¹ These data are drawn from a research project on the internal politics of a PSI federation. While the larger study also includes interviews with all the formal leaders of the party in the province, plus extensive documentary research in party archives, the data analyzed here derive from 301 interviews with a dense sample of every third member of the party in the chief commune of the province. (The number of valid responses varies somewhat from table to table.) The interviews were administered in the spring of 1963 by students of the University of Florence. The response rate was 81% after at least two callbacks.

The larger study, which treats organizational democracy and oligarchy as problems in communications, is Party Democracy: The Internal Politics of an Italian Socialist Federation (forthcoming).

² See Freeman, Howard E., Novak, Edwin, and Reeder, Leo G., “Correlates of Membership in Voluntary Associations,” American Sociologica Review, 22 (10, 1957), 228–233Google Scholar; Herbert Maccoby “The Differential Political Activity of Participants in a Voluntary Association,” ibid, 23 (October 1958), 524–532; Charles R. Wright and Herbert H. Hyman, “Voluntary Association Memberships of American Adults: Evidence from National Sample Surveys,” ibid, 23 (June 1958), 284–294.

³ The Civic Culture (Princeton: Princeton University Press, 1963), p. 320Google Scholar. Their measure of competence differs from that used in the present study.

⁴ Political Man, (Garden City, New York: Doubleday, 1960), p. 202Google Scholar.

⁵ For example, Almond and Verba, op. cit., p. 304; Campbell, Angus, Converse, Philip E., Miller, Warren E., and Stokes, Donald E., The American Voter, (New York: Wiley, 1960), pp. 475–481Google Scholar; the evidence cited in Lipset, op. cit., pp. 187–189; and Milbrath, Lester W., Political Participation (Chicago: Rand McNally and Co., 1965), pp. 122–124Google Scholar.

⁶ Organizations (New York: Wiley, 1958), pp. 72–73Google Scholar.

⁷ The Appeals of Communism (Princeton: Princeton University Press, 1955), esp. pp. 62–126Google Scholar.

⁸ This phenomenon was noted by Almond and Verba, though its implications for working class politics were not developed. Op. cit., pp. 300–322.

⁹ The eight statements were: a) Calabria has a special regional status, b) Twenty-five is the minimum age for being elected a deputy, c) There are seven countries in the European Common Market, d) The PSI abstained on the vote of confidence for the center-left government (Fanfani, 1962). e) AGIP is a private industry, f) In the past the PCI has taken part in the government, g) The Constitutional Court has yet to be established, h) Everyone who has reached the age of 21 can vote for the Chamber and the Senate.

¹⁰ The items were machine tested for scalability using the Multiple Scalogram Analysis developed by Lingoes, James C., “Multiple Scalogram Analysis: A Set-Theoretic Model for Analyzing Dichotomous Items,” Educational and Psychological Measurement, 23 (Autumn 1963), 501–524CrossRefGoogle Scholar. They were also tested for scalability by conventional counter-sorter procedures. As the items did not form an acceptable and useful Guttman-type scale, the number of correct answers given by each respondent provided his position on an index of political knowledge.

¹¹ The statements were: a) The average member has no influence at all on what the party decides, b) Sometimes politics seems so complicated that it is difficult to understand it. c) I don't think that public officials care very much about what people like me think, d) Voting is the only way for the rank and file member of the party to influence the policies of the party. The questions were taken from SRC American electoral studies. Although none of the items were reversed, there is no internal evidence of distortion due to response set.

¹² This measure of ideological sensitivity is adapted from Campbell, et al, op. cit., Chapter 10.

¹³ All measures of association herein are gamma rank order coefficients. Gamma was chosen because it does not require N×N tables in order to reach the values of ±1.0. For a discussion of gamma, see Hays, William L., Statistics for Psychologists (New York: Holt, Rinehart, Winston, 1963), pp. 655–56Google Scholar.