Announcement has just been made that Experiments on Mass Communication will be published by the Princeton University Press during the month of June. This volume is the third in the series entitled Studies in Social Psychology in World War II, prepared by former members of the Research Branch, Information and Education Division of the War Department.1As stated in the description of the first two volumes in the March issue of Items, these reports were sponsored by a special committee of the Social Science Research Council and financed by grants to the Council from the Committee of Trustees on Experimental Programs associated with the Joint Army and Navy Committee on Welfare and Recreation and from the Carnegie Corporation of New York. The members of the committee are: Frederick Osborn (chairman), Leonard S. Cottrell, Jr., Leland C. DeVinney, Carl I. Hovland, John M. Russell, and Samuel A. Stouffer.

Experiments on Mass Communication is an analysis of the impact upon the soldier of Army educational and indoctrination films, based on a series of studies conducted by the Experimental Section of the Research Branch. This Section had as its primary mission the evaluation, by means of controlled experiments, of various Army programs designed to make the soldier aware of the ideological issues behind the war. At the same time the Experimental Section was called upon by other War Department agencies to evaluate the methods and media that were being used, on an unprecedented scale, in the mass communication of purely technical instruction. Studies of both types of communication—for indoctrination and for instruction—are included in this volume.

“The studies reported are those whose methods or results have generality beyond the immediate practical objectives initially involved in the experiments.”

The studies reported are those whose methods or results have generality beyond the immediate practical objectives initially involved in the experiments. The authors have attempted to give perspective to the studies by outlining the general field of research on the effects of mass communication media and indicating requirements for an adequate systematization of the field. An extensive appendix is devoted to presentation of some of the methodological problems encountered in carrying out the work. The volume is thus of interest both to those concerned with the production and utilization of communication media and to those concerned with research on the effects of communications on the audience to which they are presented.

Extensive studies conducted to evaluate the Army’s “Why We Fight” orientation films are described, using “The Battle of Britain” to illustrate in considerable detail the procedures employed and kinds of results obtained. In these studies an analysis was made of the audience’s evaluation of the films as gauged by questionnaires and group interviews. However, this basis for evaluation was subsidiary to the main purpose of the studies, which was to determine by controlled experimental comparisons, the actual effects of the films on men’s knowledge and opinions concerning the material covered. Results for all the orientation films studied were consistent in revealing that the films produced sizable increases in men’s factual knowledge, but more limited effects on opinions or interpretations of the facts, and little or no effect on motivations. Hypotheses to account for the relative ineffectiveness of the films in the latter areas are discussed in detail, with reference to relevant data where available. A study of one of the films, in which measurements were made both shortly after the showing of the film and after a considerable lapse of time, suggests that whereas normal forgetting occurred with respect to the large immediate gains in factual knowledge, forgetting did not appear in the case of some of the opinion changes produced, and in some instances there were actual increases in effect with the lapse of time.

A considerable number of insistent requests were received from various War Department agencies for studies of the relative effectiveness of one medium of communication versus some alternative medium. The limitations of such research are discussed in terms of the inherent limitations on the generality of the findings, and it is pointed out that even for answering the empirical question whether medium A or medium B is better for some particular purpose a sizable sample of products of each medium is required, for the same reason that a sample of audience members is needed. The role of such research is shown to be primarily that of furnishing “hunches” and hypotheses for further study, and of providing means of analyzing the effects on different portions of the audience.

From a series of studies evaluating indoctrination and training films it was possible to make an extensive analysis of the role of various demographic and other factors. Surprisingly, the only variable consistently found to be closely related to the effects of the communications studied was intellectual ability, as indicated by men’s scores on the Army General Classification Test or by the number of years of schooling completed. In the transmission of information the relationship between amount learned and intellectual ability was strongly positive. Parallel analysis for opinion change, on the other hand, showed changes on some items to be positively related and on others negatively related to intellectual ability. It is shown that an additional factor termed acceptance is of importance in the analysis of opinion changes, and that the changes in belief found to be positively correlated with intelligence were those which were initially more prevalent among the more intelligent. Other changes, however, tend to be negatively related to intellectual ability.

“In these studies the method of controlled variation permitted conclusions less limited in generality than those possible from the purely evaluative studies.”

A significant portion of the book deals with experiments which were set up specifically to isolate the effects of important theoretical variables. In these studies the method of controlled variation permitted conclusions less limited in generality than those possible from the purely evaluative studies. One of these experiments was concerned with a technique for inducing the audience to participate more actively in the learning of material presented in the film. The role of this active participation procedure was simultaneously studied in relation to the effects of motivation, difficulty of the material to be learned, and the learning ability of the individual members of the audience. By means of this controlled variation of several different theoretical factors acting in combination, the findings on the role of the active participation procedure may be generalized with reference to a variety of relevant circumstances. It was found that under all conditions this procedure increased the amount learned but that the improvement was greater when motivation was low, when the material was difficult, and when the individual learner was relatively low in intellectual ability. With all variables acting in concert a gradient of effects of participation was obtained, ranging from practically no effect for intelligent, motivated men learning easy material to a maximal effect with less intelligent, unmotivated men learning difficult material.

In another study employing controlled variation of content, the effects of presenting both sides of the argument on a controversial issue were compared with the effects when a one-sided version was given. In each case the communication attempted to influence a given opinion in the same direction; the only difference between the two was the inclusion or noninclusion of arguments on the other side. The effects of the content variable were determined as a function of initial opinion and intellectual ability, two additional variables of theoretical interest. It was found that inclusion of arguments on the other side of the controversial issue was superior in producing the intended opinion change for those of greater intellectual ability, regardless of initial position. It was also found that for the group as a whole the argument with “both sides” was superior for individuals initially opposing the point of view being fostered. Further, the argument including “both sides” was least effective with the less intelligent men who were initially in favor of the opinion being fostered, the results even suggesting a slight adverse effect. Thus the most effective use of the arguments for and against a controversial issue was found to depend on the composition of the members of the audience with respect to initial opinion and intellectual ability.

As mentioned earlier, the primary purpose of the experimental studies was to provide practical answers to War Department problems. Most of these questions took the form of a request for a practical evaluation of a specific program or a comparison of existing programs. But it was found that when the experiment could be put in the form of controlled variation studies of the kind described above, the results were likely to be more valuable in providing practical suggestions for effective methods of preparing communications, largely because of the greater generality of the findings. Thus experiments designed to extend scientific knowledge appeared to have been more practical than those designed solely for purposes of “practical” evaluation.

This essay originally appeared in Items Vol. 2, No. 1 in June of 1949. Visit our archives to view the original as it first appeared in the print editions of Items.

Banner photo credit: National Archives.

References:

1
As stated in the description of the first two volumes in the March issue of Items, these reports were sponsored by a special committee of the Social Science Research Council and financed by grants to the Council from the Committee of Trustees on Experimental Programs associated with the Joint Army and Navy Committee on Welfare and Recreation and from the Carnegie Corporation of New York. The members of the committee are: Frederick Osborn (chairman), Leonard S. Cottrell, Jr., Leland C. DeVinney, Carl I. Hovland, John M. Russell, and Samuel A. Stouffer.