1 Norris, Pippa, “Through the Eye of a Needle: Comparative Legislative Recruitment in Western Democracies,” paper presented at American Political Science Association meetings, Washington, 1993; and Wilma Rule, “Why Women Don't Run: The Critical Contextual Factors in Women's Legislative Representation,” Western Political Quarterly 34 (1981), 60–77Google Scholar. There are currently six single-member district electoral systems among the central legislatures of industrialized democracies: Australia, Canada, France, New Zealand, the United Kingdom and the United States. In addition, the Japanese single nontransferable vote system effectively functions as a single-member district system, even though it is formally a multimember district system.

2 Duverger, Maurice, The Political Role of Women (Paris: UNESCO, 1955)Google Scholar; Lakeman, Enid, How Democracies Vote (London: Faber and Faber, 1970)Google Scholar; Castles, Francis, “Female Legislative Representation and the Electoral System,” Politics 1 (1981), 21–26CrossRefGoogle Scholar; Rule, , “Why Women Don't Run”Google Scholar; Rule, Wilma, “Electoral Systems, Contextual Factors and Women's Opportunity for Election to Parliament in Twenty-Three Democracies,” Western Political Quarterly 40 (1987), 477–98CrossRefGoogle Scholar; Darcy, R., Welch, Susan and Clark, Janet, Women, Elections and Representation (New York: Longman, 1987)Google Scholar; Norris, Pippa, Politics and Sexual Equality (Boulder: Lynne Rienner, 1987)Google Scholar; and Norris, Pippa, “Electoral Systems and the Parliamentary Recruitment of Women,” in Mughan, Anthony and Patterson, Samuel C., eds., Political Leadership in Democratic Societies (Chicago: Nelson-Hall, 1991), 136–44.Google Scholar

3 For a general discussion of the obstacles women face see Randall, Vicky, Women and Politics: An International Perspective (2nd ed.; Chicago: University of Chicago Press, 1987).CrossRefGoogle Scholar

4 Brodie, Janine, Women and Politics in Canada (Toronto: McGraw-Hill Ryerson, 1985)Google Scholar; Brodie, Janine with Chandler, Celia, “Women and the Electoral Process in Canada,” in Megyery, Kathy, ed., Women in Canadian Politics: Toward Equity in Representation (Toronto: Dundurn Press, 1991), 3–59Google Scholar; Bashevkin, Sylvia, Toeing the Lines: Women and Party Politics in Canada (Toronto: University of Toronto Press, 1985)Google Scholar; Erickson, Lynda, “Women and Candidacies for the House of Commons,” in Megyery, ed., Women in Canadian Politics, 101–25Google Scholar; and Young, Lisa, “Legislative Turnover and the Election of Women to the Canadian House of Commons,” in Megyery, ed., Women in Canadian Politics, 81–99.Google Scholar

5 Norris, , Politics and Sexual Equality, 126–30.Google Scholar

6 Brodie, , Women and Politics in Canada, 124Google Scholar; Young, , “Legislative Turnover and the Election of Women,” 83–84Google Scholar; and Bashevkin, Sylvia, “Women's Participation in Political Parties,” in Megyery, ed., Women in Canadian Politics, 61–79.Google Scholar

7 Young, , “Legislative Turnover and the Election of Women,” 84.Google Scholar

8 Brodie with Chandler, “Women and the Electoral Process,” 7.Google Scholar

9 Burt, Sandra, “Women's Issues and the Women's Movement in Canada since 1970,” in Cairns, Alan and Williams, Cynthia, eds., The Politics of Gender, Ethnicity and Language in Canada (Toronto: University of Toronto Press for Supply and Services Canada, 1986), 111–69Google Scholar; and Bashevkin, Sylvia, “Building a Political Voice: Women's Participation and Policy Influence in Canada,” in Nelson, Barbara J. and Chowdhury, Najma, eds., Women and Politics Worldwide (New Haven: Yale University Press, 1993).Google Scholar

10 Gotell, Lisa and Brodie, Janine, “Women and Parties: More than an Issue of Numbers,” in Thorburn, Hugh G., ed., Party Politics in Canada (6th ed.; Scarborough: Prentice Hall, 1991), 53–67.Google Scholar

11 When the NDP has won power recently on the provincial level, these victories have been accompanied by significant increases in women's representation. In 1990 the unexpected NDP majority (74 of 130 seats) obtained in the provincial election in Ontario brought 20 women to the legislature, and the Ontario New Democrats established a provincial cabinet which was more than one third female. For discussion, see Bashevkin, , “Women's Participation in Political Parties”Google Scholar; and Bashevkin, , “Building a Political Voice.”Google Scholar Similar advances for women in both legislative representation and cabinet positions occurred with the victory of the New Democrats in provincial elections in British Columbia and Saskatchewan in 1991. On the federal level the NDP had the first female party leader, Audrey McLaughlin, and in 1991 adopted a position that one half of all their candidates for the next federal election should be women, visible minorities and handicapped people.

