Hostname: page-component-8448b6f56d-tj2md Total loading time: 0 Render date: 2024-04-24T21:45:54.632Z Has data issue: false hasContentIssue false
  • English
  • Français

Class and Party in Partisan and Non-Partisan Elections: The Case of Des Moines

Published online by Cambridge University Press:  01 August 2014

Robert H. Salisbury  and
Gordon Black
Robert H. Salisbury
Affiliation:
Washington University
Gordon Black
Affiliation:
Washington University
Get access Rights & Permissions [Opens in a new window]

Extract

In a recent monograph, Professor Heinz Eulau begins his analysis by quoting two “evidently antagonistic formulations” of the theoretical underpinnings of voting behavior in the United States:

1. “A person thinks, politically, as he is socially.”

2. Crucial among the elements in the electoral decision are “traditional or habitual partisan attachments.”

These rival conceptions of primacy among politically relevant variables are often summarized by the terms “class” and “party.” As Eulau points out, “from Aristotle to Harold J. Laski, the relationship between class and party has been one of the ‘grand problems,’ so-called, of speculation about political systems. It has also remained one of the most neglected areas of systematic theory and of empirical analysis.” Data drawn from Survey Research Center surveys have recently been used to explore the relative importance and specify the interdependence of class and party in American voting. Generally, they show party to be more immediately relevant to the voting decision than class, though class position clearly shapes and sets limits to possible party identification and party-related perspectives.

Difficult problems are involved in attempting to sort out and define the two postulated independent variables. The extent to which, in some sense, class determines party orientation is perhaps the most difficult. For example, even when it is found that a certain portion of the working class prefers the Republican party, it may still be that a generation or two earlier the families of this group were Republican on class grounds, and have perpetuated the identification through the socialization process. Campbell et al., conclude that party identification has a “conserving influence,” inhibiting or, at least, slowing down the political manifestation of changes in class position. Their dat a strongly suggest that in any immediate situation class will be much less highly correlated with the vote than party preference. Campbell et al., do not attempt to control for class in relating party identification to the vote, although they do explore the separate effect of class. Eulau deals with this problem at length, but his focus is rather different. He does not attempt to specify the relative weight of each independent variable in predicting the vote, but concentrates on exploring the interrelationships of the two variables.

Type
Research Article
Information
American Political Science Review , Volume 57 , Issue 3 , September 1963 , pp. 584 - 592
Copyright
Copyright © American Political Science Association 1963

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 Class and Party in the Eisenhower Years (New York, 1962), p. 1Google Scholar. The first statement is drawn from Lazarsfeld, Paul F., Berelson, Bernard, and Gaudet, Hazel, The People's Choice (New York: Columbia University Press, 1948), p. 27Google Scholar; the second from Key, V. O. Jr. and Munger, Frank, “Social Determinism and Electoral Decision: The Case of Indiana,” in Burdick, Eugene and Brodbeck, Arthur J., eds., American Voting Behavior (New York, 1959), pp. 281–2Google Scholar.

2 Op. cit., p. 4.

3 See Campbell, Angus, Converse, Phillip E., Miller, Warren E., and Stokes, Donald E., The American Voter (New York, 1960), esp. ch. 13Google Scholar.

4 Ibid., pp. 364–8.

5 In Des Moines it is possible to register without declaring party affiliation, and about 13 percent of the electorate does so. There is very little variation, however, in the proportion of “no-party” registrants by precincts, and so we have excluded them in our calculations of party registration.

6 All correlations are Pearson product-moment correlations using percentages of voters registered Republican and neighborhood housing rating for each of 68 precincts as independent variables. The data are arranged so that positive corelations indicate positive relationships. Partial correlations greater than.3 are significant at the .01 level. Partial correlations greater than .23 are significant at the .05 level. Computer work was performed under a National Science Foundation grant to Washington University for unsponsored research.

We have chosen to report our data in terms of partial and multiple r's. An alternative method would utilize Beta scores which would indicate both the closeness of fit and the slopes of the regression lines, whereas the partial r's indicate only the former. The substance of our conclusions would not be altered by this method although the numerical values would. Partial r's are somewhat more familiar to social scientists and may provide more ready comparability.

