| Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting |
|
We are interested here in what holds people together in an interpersonal communications network, and we are interested in the variety of communications pat terns that take shape under different social conditions . . . . We shall consider these matters under three separate headings, the first of which is concerned with the influence on interpersonal communication of the structure of social connections among individuals. STRUCTURAL CONNECTIONS: THE NETWORKS OF COMMUNICATIONS FLOW Differences in the degree of mutual attraction among individuals, differences in the degree of their interdependence, differences in status, and, of course, mere differences in such things as propinquity or group size will make for significant differences in the rate of contact and communication and often, too, in the con tent of what is communicated. These are some of the elements of group structure, and there are a variety of studies to show how they are related to interpersonal communication. ----------- From Elihu Katz and Paul Lazarsfeld, "Interpersonal Networks: Communicating Within the Group," in Personal Influence (New York: The Free Press, 1955), 84-95. Reproduced with permission of the authors and publisher. ---------------- Festinger et al. demonstrate, for example, how friendship ties operate as links in a communications network, in their studies of the transmission of rumors in two communities.' In "Regent Hill," a rumor spread to 62 percent of those who had close friends in the community, to 42 percent of those who had only acquaintances and to 33 percent of those who claimed no friends at all in the neighborhood. In their Westgate study of the students' housing development, sociometric friendship choices accounted for the direction taken by almost half (six of fourteen) of the instances in which a planted rumor was passed on. Moreno (1953), too, studied a rumor and found that it followed very closely the path he had predicted for it in terms of the "psycho-social networks" which he had mapped out. "Friendship between two people," says Festinger, "implies the existence of an active channel of communication." The sociometric method permits the study of communications flow in terms of an objectively delineable pattern of individual relationships. Thus, a rumor or a bit of news or an action-stimulant introduced into a social group whose socio metric connections are known, can be watched almost the way the doctor watches the flow of chalky liquid introduced into the human body during a fluoroscopic examination. --------------------- 1 Festinger, Schachter and Back (1950), Chap. 7. 2 For an interesting discussion of these networks and their relationship to both flow-of influence and individual and group action, see Moreno's (1953) statement, pp. 440-450. 3 Festinger, Schachter and Back (1950), pp. 125-127. 4 Jennings calls this the "psyche group," contrasting it with a larger, more goal oriented, more formally organized "socio group." The "psyche group" would seem to correspond, generally, to our idea of the primary group. 5 Festinger, Schachter and Back (1950), p. 127. -------------- Rumor is not the only kind of interpersonal communication that has been studied this way. Jennings (1952), for example, tries to trace the transmission process by which relevant knowledge about girls running for election was spread among the 400 occupants at the New York State Training School for Girls. Jennings puts the problem this way: "Without a certain minimum publicity being spread to others by those who know and approve the individual's 'way of leading,' no individual can have a leadership position beyond his immediate interrelationships." She observes that the mechanism for publicity of this sort is the immediate friendship group--"an interpersonal structure where the uniqueness of the individual as a personality is appreciated. . . . " Now, when the over lap of these groups is extensive, that is, to the extent that individuals belong to more than one such group, there will be considerable shared knowledge; but we are told, where it is not well developed, shared knowledge will be restricted. Festinger's finding that rumors do not usually travel outside of self-contained clique structures is another way of expressing this same observation.5 An approach which examines variations of structure in the small group and the effect these have on the character of communications can also be singled out from the prolific work of Festinger and his associates. Here the key variable is the degree of group cohesiveness; and cohesiveness, one might say, is the cement of socio-metric structure. In an experiment by Kurt Back (1952) . . . we learn that when people are more attached to each other, they exert greater influence over each other's opinions and, moreover, are more effective in their influencing. Thus, it will be recalled, when the groups were highly cohesive, the number of at tempts at changing a partner's opinion were both greater in number and more successful than when the interpersonal relations were less cohesive. Similarly, the Westgate study demonstrates that there is greatest uniformity of opinion in those residential courts that are most cohesive, thus implying that there is more, and more effective, communication among the members of these courts.6 Cohesiveness in this case was judged from the allocation of intra-court friendship choices. In addition to friendship channels, there are of course other sorts of inter personal channels. Back, et al. (1950), for example, studied the processes of rumor transmission in an industrial organization. The factory's hierarchy consisted of five levels and a total of 55 members. Information was obtained from the director and from several others about the formal and the informal structures that existed within the organization, and on the basis of this information, a small number of strategically placed "cooperating observers" were recruited by the experimenters. These observers were pledged to secrecy regarding their part in the study and were instructed to make a careful record of the content and the trans mission channels of every rumor-planted one at a time by the experimenters over a four month period-that was passed on to them or that they overheard being told to others.' In this way, there were recorded 17 acts of communication resulting from several rumors whose content related to the entire organization. ----------------------- 6 Ibid., p. 92. 