12 In short, there could be a contagion from the left in nominating female candidates without such parties having to lead in electing female candidates. See Studlar, Donley T. and Welch, Susan, “The Party System and the Representation of Women in English Metropolitan Boroughs,” Electoral Studies 11 (1992), 63–69.CrossRefGoogle Scholar

13 LeDuc, Lawrence, “Canada: The Politics of Stable Dealignment,” in Dalton, Russell J., Flanagan, Scott C. and Beck, Paul Allen, eds., Electoral Change in Advanced Industrial Democracies: Realignment or Dealignment? (Princeton: Princeton University Press, 1984), 402–24Google Scholar; and Clarke, Harold D., Jenson, Jane, LeDuc, Lawrence and Pammett, John H., Absent Mandate: Interpreting Change in Canadian Elections (2nd ed.; Toronto: Gage, 1991).Google Scholar

14 After 1988, the Bloc Québécois formed as a federal splinter party from the Conservatives, and the Reform party surged in popular opinion eastward from its Alberta base.

15 Courtney, John, “Reinventing the Brokerage Wheel: The Tory Success in 1984,” in Penniman, Howard R., ed., Canada at the Polls, 1984 (Washington: American Enterprise Institute, 1988), 190–208Google Scholar; and Blake, Donald E., “Party Competition and Electoral Volatility: Canada in Comparative Perspective,” in Bakvis, Herman, ed., Representation, Integration and Political Parties in Canada (Toronto: Dundurn Press, 1991), 253–73.Google Scholar

16 Brodie, , Women and Politics in Canada, 19–21.Google Scholar

17 Gotell, and Brodie, , “Women and Parties,” 65.Google Scholar

18 Williams, Robert J., “Candidate Selection,” in Penniman, Howard R., ed., Canada at the Polls, 1979 and 1980 (Washington: American Enterprise Institute, 1981), 86–120Google Scholar; Erickson, Lynda and Carty, R. K., “Parties and Candidate Selection in the 1988 Canadian General Election,” this Journal 24 (1991), 331–49Google Scholar; and Erickson, , “Women and Candidacies for the House of Commons,” 103–04Google Scholar. In 1993, the Liberal party leader, Jean Chrétien, acting within new powers granted him by the party, appointed 14 Liberal parliamentary candidates, 10 of whom were women. There was considerable intraparty controversy about this procedure.

19 Welch, Susan, “Women as Political Animals? A Test of Some Explanations for Male-Female Political Participation Differences,” American Journal of Political Science 21 (1977), 711–30CrossRefGoogle Scholar; Welch, Susan, “Recruitment of Women to Public Office: A Discriminant Function Analysis,” Western Political Quarterly 31 (1978), 372–80CrossRefGoogle Scholar; Brodie, , Women in Politics in CanadaGoogle Scholar; and Randall, , Women and Politics, 96–107.Google Scholar

20 Brodie, , Women and Politics in Canada, 2Google Scholar; and Bashevkin, , “Building a Political Voice.”Google Scholar

21 Wilson-Smith, Anthony, “The Tory Revival in Quebec,” MacIean's, April 9, 1984, 14–16.Google Scholar

22 For example, in Ontario the three-party vote in 1980 was Conservative 35 per cent, Liberal 42 per cent and NDP 22 per cent, while the early 1984 Gallup totals were Conservative 54 per cent, Liberal 33 per cent and NDP 13 per cent. Therefore in each Ontario riding the 1980 Conservative vote total was increased by 54 per cent (54/35), the Liberal vote total was decreased by 21 per cent (33/42) and the NDP vote total decreased by 41 per cent (13/22).

23 We examined the sensitivity of our results by testing for a change in gender make-up if the 32 races where the margin of victory was less than 10 per cent had been won by the party which finished second. In four cases women would have lost seats, but in four other cases the new winning MP would have been a woman.

