Hostname: page-component-8448b6f56d-m8qmq Total loading time: 0 Render date: 2024-04-24T13:22:44.879Z Has data issue: false hasContentIssue false
  • English
  • Français

An Introduction to the Theory and Measurement of Influence*

Published online by Cambridge University Press:  02 September 2013

James G. March
James G. March
Affiliation:
Carnegie Institute of Technology
Get access Rights & Permissions [Opens in a new window]

Extract

The science of politics is a science of human behavior. It concerns itself with a specific segment of the activities of humans—those which either take place in, or have a clearly discernible effect upon, the formal governmental machinery of the community. The characteristic feature of a political scientist, therefore, is not his unique theoretical framework but his special empirical interest. Two main consequences follow. First, it is trivially true, and widely recognized, that the major concepts of other behavior sciences are necessarily an integral part of the study of political behavior. Second, it is equally true that, within the social sciences, it is the responsibility of political science to develop those elements of behavior theory that are particularly relevant for the analysis of action in the sphere of politics.

Much of current empirical and theoretical work in political analysis is organized around the observation that many political data can be conceived to represent results of mechanisms for decision-making used (consciously or unconsciously) by individuals or collectivities. In a similar fashion, students of a significant number of other types of behavior have tended to formulate their problems within a decision-making framework. When one examines these apparently disparate branches of behavior theory, it is difficult to avoid the conclusion that there exist potentially fruitful parallelisms among such theories as those of consumer behavior, administrative behavior, price setting, legislative enactments, propaganda, learning, foreign affairs, and social control.

Type
Research Article
Information
American Political Science Review , Volume 49 , Issue 2 , June 1955 , pp. 431 - 451
Copyright
Copyright © American Political Science Association 1955

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See, for example, Dahl, Robert A., Congress and Foreign Policy (New York, 1950)Google Scholar, Part I; Lasswell, Harold D. and Kaplan, Abraham, Power and Society (New Haven, 1950), pp. 74–83Google Scholar; Lipset, Seymour M., Lazarsfeld, Paul F., Barton, Allen H., and Linz, Juan, “The Psychology of Voting: An Analysis of Political Behavior,” in the Handbook of Social Psychology, ed. Lindzey, (Cambridge, 1954), Vol. 2, pp. 1124–75Google Scholar; Shubik, Martin, Readings in Game Theory and Political Behavior (New York, 1954)Google Scholar; Simon, Herbert A., Smithburg, Donald W., and Thompson, Victor A., Public Administration (New York, 1950)Google Scholar, Ch. 3; Snyder, Richard C., Bruck, H. W., Sapin, Burton, Decision-Making as an Approach to the Study of International Politics (Princeton, 1954)Google Scholar.

2 Edwards, Ward, “The Theory of Decision-Making,” Psychological Bulletin, Vol. 51, pp. 380–417 (July, 1954)CrossRefGoogle ScholarPubMed; Lazarsfeld, Paul F., Katz, Elihu et al. , Personal Influence (Glencoe, 1954)Google Scholar; Savage, L. J., The Foundations of Statistics (New York, 1954)Google Scholar; Decision Processes, ed. Thrall, Coombs, and Davis, (New York, 1954)Google Scholar.

3 Note that this does not commit one to a Newtonian psychology; nor is it anticipated that an analogical approach to the theory of influence will prove particularly useful.

4 Because it partially avoids this pitfall by relying heavily on learning theory, the recent Yale volume on persuasion has considerable merit. See Hovland, Carl I., Janis, Irving L., and Kelley, Harold H., Communication and Persuasion (New Haven, 1953)Google Scholar.

5 See, for example, Koopmans, Tjalling C., “Identification Problems in Economic Model Construction,” in Studies in Econometric Method, ed. Hood, and Koopmans, (New York, 1953), pp. 27–48Google Scholar.

6 A ‘closed loop’ system is one in which the variables within the system are related in such a way that one can trace a chain of relationships from a given variable through other variables and back to the original starting point without retracing one's steps. Thus a fairly common set of hypotheses in voting behavior forms the following closed loop: voting intention→exposure to party position on the state of the economy→perception of the state of economy→voting intention. An ‘open loop’ system is one in which such circles do not exist. The terms were borrowed originally from servo-mechanics, where they are used primarily to describe different types of automatic control devices.

