¹ Lipset, S. M. and Rokkan, S., ‘Cleavage Structures, Party Systems and Voter Alignments: An Introduction’, in Lipset, S. M. and Rokkan, S., eds., Party Systems and Voter Alignments (New York: Free Press, 1967), pp. 1–64.Google Scholar

² For a similar discussion, see Shively, W. P., ‘Party Identification, Party Choice and Voting Stability: the Weimar Case’, American Political Science Review, LXVI (1972), 1203–25.CrossRefGoogle Scholar

³ Rose, Richard, ‘Comparability in Electoral Studies’ in Rose, Richard, ed., Electoral behavior (New York: Glencoe, 1974), Table 1, p. 17.Google Scholar Budge and Farlie find Australian voters to be rather more predictable than Canadian voters. See Budge, Ian and Farlie, Dennis, Voting and Party Competition (London: Wiley, 1977), Chap. 8, esp. Tables 8·6 and 8·7.Google Scholar

⁴ Many explicit comparisons are made in Epstein, Leon D., ‘A Comparative Study of Australian Parties’, British Journal of Political Science, VII (1977), 1–21.CrossRefGoogle Scholar See also his earlier ‘A Comparative Study of Canadian Parties’, American Political Science Review, LVIII (1964), 46–59Google Scholar and Albinski, H. F., Canadian and Australian Politics in Comparative Perspective (New York: Oxford University Press, 1973), 145–286.Google Scholar

⁵ Alford, Robert, Party and Society (Chicago: Rand McNally, 1963), 172–8Google Scholar; Sundquist, James L., Dynamics of the Party System (Washington, D.C.: Brookings Institute, 1973), pp. 26–36, 275–98Google Scholar; Aitkin, Don, Stability and Change in Australian Politics (Canberra: Australian National University Press, 1977), pp. 162–6.Google Scholar

⁶ Campbell, A. et al. , The American Voter, abridged edn. (New York: Wiley, 1964), pp. 89–96Google Scholar; Greenstein, Fred I., Children and Politics, rev. edn. (New Haven, Conn.: Yale University Press, 1969), pp. 55–84.Google Scholar

⁷ Beck, Paul Allen, ‘A Socialization Theory of Partisan Realignment’, in Niemi, Richard G. and Weisberg, Herbert F., eds., Controversies in American Voting behavior (San Francisco: W. H. Freeman, 1976), pp. 396–411.Google Scholar

⁸ Miller, Warren E. and Stouthard, Philip, ‘Confessional Attachment and Electoral Behavior in the Netherlands’, European Journal of Political Research, III (1975), 219–58.CrossRefGoogle Scholar

⁹ Goldberg, A. S., ‘Social Determinism and Rationality as Bases for Party Identification’, American Political Science Review, LXIII (1969), 5–25.CrossRefGoogle Scholar

¹⁰ For a similar sort of model, see Campbell, et al. , The American Voter, pp. 161–82.Google Scholar

¹¹ Bishop, Y. M. M., ‘Full Contingency Tables, Logits and Split Contingency Tables’, Biometrics, XXV (1969), 383–400CrossRefGoogle Scholar; Bishop, Y. M. M. and Mosteller, F., ‘Smoothed Contingency Table Analysis’ in Bunker, J. P. et al. , eds., The National Halothene Study (Washington, D.C.: US Government Printing Office, 1969), Chap. IV-3Google Scholar; Fienberg, S. E., The Analysis of Cross-Classified Categorical Data (Cambridge, Mass.: MIT Press, 1977)Google Scholar; Goodman, L. A., ‘The Analysis of Multidimensional Contingency Tables…’, Technometrics, XXXI (1977), 33–62Google Scholar; Bishop, Y. M. M., Fienberg, S. E. and Holland, P. W., Discrete Multivariate Analysis: Theory and Practice (Cambridge, Mass.: MIT Press, 1975)Google Scholar; Reynolds, H. T., The Analysis of Cross-Classifications (New York: Free Press, 1977).Google Scholar For a relatively less technical explanation of the method see Gold, H., Correlates of Support for Separatism Amongst French Canadians in Quebec (unpublished MA thesis, Queen's University, Kingston, 1976), Chap. 2Google Scholar, and Reynolds, H. T., Analysis of Nominal Data, Sage University Papers Series on Quantitative Analysis in the Social Sciences (Beverly Hills, Calif.: Sage, 1977).Google Scholar

¹² A more comprehensive justification for the choice of Model 6 than given in Table 6 is available on request from the authors. Model 6 can be shown to be superior by our criteria, not only to the seven alternatives considered in Table 6, but also to an exhaustive set of twenty other neighbouring possibilities.

¹³ See Mol, Hans, Religion in Australia: A Sociological Investigation (Melbourne; Nelson, 1971), pp. 289–300Google Scholar; Aitkin, Don, Stability and Change in Australian Politics, pp. 170–5.Google Scholar

¹⁴ Similar findings have been reported for surveys done in 1965 and 1968. See Irvine, William P., ‘Explaining the Religious Basis of Canadian Partisan Identity’, Canadian Journal of Political Science, VII (1974), 560–3CrossRefGoogle Scholar; and Irvine, William P., ‘The Decline of the Religious and Cultural Bases for Political Partisanship in Canada’ (unpublished paper presented at the ECPR Workshop on Language and Religion in Politics, London, 04 1975).Google Scholar

¹⁵ For more evidence and discussion of a decline over time of the class cleavage in Australian politics, see David Kemp, ‘Social Change and the Future of Political Parties: the Australian Case’ in Maisel, Louis and Sacks, Paul, The Future of Political Parties, Sage Electoral Studies Yearbook, Vol. 1 (Beverly Hills, Calif.: Sage Publications, 1975), pp. 124–64.Google Scholar

¹⁶ Among those whose occupations are currently working-class, 56 per cent of those from Labor families and 48 per cent of those from non-Labor families inherited their partisanship. Of those who are now middle-class, 27 per cent stuck to a Labor partisanship, while 84 per cent of those whose parents were non-Labor inherited a similar non-ALP stance.

