist is interested. It seems probable that effects of propaganda among actual reactors can in most cases only be estimated, not “measured” scientifically, .and that the most valid estimates are likely to be made by persons combining considerable training in the methods of social science with considerable direct experience among the reactors un.} analysis.