7 By interviewing the entire membership of the organization at the end of the experimental period, the authors were able to ascertain that these cooperators had failed to record only 22 percent of the communications that occurred; in other words, this method yielded 78 percent of the desired information. In general, there was not a great deal of communication concerning these rumors. The authors discuss several reasons for this, including their suspicion that the cooperators themselves-who were chosen because of their key locations hut who were not permitted to pass on anything they heard-may under ordinary circumstances have been important transmitting agents. Back's brief discussion of the assets and liabilities of several of the methods that have been used, or might be used, for the study of rumor transmission is very valuable. ------------------------- Of these, 11 were directed upward in the status hierarchy, four were directed to peers in the hierarchy, and only two communications were directed downward. This study, it is important to remember, is concerned with informal communication patterns in the setting of a highly formal organization. It is interesting to speculate whether these upward-directed acts of informal communication reflect the more formally prescribed work-interaction patterns, or mobility aspirations, or perhaps, friendship patterns which cross status lines. To what extent would communications which are not directly relevant to the organization-say, communications about presidential elections, or World Series scores-flow in the same way? To what extent do communications between, say, social classes in a community or between more and less prestigeful members of a small group parallel this case of the factory? 8 All these are questions for which we do not have simple answers. But since we are concerned specifically with communication in informally organized small groups, perhaps we should attempt, at least, to place these problems in the small group context. If we ask whether communication even in small groups tends to be directed upward (from lower status to higher status members), the answer seems to be yes--but the yes needs considerable elaboration. Homans' analysis of this matter, for example, leads him to suggest, first, that "any single person inter acts most of all with his equals." At the same time, however, the higher a per son's status, the more people will seek to communicate with him. In other words, higher ranking individuals are targets for communications from those below them in rank. Homans adds that these high ranking members, in turn, tend to ad dress themselves to a larger number of group members than to lower ranking individuals.'° Bales' studies at Harvard seem to corroborate these observations, though it is not quite clear to what extent his findings-which are based on discussion groups and problem-solving conferences--are generalizable to the more informal influencing that goes on between friends and neighbors, husbands and wives, workers on the job, etc." Using a device for recording interaction, Bales is able to show that if the members of a group are ranked in order of the frequency with which they speak to others, then it turns out that this same rank order holds for the frequency with which others speak to them. But the most frequent speakers tend to address themselves to the entire group; indeed, it turns out others tend to talk to them as individuals, but their own talk is more often directed to the group as a whole. Furthermore, it is important to note, popularity in the group is also distributed according to the rank order of frequency of speaking, so that the most frequent speakers are also most popular. It follows that person-to-person messages are directed at the most popular group members and thus may be said to move upward in the hierarchy, while communication from one person to several others tends to flow down. ------------- These questions, of course, are raised again in later parts of this book. Specifically, it is asked whether interpersonal influencing in an urban community tends to take place between individuals of like status tor age, or gregariousness) or between individuals of differing status. And, again, it is asked whether the channels of interpersonal communication on one subject-say, fashions--are the same as those employed for another subject-say, politics. See Part Two, Section Three. Later parts of the present section, too, treat the question of multiplicity of channels (see the "situational" aspects of communications, p. 94 if. and p. 100 ff.). For discussion of content factors in upward and downward communication in informal organizations, see, e.g., Homans (1950), particularly p. 461, Kelley (1950), Thibaut (1950). These latter two are important studies of the relationship between member ship in one or another of several hierarchically arranged groups (experimentally induced) and the frequency and content of communications behavior. 9. Homans (1950), p. 184. 10. Ibid., pp. 182-183. In Homans' own words: "The higher a man's social rank, the larger will be the number of persons that originate interaction for him, either directly or through intermediaries . . . (and) the larger the number of persons for whom he originates inter action, either directly or through intermediaries." Authority for substituting "communication" for "interaction" will be found, Ibid., p. 37. 