¹⁷ Opinion polls since August 1978 suggest a revival in the ALP's popularity. See The Bulletin, 13 02 1979, p. 34.Google Scholar

¹⁸ Jennings, M. Kent and Niemi, Richard, ‘The Transmission of Political Values from Parent to Child’, American Political Science Review, LXII (1968), 169–84.CrossRefGoogle Scholar

¹⁹ A most stimulating characterization of the Michigan paradigm is offered by W. D. Burnham in ‘Contributions of the SRC to the Development of Voting Theory’ (unpublished paper presented to the 1974 annual meeting of the American Political Science Association).

²⁰ For a review of the literature and some evidence, see Nie, Norman H., Verba, Sidney and Petrocik, John R., The Changing American Voter (Cambridge, Mass.: Harvard University Press, 1976), pp. 156–73. See also pp. 187–93Google Scholar for evidence that there may have been an earlier period of high issue voting in the United States.

²¹ Burnham, Walter Dean, ‘Political Immunization and Political Confessionalism: The United States and Weimar Germany’, Journal of Interdisciplinary History, III (1972–1973), 1–30, p. 15.CrossRefGoogle Scholar

²² Rose, Richard and Urwin, D. W., ‘Persistence and Change in Western Party Systems since 1945’, Political Studies, XVIII (1970), 287–319.CrossRefGoogle Scholar

²³ Reports of ‘very strong’ partisanship are offered by 23 per cent of respondents in 1965 and 24 per cent in 1968. Unlike in 1974, neither of these electorates contained 18–20 year olds. This expansion in 1974 ought to have diminished aggregate strength of partisanship. See Irvine, William P., ‘Explaining the Brittleness of Partisanship in Canada’ (unpublished paper presented to the 47th annual meeting of the Canadian Political Science Association, Edmonton, Alberta, 1975), Table 1.Google Scholar

²⁴ We continue to exclude, as we did earlier in the paper, those who have no partisan identity and can recall none for either parent.

²⁵ This is similar in conceptual status to our previous variable which we labelled ‘class integration’. That variable, it will be recalled, was coded ‘high’, ‘medium’ and ‘low’. In this case, we code the environment as ‘working class’ if the respondent identifies with and draws the majority of his friends from that class; as ‘middle class’ if this is the course of his friends and identification, and ‘mixed’ in all other cases. In this section of the paper, we have sought to preserve the directionality of all the variables.

²⁶ To obtain the estimates in Panel 2 of Table 10 (but not for Panel 1), a small constant (0·5) was added to each cell of the cross-tabulation to deal with seven zero marginals encountered there. Goodness-of-fit tests, however, were carried out on the actual cell observations but the degrees of freedom were adjusted to take account of zero marginals.

²⁷ In an early test, we found a weak effect whereby Roman Catholic non-Labor identifiers displayed less voting fidelity than other religious groups. On inspection, this effect was found to depend entirely on sixteen respondents who had switched from the Australian Labor Party to the Democratic Labor Party when the split between the two occurred. It is very difficult to consider these as floating voters, given that they simply followed a party schism, so we excluded them from the analysis of Table 10 (2). On doing so, the religious effect disappeared. Even in the earlier version, there was no evidence of any effect from frequency of religious practice, either directly or in interaction with other variables.

²⁸ As for Table 10, Panel 2, a small constant (05) was added to each cell of the cross-tabulation to obtain the estimates in Table 11. Three zero marginals had been obtained in the cross-tabulation. Goodness-of-fit tests were carried out on actual observations with adjusted degrees of freedom.

²⁹ We did estimate the parameters for a model containing an interactive effect between religion and party. Summing the effect on voting fidelity of being a Roman Catholic, plus the effect of being a Roman Catholic Liberal, we obtain the following effects: R.C./Lib. = ·068; R.C./non-Lib. = –·040; non-R.C./Lib. = –·068; and non-R.C./non-Lib. = ·040. Although the signs are all consistent with a model of social reinforcement or cross-pressures, the effects are not significantly different from zero.

³⁰ Burnham, , ‘Politial Immunization and Political Confessionalism’, p. 16.Google Scholar

³¹ Epstein, , ‘A Comparative Study of Australian Parties’, p. 9.Google Scholar

³² Unless, of course, the splitters aim to defeat their former partners. By directing second preferences to the Liberal or Country Parties, the DLP was able to weaken the ALP's presence in the House of Representatives below what would have been obtained in a first-past-the-post system. By winning Senate seats on its own account, the DLP helped to deny control of the Senate to Whitlam's Labor government.

³³ For a discussion of Australian third parties up to 1975 and a similar argument about the effect of the electoral system, see Reynolds, Paul, ‘The Role of the Minor Parties’, in Penniman, Howard, ed., Australia at the Polls (Washington, D.C.: American Enterprise Institute, 1977), pp. 159–69.Google Scholar