Hostname: page-component-848d4c4894-sjtt6 Total loading time: 0 Render date: 2024-06-22T12:09:37.806Z Has data issue: false hasContentIssue false
  • English
  • Français

Discerning a Causal Pattern among Data on Voting Behavior*

Published online by Cambridge University Press:  01 August 2014

Arthur S. Goldberg
Arthur S. Goldberg*
Affiliation:
University of Rochester
Get access Rights & Permissions [Opens in a new window]

Extract

The present analysis is devoted to making an empirically based choice among alternate causal explanations. This entails making causal inferences from statistical correlations. While this might, at one time, have constituted a heresy, I believe that the procedure to be followed here will soon be a part of statistical orthodoxy.

This is not the place for an extended philosophical discussion of the problem of causality. Yet I would like to make my position on the problem as clear as concise presentation will permit. My basic sympathies are with that school which argues that scientifically relevant causal explanation inheres only in our theories, i.e., that the explained event takes the shape which it does because our postulates and logic preclude any other shape on pain of being themselves incorrect. However, the development of such theory, containing such postulates, is usually the product of an inspired insight on the part of one thoroughly immersed in the manifestations of the empirical phenomenon under consideration. The production and verification of such insight in a systematic and reproducible way is the goal of inductive research. Where controlled experimentation is possible, Mill's canons may apply. Where such experiments are either impossible or impracticable, statistical inference becomes necessary. It is in this situation that the present approach, based upon a model developed by Herbert Simon and others, seems justified.

Simon's model is designed to capture the asymmetry in our notions of causality. When one speaks of A as a cause of B, one usually has in mind a unidirectional forcing, and not merely a covariation, or phased covariation.

Type
Research Article
Information
American Political Science Review , Volume 60 , Issue 4 , December 1966 , pp. 913 - 922
Copyright
Copyright © American Political Science Association 1966

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 Cf. Miller, Warren E. and Stokes, Donald E., “Constituency Influence in Congress,” this Review, (03, 1963), 45–56Google Scholar; Pelz, Donald C. and Andrews, Frank M., “Causal Priorities in Panel Study Data,” American Sociological Review, 24 (12, 1964), 836–847CrossRefGoogle Scholar; Alker, Hayward R. Jr., Mathematics and Politics (New York: The Macmillan Company, 1965), chap. VIGoogle Scholar.

2 Cf. Brown, Robert, Explanation in Social Science (Chicago: Aldine Publishing Company, 1963), chap. XIGoogle Scholar. See also Hanson, Norwood Russell, Patterns of Discovery (Cambridge: Cambridge University Press, 1958, chap. IIIGoogle Scholar; and Riker, William H., “Causes of Events,” Journal of Philosophy, 55 (1958), 281–291CrossRefGoogle Scholar.

3 See Simon, Herbert A., Models of Man (New York: John Wiley & Sons, Inc., 1957), chaps. I–IIIGoogle Scholar; and Blalock, Hubert M. Jr., Causal Inferences in Nonexperimental Research (Chapel Hill: The University of North Carolina Press, 1964), chaps. I–IIIGoogle Scholar. These authors, in turn, have drawn heavily upon the work of econometricians. Among basic sources are, for example, Frisch, Ragnar, Statistical Confluence Analysis by Means of Complete Regression Systems (Oslo: Universitets Økonomiske Institutt, 1934)Google Scholar and Koopmans, T. C. (ed.), Statistical Inference in Dynamic Economic Models (New York: John Wiley and Sons,1950)Google Scholar.

4 Simon, op. cit., chap. I.

5 Blalock, op. cit., pp. 52–60.

6 See ibid., pp. 46–54.

7 See, for example, Holte, Fritz C., Economic Shock-Models (Oslo: Norwegian Universities Press, 1962), pp. 14–17Google Scholar.

8 See Hanson, op. cit., chap. IV.

9 The body of data used consisted of that generated in the 1956 election study (#417) by the Survey Research Center at The University of Michigan, and made available to me through the Inter-University Consortium for Political Research. The exact operationalization of the variables is described in the Technical Note at the end of the article. The number of respondents in the present analysis is 645 out of the 1762 provided in the SRC survey. The 645 are those who had full information on the full set of variables. There is thus a slight overrepresentation of Republicans and upper SES respondents. This bias is slight and there appears to be no other systematic bias in the sub-sample. For more detailed discussion, see Goldberg, Arthur S., The Intergenerational Transmissien of Party Identification (unpublished doctoral dissertation, Yale University, 1966), Appendix BGoogle Scholar.