7 Hovland, Communication and Persuasion, Ch. 8, and the articles there cited.

8 To be sure, one can define something in any way he pleases. What is meant is that a uni-dimensional influence index will rarely be of much utility.

9 March, James G., “Husband-Wife Interaction over Political Issues,” Public Opinion Quarterly, Vol. 17, pp. 461–70 (Winter, 1953–1954)CrossRefGoogle Scholar; Rubenstein, Albert H., “Problems in the Measurement of Interpersonal Communication in an Ongoing Situation,” Sociometry, Vol. 16, p. 86 (Feb., 1953)CrossRefGoogle Scholar.

10 Galloway, George, The Legislative Process in Congress (New York, 1953)Google Scholar, Ch. 14.

11 If the index has only two values, it is possible to talk of a set theoretical model for describing the relationship, denning a partitioning of the joint decisions of A and B into mutually exclusive sets of decisions made by either A or B alone. Comparison of influence positions then depends either upon the development of comparison values or upon the existence of a situation in which the set of decisions made by A alone at time t 0 is properly included in the set made by A alone at time t 1, or in the set made by B alone at some given time. If a multi-valued index is used, vector notation seems to be indicated and the development of standard measurement units to be of central importance.

12 Lasswell, and Kaplan, , Power and Society, p. 76Google Scholar; Truman, David B., The Governmental Process (New York, 1951), pp. 188–93Google Scholar.

13 For the moment, at least, we will not consider the complications introduced by the possibility that the President's announcement is induced by certain previous activities on the part of the Congressman.

14 Wold, H. O. A., “Causality and Econometrics,” Econometrica, Vol. 22, pp. 162–77 (April, 1954)CrossRefGoogle Scholar. For a general discussion of the function of explanatory laws in the construction of theory, see Hull, Clark L., Principles of Behavior (New York, 1943)Google Scholar, Ch. 1.

15 Simon, Herbert A., “On the Definition of the Causal Relation,” Journal of Philosophy, Vol. 49, pp. 517–28 (July 31, 1952)CrossRefGoogle Scholar; and “Notes on the Observation and Measurement of Political Power,” Journal of Politics, Vol. 15, pp. 500–16 (Nov., 1953)CrossRefGoogle Scholar.

16 Simon, , “Notes” (cited in note 15), pp. 513–14Google Scholar; March, , “Husband-Wife Interaction” (cited in note 9), pp. 469–70Google Scholar.

17 Note the problem of spurious correlation that exists here as elsewhere in the treatment of causality. The fact that the behaviors of the premiers of Hungary and Roumania appear to co-vary is not sufficient evidence upon which to establish an influence relationship between them.

18 Special acknowledgement is due to Professors Harold Guetzkow and Herbert Simon and to Messrs. Douglas K. Mims, Frederick Stern, and Fred M. Tonge for their criticisms and suggestions with respect to the model outlined in this section.

19 For a similar definition, see Lindsley, Donald B., “Electroencephalography,” in Personality and the Behavior Disorders, ed. Hunt, (New York, 1944), Vol. 2, p. 1033Google Scholar.

20 Excluded, consequently, from the present discussion is the influence analysis of a wholly self-contained system—that is, one in which all variables are determined by system variables exclusively. This applies whether the system involved is the individual personality or a number of such personalities.

21 Osgood, Charles E., Method and Theory in Experimental Psychology (New York, 1953)Google Scholar.

22 Lindsley, Donald B., “Emotion,” in the Handbook of Experimental Psychology, ed. Stevens, (New York, 1951), pp. 473–516Google Scholar.

23 Ashby, William R., Design for a Brain (New York, 1952)Google Scholar; Hebb, Donald O., The Organization of Behavior (New York, 1949)Google Scholar.