11. Bales (1952) summarizes major findings from his studies which seek to describe uniformities of the following three kinds: (1) "Profile"-the relative frequency of different sorts of substantive acts during the course of a discussion-the ratio of questions to answers, for example; (2) "Phase Movements"-the distribution through time of the group's attention to three qualitative areas of discussion, namely, Orientation (what is it). ---------------------- To his analysis of the direction of flow, Bales adds an analysis of content. It becomes evident that the high ranking people-those who are most popular and most frequent talkers-are also likely to say different things than the low ranking people: The former tend to offer information and proffer opinion, while the latter typically request information and opinion, and express agreement or disagreement. The infrequent speakers, in other words, tend to "react" rather than "initiate," while the more frequent speakers seem to make more influence-attempts. These studies show that social stratification-status and rank-plays an important part in channeling communications flow in small, informal groups, just as it does in more formally organized groups. But now that we have seen some thing of the influence of formal and informal hierarchies and of mutual attractions (friendship, cohesiveness) in making for variations in the networks of inter personal communication, let us consider the consequences for communications behavior of another of the important dimensions of group structure: group size. Again, Bales talks to our point. As groups get larger (in this case, ranging from 3 to 8 people) he finds that more and more communication is directed to one member of the group (the most frequent communicator), thus reducing the relative amount of interchange among all members with each other. At the same time the recipient of this increased attention begins to direct more and more of his re marks to the group as a whole, and proportionately less to specific individuals. ------------------ Evaluation (how do you feel about it) and Control (what shall we do about it); and (3) "Who-to-Whom Matrix"-who says what and how often to whom. 12. Ibid., p. 155. 13. Hare (1952). ------------------------- "The communication pattern," says Bales, "tends to 'centralize,' in other words, around a leader through whom most of the communication flows." 12 But though the discussion in the larger group tends to focus more around one individual, it does not necessarily follow that the degree of consensus achieved will be greater. Quite the contrary, according to one experiment on this subject, which compared five-man and twelve-man discussion groups.'3 The discussants were Boy Scouts who came together to iron out differences in opinion regarding the relative merits of various types of camping equipment, and it was found that the five-man groups achieved a significantly greater degree of consensus than the twelve-man groups. Furthermore, the author reports-on the basis of a questionnaire-that there was greater dissatisfaction with the meeting ex pressed among members of the larger groups. There is evidence, too, that the dissatisfaction was due to the lower level of participation that was imposed on the larger groups by the time limit. This leads the author to suggest that lowered participation results in lowered consensus." Here, then, are a few brief illustrations of the ways in which different structural "arrangements" affect the patterns of interpersonal communication. In our discussion of the strategic points of communication below, we shall return to a consideration of how "location" within a given communication structure constitutes one of the major keys to the study of strategic communications roles- particularly those of leaders and influential. The concept of "centrality," which has already been touched upon, will be found to play a major role. COMPARATIVE CLIMATES: GROUP CULTURE AND COMMUNICATIONS The idea that varying the links between the members of small groups might also result in significant variations in the patterns of communications flow, the volume of communication and even the content has occurred to different re searchers in different ways. The studies we have just finished reporting were primarily concerned with the structure of the group. Here we shall turn to consider the effects on interpersonal communication of varying the "climate," or "culture," within which a group of individuals meets, and the first study we shall re port is a well-known one by Kurt Lewin and his associates. ----------------------- 14 This same study informs us that leaders in the large groups had less influence over the opinions of their groups than leaders of the small groups. Confronting this finding with Bales' report that leaders in larger groups tend to command a position of "centrality" in the communications channels of their groups, seemingly we are led to the conclusion that as the leader gains in centrality he loses in influence. As we shall see later below when we consider the relationship between social "location" and influence, this is not necessarily so. Intuitively, we may suggest that the extent to which there is a "pressure toward uniformity" in the group-that is, the extent that members are dependent on each other for the solution of some problem, or the attainment of a goal-is an important variable here. In the present experiment, a group vote was required, but there was no call for unanimity, or what's more (and this is a point we shall see illustrated below) no interdependence among group members was required insofar as obtaining an "answer" was concerned. Our suspicion is, therefore, that where there is strong pressure toward uniformity and/or considerable inter dependence the more central role of large group leaders is also more influential, whereas under conditions where there is little pressure toward uniformity, the influentiality of the large-group leader is reduced. ------------------------- Together with Ronald Lippitt and Ralph K. White, Lewin set up what is per haps his most famous experiment-that on "experimentally created social climates": "democratic," "authoritarian" and "laissez-faire." 