10 Surveys of the literature on this point are available inHyman, Herbert H., Political Socialization (Glencoe, Illinois: The Free Press. 1959)Google Scholar; and Lane, Robert E. and Sears, David O., Public Opinion (Englewood Cliffs: Prentice-Hall, Inc., 1964)Google Scholar.

11 See, for example, Berelson, Bernard R., Lazarsfeld, Paul F., and McPhee, William N., Voting (Chicago: The University of Chicago Press, 1954), chaps. IV–VIIGoogle Scholar. See also Lipset, Seymour Martin, Political Man (Garden City, New York: Doubleday & Company, Inc., 1960), Part IIGoogle Scholar.

12 See, for example, Campbell, Angus, Gurin, Gerald, and Miller, Warren E., The Voter Decides, (Evanston, Illinois: Row, Peterson and Company, 1954), chap. VIIGoogle Scholar. See also Campbell, Angus, Converse, Philip E., Miller, Warren E., and Stokes, Donald E., The American Voter (New York: John Wiley & Sons, Inc., 1960), chap. VIGoogle Scholar.

13 The American Voter, chap. IV. See also Stokes, Donald E., Campbell, Angus, and Miller, Warren E., “Components of Electoral Decision,” this Review, 52 (06, 1958), 367–387Google Scholar.

14 Cf. The American Voter, pp. 24–37.

15 Cf. Berelson, Lazarsfeld, and McPhee, loc. cit.

16 Cf. Lane and Sears, op. cit., pp. 18–19.

17 Although the aid of a computer was enlisted for the statistical computations in the present article, the lower order partials could be calculated with a slide rule or desk calculator. The appropriate formulae are available in Blalock, Hubert M., Social Statistics (New York: McGraw-Hill Book Company, Inc. 1960), pp. 333–336Google Scholar. For those who wish to check the computation of some of the lower order partials, the matrix of simple correlations is provided below:

18 Assuming that reciprocal causation is not involved, there are always n!/2(n−2)! possible arrows among n variables.

19 The American Voter, p. 35. The relative size of this impact is a function of factors which will be dealt with later in the present analysis. See pp. 918–919, below.

20 Blalock, , Causal Inference …, p. 59Google Scholar.

21 Lane and Sears, loc. cit. Greenstein, Fred I., “The Benevolent Leader: Children's Images of Political Authority,” this Review, 54 (1960), 934–943Google Scholar.

22 The decision to accept or reject linkages on the basis of significance tests poses some problems. The significance test is designed to deal with the type I error (rejection of the null hypothesis when it is true). That is to say, it provides a statement of the probability that a strength of association as great as that found in the sample could have been drawn from a population in which there was in fact no relationship between the variables in question. However, in deciding to omit a link, one is concerned with the risk of a type II error (acceptance of the null hypothesis when it is false). The information desired in this case is the probability that a strength of association as weak asthat in the sample could have been drawn from a population in which there was a stronger association. Unless one has an a priori expectation of the strength of association in the population, it is not possible to calculate this probability.

23 The American Voter, pp. 32–37.

24 Cf. Berelson, Lazarsfeld, and McPhee, op. cit., pp. 227–233. See also The American Voter, pp. 128–131; and Lane and Sears, op. cit., pp. 81–82.

25 Note that this “net impact” refers to a population parameter, or statistic thereof, rather than to the impact of party identification on the vote of any single individual in the population. While the latter would doubtless be of interest, it would be extremely difficult to come by innon-experimental data.

26 See The American Voter, pp. 139–142.

27 Key, V. O. Jr., “A Theory of Critical Elections,” Journal of Politics, 18 (1955), 3–18CrossRefGoogle Scholar.

28 Blalock, , Causal Inferences …, pp. 53–57Google Scholar.

29 Ibid., pp. 46–47. See also Wold, Herman and Jureen, Lars, Demand Analysis (New York: John Wiley & Sons, 1953), pp. 37–38Google Scholar, and Fritz C. Holte, loc. cit.

Submit a response

Comments

No Comments have been published for this article.