24 Wiener, Norbert, Cybernetics (Cambridge, 1948)Google ScholarPubMed.

25 Rudolph Carnap and Y. Bar-Hillel, An Outline of a Theory of Semantic Information, Technical Report No. 247, Research Laboratory of Electronics, Massachusetts Institute of Technology, October 27, 1952.

26 Lewin, Kurt, The Conceptual Representation and the Measurement of Psychological Forces (Durham, 1938)CrossRefGoogle Scholar.

27 Arrow, Kenneth J., “Alternative Approaches to the Theory of Choice in Risk-Taking Situation,” Econometrica, Vol. 19, pp. 404–32 (Oct., 1951)CrossRefGoogle Scholar; von Neumann, John and Morganstern, Oskar, Theory of Games and Economic Behavior (Princeton, 1947), pp. 15–31Google Scholar.

28 The latter can be treated as a special case of the former if the probability web be taken as completely denned but having frequent instances where p= 0.

29 One example of such clarification is presented in the present section. The writer is persuaded it is far from an isolated case. For example, some of the respects in which the problem of generality is attacked have been indicated in the discussion of influence processes (e.g., the elements of similarity between habit—and therefore the extended literature on habit—and legitimacy). With respect to the dimensionality of an influence relationship, the model suggests that in addition to the subject matter under consideration it may be necessary to differentiate influence over alternative suggesting from influence over choice among given alternatives, influence over highly specific acts or attitudes from influence over underlying value premises, etc.

30 Friedrich, Carl J., Constitutional Government and Democracy (Boston, 1941), pp. 589–91Google Scholar; Simon, , “Notes” (cited in note 15), pp. 505–6Google Scholar.

31 For example, Hunter, Floyd, Community Power Structure (Chapel Hill, 1953)Google Scholar; Hurwitz, J. I. et al. , “Some Effects of Power on the Relations among Group Members,” in Group Dynamics, ed. Cartwright, and Zander, (Evanston, 1953), pp. 483–92Google Scholar; Ronald Lippitt et al., “The Dynamics of Power,” ibid., pp. 462–82.

32 For example, Bieri, James, “Changes in Interpersonal Perceptions Following Social Interaction,” Journal of Abnormal and Social Psychology, Vol. 48, pp. 61–66 (Jan., 1953)CrossRefGoogle ScholarPubMed; Cromwell, Harvey and Kunkel, Richard, “An Experimental Study of the Effect on the Attitude of Listeners of Repeating the Same Oral Propaganda,” Journal of Social Psychology, Vol. 35, pp. 175–84 (May, 1952)CrossRefGoogle Scholar; Davies, James C., “Some Relations between Events and Attitudes,” this Review, Vol. 46, pp. 777–89 (Sept., 1952)Google Scholar; Festinger, Leon et al. , “The Influence Process in the Presence of Extreme Deviates,” Human Relations, Vol. 5, pp. 327–46 (Nov., 1952)CrossRefGoogle Scholar.

33 For example, Kelley, Harold H., “The Warm-Cold Variable in First Impressions of Persons,” Journal of Personality, Vol. 18, pp. 431–39 (June, 1950)CrossRefGoogle ScholarPubMed; Moos, Malcolm and Koslin, B., “Prestige Suggestion and Political Leadership,” Public Opinion Quarterly, Vol. 16, pp. 77–93 (Spring, 1952)CrossRefGoogle Scholar; Parrish, Jack A. and Campbell, Donald T., “Measuring Propaganda Effects with Direct and Indirect Attitude Tests,” Journal of Abnormal and Social Psychology, Vol. 48, pp. 3–9 (Jan., 1953)CrossRefGoogle ScholarPubMed; Schachter, Stanley and Hall, Robert, “Group-Derived Restraints and Audience Persuasion,” Human Relations, Vol. 5, pp. 397–406 (Nov., 1952)CrossRefGoogle Scholar.

