¹ Nonpartisan elections refer specifically to councilmanic offices only, but in cities with nonpartisan elections for the city council the office of the mayor and other city officers are usually nonpartisan as well. See Lee, Eugene C., The Politics of Nonpartisanship: A Study of California City Elections (Berkeley and Los Angeles: University of California Press, 1960), p. 25Google Scholar. A historical account of the spread of nonpartisanship is given in this work. See also: Stone, Harold A. et al. , City Manager Government in the United States: A Review after Twenty-Five Years, Chicago: Public Administration Service, 1940. The Municipal Year Book, 1958 (Chicago: International City Managers' Association, 1958), is the data source used for electoral rules in the cities studied in this paper.Google Scholar

² Research on the consequences of the nonpartisan system of election rules may be found in: Adrian, Charles R., “Some General Characteristics of Nonpartisan Elections”, American Political Science Review, XLVI (09, 1952), pp. 766–776CrossRefGoogle Scholar. A later and more comprehensive work on the effects of nonpartisan elections may be found in Charlotte Frank, “Politics in the Non-partisan City”, unpublished Masters Thesis, University of Chicago, 1957/58. For additional data on the consequences of nonpartisanship see Lee, op. cit., Williams, Oliver P. and Adrian, Charles R., “The Insulation of Local Politics under the Nonpartisan Ballot”, American Political Science Review, LIII (12, 1959), pp. 1052–63CrossRefGoogle Scholar, and Freeman, J. Leiper, “Local Party Systems: Theoretical Considerations and a Case Analysis”, American Journal of Sociology, LXIV, (11, 1958), 282–289.CrossRefGoogle Scholar

³ See Schattschneider, E. E., Party Government, New York: Farr and Rinehart, 1942Google Scholar; Gosnell, H. F., Grass Roots Politics, Washington, D.C.: American Council on Public Affairs, 1942Google Scholar, Key, V. O., Jr., Southern Politics, New York: Alfred Knopf, 1949Google Scholar, and Key, V. O., Jr. American State Politics: An Introduction, New York: Alfred Knopf, 1956.Google Scholar

⁴ We used the average of the vote for U.S. Representatives in the state as a whole for all Congressional elections from 1944 through 1950 as our indicator of the condition of the parties in each state. The data were taken from Gallup, George, The Political Almanac, New York: B. C. Forbes & Sons, 1952.Google Scholar These statistics gives us the classification of the 48 states listed in Table 1. Reliable data are not available for the electoral system changes before 1940 and in view of the continuing change in the direction of nonpartisan balloting it seemed reasonable to use the post-war vote as an indicator of the long term competitive status of a state. Careful examination of the states in each category should persuade the reader that with the exception of Rhode Island the 14 democratic states have been one-party states since the Civil War. The 12 Republican states have long term histories as Republican strongholds. V. O. Key has a useful list of the average vote for Governor from around 1910 to around 1952 for 21 non-Southern states and all of the states he includes as having less than 45 per cent Democratic voters during that period are included in my Republican category. His listing would exclude only New Jersey placing it in the competitive group. See Key, V. O.Jr., American State Politics, op. cit. p. 99. Any criteria will necessarily define the results of a classification; in the present case I assumed that a post-depression, post-war vote covering four elections at two-year intervals would reflect the long term predispositions of the state. We will see in a later section that some “Republican” states have developed two-party politics since 1950. This later development does not mean that the original classification is false, only that state political systems are subject to changes (given the appropriate types of social structures, as we will point out).Google Scholar

⁵ Data on voting in these elections were taken from Edgar Robinson, E., The Presidential Vote, 1896–1932, Stanford: Stanford University Press, 1934. It is interesting to note that present day “ticket-splitting” by voters is often found in the same states we have called “low-loyalty” states. Ticket-splitting may also be an indicator of low party loyalty. A separate indicator of differences among southern states in party loyalty was not available to this writer. Further analysis of party loyalty differences among southern states and the effect of these differences would be desirable.Google Scholar

⁶ Lipset, S. M., Political Man, New York: Doubleday, 1960Google Scholar, chapter 1. and Kornhauser, William, The Politics of Mass Society, Glencoe, Illinois, The Free Press, 1959.Google Scholar

⁷ Coleman, James, Community Conflict, Glencoe, Illinois: The Free Press, 1957.Google Scholar

⁸ Peter H. Rossi, “Theory and Method in the Study of Power in the Local Community”, a paper presented at the 1960 Annual Meeting of the American Sociological Association, (New York, 1960). Case studies of conflict over the form of government that a community adopts also support the idea that movements to return to a partisan political system stem from working class organizations. On this point see Stone, Edwin O. and Floro, George K., Abandonments of the Manager Plan, Lawrence, Kansas: University of Kansas Publications, 1953.Google Scholar

⁹ Rossi, ibid., p. 30.

¹⁰ This corresponds to the economic classification of cities as “industrial” and “manufacturing” as opposed to all other types of cities. These data were taken from The Municipal Year Book, 1953, op. cit. Data on union organization and political action in manufacturing as opposed to nonmanufacturing cities would possibly clarify this relationship.

¹¹ Religious data are taken from the U.S. Department of Commerce, Religious Bodies, 1936, Vol. 1, Washington: U.S. Government Printing Office, 1941. The number of Roman Catholics and Greek Orthodox Catholics as the percent of the total city population is used if the city had 25,000 inhabitants at the time of the 1930 census. If the city's population was less, we necessarily used similar data from the county in which the city was located as the best available estimate of the city's Catholic population.

¹² A failure to control variables properly may therefore account for the lack of comparability between investigations of attributes associated with two-party competition in different states. For example, an index of urbanization is associated with party competition in Ohio but not in Iowa. This suggests that religion might be introduced as a third variable. See Gold, David and Schmidhauser, John, “Urbanization and Party Competition: The Case of Iowa”, Midwest Journal of Political Science, IV, (02, 1960), pp. 67–72,Google Scholar and Eulau, Heinz, “The Ecological Basis of Party Systems: The Case of Ohio”, Midwest Journal of Political Science, I, (08, 1957), pp. 125–135.CrossRefGoogle Scholar