7 Primarily because of its convenience for our purposes, we have used the neighborhood classification prepared by Nelson, Howard J., reported in The Livelihood Structure of Des Moines, Iowa, unpublished Ph.D. diss., University of Chicago, 1949, pp. 110–15Google Scholar and map, figure 2. Nelson, in turn, followed the field classification procedures and categories developed by Jones, W. D., “Field Mapping of Residential Areas in Metropolitan Chicago,” Annals of the Association of American Geographers, Vol. 21 (1931), pp. 207–14CrossRefGoogle Scholar. Nelson utilizes five neighborhood categories and bases his classification of blocks on size of housing units, spacing, state of upkeep, need of major repairs or new facility, age, and median rental. In virtually every case, however, the assignment of a block to one category or another hinged upon the relative condition of housing in that block to other Des Moines housing. No absolute standard is in volved.

8 See, particularly, Kahl, Joseph A. and Davis, James A., “A Comparison of Indexes of Socio-Economic Status,” American Sociological Review, Vol. 20(1955), pp. 317–25CrossRefGoogle Scholar.

9 It is premature, of course, to conclude that a Shevky-Bell index would not provide an adequate measure of class differentiation in Des Moines. It does appear, however, that as the index was designed for use in larger and more heterogeneous metropolitan areas, its ability to discriminate among rather narrowly differentiated class groupings would not be high. See Shevky, Eshref and Bell, Wendell, Social Area Analysis (Stanford: Stanford University Press, 1955)Google Scholar.

10 For an excellent recent discussion of this problem, see Ranney, Austin, “The Utility and Limitations of Aggregate Data in the Study of Electoral Behavior,” in Ranney, , ed., Essays in the Behavioral Study of Politics (Urbana: University of Illinois Press, 1962), pp. 91–102Google Scholar.

11 See Lee, Eugene C., The Politics of Non-Partisanship (Berkeley and Los Angeles: University of California Press, 1960)Google Scholar; Adrian, Charles R., “A Typology for Nonpartisan Elections,” Western Political Quarterly, Vol. 12 (1959), pp. 449–458CrossRefGoogle Scholar; Charles R. Adrian, “Some General Characteris-tics of Nonpartisan Elections,” this Review, Vol. 46 (1952), pp. 766–76; Gilbert, Charles E., “Some Aspects of Nonpartisan Elections in Large Cities,” Midwest Journal of Political Science, Vol. 6 (1962), pp. 345–62CrossRefGoogle Scholar.

12 See, for example, the discussion in Kneier, C. M., City Government in the United States, rev. ed. (New York, 1947), pp. 497–508Google Scholar; Penniman, Howard R., Sait's American Parties and Elections, 4th ed. (New York, 1948), pp. 203–5Google Scholar. But cf. Adrian, “Some General Characteristics …,” loc. cit., pp. 773–4.

13 See, for example, Lane, Robert E., Political Life (Glencoe, Illinois, 1959), pp. 269–71Google Scholar.

14 “A Typology …,” loc. cit. In Adrian's classification, Des Moines elections would appear to be Type III. “Elections where slates of candidates are supported by various interest groups, but political parly organizations have little or no part in campaigns, or are active only sporadically.” p. 455, italics in original. Adrian notes that this is a very common type of non-partisan election, p. 456.

15 For a non-partisan city with highly structured local party organizations closely mirroring national party lines, see Freeman, J. Lieper, “Local Party Systems: Theoretical Considerations and a Case Analysis,” American Journal of Sociology, Vol. 44 (1958), pp. 282–9CrossRefGoogle Scholar.

16 In the 1949 election on the adoption of the city manager system, 49.1 percent of the registered voters went to the polls. In subsequent municipal elections the turnouts were as follows: 34.3 percent in 1951; 20 percent in 1953; 39.3 percent in 1955; 45.7 percent in 1957; 39.3 percent in 1959; and 36.3 percent in 1961.

17 Oliver P. Williams and Charles R. Adrian found that in four Michigan communities with non-partisan elections, “the more assiduously issues in the non-partisan elections were pursued, the more the resulting vote conformed to the partisan pattern.” “The Insulation of Local Politics under the Nonpartisan Ballot,” this Review Vol. 53 (1959), pp. 1052–63 at p. 1058. Our data suggest that not all issues have this effect in Des Moines, but the question of “labor power” certainly does.

18 Ibid., pp. 1059–61.

19 Op. cit., pp. 333–80. For evidence that depolarization may not actually have occurred in any consistent manner over time, see Alford, Robert R., “The Role of Social Class in American Voting Behavior,” Western Political Quarterly, Vol. 16 (1963), pp. 180–95.CrossRefGoogle Scholar

20 Especially, Williams and Adrian, op. cit