15 We shall not enter into a detailed discussion of the experimental design, except to indicate that each of four matched clubs of 11-year-old children were subjected, in varying sequences, to each of the three "climates," and that these "climates" were deter mined primarily by the carefully rehearsed behavior of the adult group leader. Moreover, the variations that might be products of the leader's personality were controlled by assigning more than one group to each leader and making him responsible for creating more than one kind of "climate." Each club met in the same clubroom setting and each engaged in the same activity. Records of all kinds were kept, including a minute-by-minute analysis of sub-groupings, a quantitative running account of the behavior of the adult leader and of all group interaction, and a continuous stenographic record of all conversations. Although this study was not formulated by its authors in terms of communications problems, many of its results will be of direct import for us. We learn, for example, that the sheer volume of conversation among group members was more restricted in the authoritarian atmosphere 16 than in the democratic or the laissez-faire situations. The more interesting communications data, however, must be described in somewhat more substantive terms. Thus, in the authoritarian groups there were more attempts to attract and hold the leader's attention, although the character of the interaction with the leader was relatively less confiding and less "personal." In the democratic and laissez-faire atmospheres, on the other hand, members made more requests for attention and approval from fellow members than in the authoritarian climates. The laissez-faire groups exceeded the other groups by far in the category of requesting information of the leader, while the democratic groups were freest of all in making suggestions to the leader about group policy. This evidence implies that the democratic group, by virtue of the processes of interchange and decision which were part of its group "climate," probably went much further than the other groups in establishing group-enforced norms of behavior, independent of the adult leader. The pattern of communication deter mined by the democratic "climate," in other words, did not center exclusively on the leader. This interpretation seems to be supported by the results of one of the "test episodes" in this experiment, whereby the leader was summoned from his group unexpectedly, and the reaction, in each "climate," to the leader's sudden departure was observed. When the leader left the authoritarian groups, the authors tell us, " 'working time' dropped to a minimum . . . and most of what was done was in the minutes just after the leader had left the room . . . (but) in the democratic atmosphere the absence or presence of the leader had practically no effect." Primarily, of course, this is evidence that the authoritarian groups de pended completely on the leader's direction and lacked any sort of productive initiative except when the leader was present. Apparently, the group was not strictly a group-or at least a production group-except when cemented together by the leader. The democratic group, on the other hand, seemed to be able to function as an interacting goal-oriented unit without the adult leader. There seems to be evidence here that as a result of group-derived decisions which were the basis of the democratic group "climate," the group had transmitted to its members a set of shared "traditions" which became group property and were maintained even when the leader was absent. ------------------------ 15 For a summary of these studies, see Lippitt and White (1952). 16 Actually, the authoritarian groups did not all behave alike. Some behaved very aggressively during the experiment and others apathetically, but we shall not go into the details of this distinction. --------------------------------- Another climatic difference built into an experimental situation can be examined in Festinger and Thibaut's (1952) study of the effect of varying the "pressure toward uniformity of opinion." We know, for example, that some cultures more than others demand unanimity of opinion in their members; and we know, too, that even a single group in a single culture experiences some situations as permissive and others as demanding, as far as homogeneity of opinion is concerned. The authors "created" three different kinds of experimental "climates." One set of groups was told that the experimenter's interest was in observing "how a group went about coming to a unanimous decision." Thus, a "high" pressure toward uniformity of opinion was induced in these groups. A second set of groups was informed about the solution that some experts had proposed for a particular problem and were told that the group would be rated according to the number of its members who arrived at the "correct" solution; these instructions were designed to introduce "medium" pressure toward uniformity. The third set of groups was merely informed that the experimenter wanted to study a problem-solving group, and thus no external pressure toward uniformity was introduced. The results of the study clearly demonstrate the authors' hypothesis that "as pressure toward uniformity increases, both pressure to communicate and readiness to change also increase. Since both of these factors are conducive to change, there should be increasing change toward uniformity of opinion as the pressure toward uniformity increases." By means of these studies we have tried to show that the "climates" or "cultures" within which individuals find themselves, influence the patterning of inter personal communication. Like the structural characteristics of interpersonal relations, "climatic" characteristics must be accounted for if we are properly to analyze the varied character of the channels of flow of information and influence among interacting individuals. ----------------------- 17. Lippitt and White (1952), p. 348. ------------------------- SITUATIONS: COMMUNICATIONS CONTENT AND INTERPERSONAL RELATIONS Individuals in modern society are usually members of more than one group, and have interpersonal ties in different sorts of situations. Groups of individuals, furthermore, share certain interests on the level of the group as a whole, and di vide into sub-groups, as far as certain other interests are