34 One variation of the more standard opinion-change techniques warrants specific mention. Where the influence relationship studied involves a simple dyad, it is possible to use the revealed-differences method. Two respondents are presented with problems, each of which admits of only two possible responses. They are then asked to reconcile any differences that arise and arrive at a joint position for each question. The proportion of decisions “won” by the most-winning person forms an index of the influence differential. To the extent that it is possible to assume the equal weighting of decisions, such a method is feasible and provides a reasonably straightforward way of comparing different relationships. See Strodtbeck, Fred L., “Husband-Wife Interaction over Revealed Differences,” American Sociological Review, Vol. 16, pp. 468–73 (Aug., 1951)CrossRefGoogle Scholar. Recently, Strodtbeck has been experimenting with a similar technique in three-person groups.

35 Stouffer, Samuel A. et al. , Measurement and Prediction (Princeton, 1950)Google Scholar.

36 Kelman, Herbert C., “Attitude Change as a Function of Response Restriction,” Human Relations, Vol. 6, pp. 185–214 (Aug., 1953)CrossRefGoogle Scholar.

37 Gorden, Raymond L., “Interaction between Attitude and the Definitioi of the Situation in the Expression of Opinion,” American Sociological Review, Vol. 17, pp. 50–58 (Feb., 1952)CrossRefGoogle Scholar.

38 Jarrett, Rheem F. and Sheriffs, Alex C., “Propaganda, Debate, and Impartial Presentation as Determiners of Attitude Change,” Journal of Abnormal and Social Psychology, Vol. 48, pp. 33–41 (Jan., 1953)CrossRefGoogle ScholarPubMed.

39 Bovard, Everett W., “Group Structure and Perception,” Journal of Abnormal and Social Psychology, Vol. 46, pp. 398–405 (July, 1951)CrossRefGoogle ScholarPubMed.

40 Janis, Irving L. et al. “Effects of Preparatory Communications on Reactions to a Subsequent News Event,” Public Opinion Quarterly, Vol. 15, pp. 487–518 (Fall, 1951)CrossRefGoogle Scholar.

41 Preston, Malcolm G. and Heintz, Roy K., “Effects of Participatory vs. Supervisory Leadership on Group Judgment,” Journal of Abnormal and Social Psychology, Vol. 44, pp. 345–55 (July, 1949)CrossRefGoogle ScholarPubMed.

42 Hovland, Carl I. and Weiss, Walter, “The Influence of Source Credibility on Communication Effectiveness,” Public Opinion Quarterly, Vol. 15, pp. 635–50 (Winter, 1951–1952)CrossRefGoogle Scholar.

43 Asch, Solomon E., “Effects of Group Pressure upon the Modification and Distortion of Judgments,” in Groups, Leadership, and Men ed. Guetzkow, (Pittsburgh, 1951) pp. 177–90Google Scholar.

44 One method of avoiding this difficulty is to prohibit communication between respondents and present them with problems completely alien to their experience. Such procedures pose considerable problems of interpretation, but they do provide interesting and important data on the mechanisms of personal adjustment used by the satellite individual in a situation of uncertainty.

45 The recording of interactive events has been used by several observers in one form or another, probably the best known schemes being those developed by Bales and Carter. See Bales, Robert F., Interaction Process Analysis (Cambridge, 1950)Google Scholar; Carter, Launor et al. , “A Note on a New Technique of Interaction Recording,” Journal of Abnormal and Social Psychology, Vol. 46, pp. 258–60 (April, 1951)CrossRefGoogle ScholarPubMed. In essence, the observation of influence in this case is made by attempting to observe the in-process interaction.

46 Mills, Theodore M., “Power Relations in Three-Person Groups,” American Sociology Review, Vol. 18, pp. 351–57 (Aug., 1953)CrossRefGoogle Scholar; Polansky, Norman et al. , “An Investigation of Behavioral Contagion in Groups,” Human Relations, Vol. 3, pp. 319–48 (Nov., 1950)CrossRefGoogle Scholar.

47 Back, Kurt W., “Influence through Social Communication,” Journal of Abnormal and Social Psychology, Vol. 46, pp. 9–23 (Jan., 1951)CrossRefGoogle ScholarPubMed.

48 Festinger et al., “Influence Process” (cited in note 32).

49 Lippitt et al., “Dynamics of Power” (cited in note 31).