concerned. Therefore, the very same group may engender quite different communications networks relative to different interests and different situations. Important questions emerge from this situational approach to interpersonal relations: How do communication patterns change in different social situations? What kinds of communication con tent flow through which kinds of interpersonal networks? For example, is the family group important in generating or transmitting political opinions or is it the work group-and for which kinds of people? Because most small-group studies are conducted in laboratories, there are very few answers for this kind of question in actual research reports. Some leads, however, are available to us, notably from those who have stepped outside the confines of the small-group laboratory to study "natural" groups. The cardinal principle that emerges from all work in this area is that shared interest in a given subject is the basis for interpersonal networks of communication. In other words, as Allport and Postman (1947) put it in their study of rumor, "A rumor public exists wherever there is a community of interest." For example, financial rumors circulate among those who are likely to be affected by financial ups and downs." Festinger and his associates (1950) provide several illustrations for this principle in their studies of housing communities." When a rumor concerning the future of the children's nursery school was circulated, these investigators were able to ascertain that it had reached 62 percent of the people in the community with children of nursery school age, and only 28 percent of the people who had no children of that age. When a rumor hostile to the existence of the Tenants' Council was "planted" in Westgate, it tended to reach people who were particularly associated with the organization and among those who heard it, it was passed on, in turn, only by those who were active participants in the organization, or who were members of a family with active participants, but not at all by any others. Similarly, in the study of "planted" rumors in an industrial organization, Back et al. (1950) find that when a rumor pertained specifically to a small group, it tended to spread very quickly to the members of that group and not to go beyond the few people who were vitally concerned. -------------------- "Allport and Postman (1947), p. 180. 19 Festinger, Schachter and Back 119501, Chap. 7. --------------------- Here, then, it appears that networks of communication exist not only within the web of friendship networks but within the web of shared interests and concerns as well. Festinger et al. (1950), however, suggest several important amendments to this conclusion, though their proof for these is not quite conclusive. In terms of their Westgate study of attitudes toward a Tenants' Council, they tell us, first of all, that information favorable to the Tenants' Council was much more actively communicated within those residential courts whose members were favorable to the Council than those courts which were unfavorable. Secondly, they say that there seemed to be little or no communication on matters relevant to the Tenants' Council between members of courts which had conflicting court attitudes toward the Council. On a more abstract level, these findings seem to point to the generalization that there will be a greater amount of communication concerning a given matter among people who share a concern in the matter when (a) their point of view on a particular concern is homogeneous and (b) when the content of the communication is favorable to their shared point of view." REFERENCES ALLPORT, GORDON W. and Leo J. POSTMAN (1947), The Psychology of Rumor, New York: Holt. BACK, KURT W. et al. (1950), "A Method of Studying Rumor Transmission," in Festinger et al., Theory and Experiment in Social Communication, Ann Arbor, Mich.: Research Center for Group Dynamics, University of Michigan. BACK, KURT W. (1952), "Influence Through Social Communication," in Swanson, Newcomb and Hartley, eds., Readings in Social Psychology, New York: Holt. BALES, R. F. (1952), "Some Uniformities of Behavior in Small Social Systems," in Swanson, Newcomb and Hartley, eds., Readings in Social Psychology, New York: Holt. PESTINGER, LEON, S TANLEY S CHACHTER and KURT BACK (1950), Social Pressures in Informal Groups, New York: Harper & Row. HOMANS, GEORGE C. (1950), The Human Group, New York: Harcourt, Brace & World. JENNINGS, HELEN H. (19521, "Leadership and Isolation," in Swanson, Newcomb and Hartley, eds., Readings in Social Psychology, New York: Holt. KELLEY, HAROLD H. (1950), "Communication in Experimentally Created Hierarchies," in Festinger et al., Theory and Experiment in Social Communication, Ann Arbor, Mich.: Research Center for Group Dynamics, University of Michigan. L IPPITT, RONALD and RALPH K. WHITE (1952), "An Experimental Study of Leadership and Group Life," in Swanson, Newcomb and Hartley, eds., Readings in Social Psychology, New York: Holt. THIBAUT, JOHN (1950), "An Experimental Study of the Cohesiveness of Under-priviledged Groups," Human Relations, Vol. 3, pp. 251-278. Festinger, Schachter and Back (1950), p. 129. We are not told, however, that communications unfavorable to the Council were more actively spread in the unfavorable courts. Moreover, what we are told is that communications hostile to the Council spread most of all and most quickly to those who were active participants in the Council, most but not all of whom, we know, were favorable, too. That is one objection to the conclusions reported. A second objection stems from the inference that since rumors did not spread from the favorable to the unfavorable courts, or vice versa, that communication between groups with different standards is therefore limited. This cannot he conclusively demonstrated since groups with different standards tended also to correspond to different friendship cliques; hence it would be sufficient, therefore, to point out that communications were limited by friendship channels-something we have already been told. Also in Part 5:
|
|
Top of Page | Index | Home |