24 Morley, J. Terence, “Annihilation Avoided: The New Democratic Party in the 1984 Federal General Election,” in Penniman, Howard R., ed., Canada at the Polls, 1984 (Washington: American Enterprise Institute, 1988), 120–36.Google Scholar

25 Statistics Canada, Federal Election Districts—1981 Selected Social and Economic Characteristics (Ottawa: Statistics Canada, 1983)Google Scholar; and Statistics Canada, Federal Election Districts—1987 Representation Order: Part 2, Profiles (Ottawa: Statistics Canada, 1988).Google Scholar

26 Hanushek, Eric A. and Jackson, John A., Statistical Methods for Social Scientists (New York: Academic Press, 1977), 179–216.CrossRefGoogle Scholar

27 Andersen, Kristi and Thorson, Stuart A., “Congressional Turnover and the Election of Women,” Western Political Quarterly 37 (1984), 143–56CrossRefGoogle Scholar; Darcy, R. and Choike, James R., “A Formal Analysis of Legislative Turnover: Women Candidates and Legislative Representation,” American Journal of Political Science 30 (1986), 237–55CrossRefGoogle Scholar; and Darcy, , Welch, and Clark, , Women, Elections and Representation, 138–42.Google Scholar

28 Bashevkin, , “Building a Political Voice”Google Scholar; Young, , “Legislative Turnover and the Election of Women,” 86–96Google Scholar; and Franks, C. E. S., The Parliament of Canada (Toronto: University of Toronto Press, 1987), 73–74.CrossRefGoogle Scholar

29 Bashevkin, , Toeing the Lines, 72–74Google Scholar; Brodie, , Women and Politics in Canada, 106–19Google Scholar; see, however, Erickson, , “Women and Candidacies for the House of Commons”Google Scholar; and Pelletier, Réjean and Tremblay, Manon, “Les femmes sont-elles candidates dans des conscriptions perdues d'avance? De l'examen d'une croyance,” this Journal 25 (1992), 249–67.Google Scholar

30 There was a reapportionment of the federal electoral districts between the 1984 and 1988 elections. The number of constituencies was expanded from 282 to 295 and numerous borders were redrawn. Since the party competitiveness measure depends upon data from the previous election, this causes some problems for our 1988 probits. Elections Canada, however, has taken the 1984 electoral results and transposed them onto the newly reapportioned districts, providing estimates for party vote for all of the 295 districts based on the 1984 election results. We use these hypothetical results for the 1984 elections to estimate party competitiveness for 1988.

31 Moncrief, Gary F. and Thompson, Joel A., “Urban and Rural Ridings and Women in Provincial Politics: A Research Note on MLAs,” this Journal 24 (1991), 831–37.Google Scholar

32 Volgy, Thomas J., Schwarz, John E. and Gottlieb, Hildy, “Female Representation and the Quest for Resources: Feminist Activism and Electoral Success,” Social Science Quarterly 71 (1990), 156–68.Google Scholar

33 The census was taken in 1981 and 1986. The data from the 1981 census is used for both the 1980 and 1984 probit results. The 1986 census is used for the 1988 probit results. This does introduce some slippage as the data are not collected in the election year. Since these variables are primarily demographic characteristics of each riding, however, we are reasonably confident that there is unlikely to be major change in the two- or three-year period from data collection to the election in which they are used.

34 See Guppy, Neil, Freeman, Sabrina and Buchan, Shari, “Representing Canadians: Changes in the Economic Backgrounds of Federal Politicians, 1965–1984,” Canadian Review of Sociology and Anthropology 24 (1987), 417–30.CrossRefGoogle Scholar

35 Andersen, Kristi, “Working Women and Political Participation, 1952–1972,” American Journal of Political Science 19 (1975), 439–53CrossRefGoogle Scholar; and Welch, , “Women as Political Animals?”Google Scholar

36 Rule, , “Why Women Don't Run,” 69–77Google Scholar, and Rule, , “Electoral Systems,” 481–83, 490–91Google Scholar. We use slightly different estimates of university education than Rule. Rule used women with university education; we use percentage of total adult population with university education.

37 Norris, , Politics and Sexual Equality, 131.Google Scholar

38 Rule, Wilma, “Why More Women Are State Legislators,” Western Political Quarterly 43 (1990), 437–48.CrossRefGoogle Scholar

39 Rule, , “Why Women Don't Run,” 69–77Google Scholar, and Rule, , “Electoral Systems,” 481–83, 490–91.Google Scholar

40 Engstrom, Richard L., “District Magnitudes and the Election of Women to the Irish Dail,” Electoral Studies 6 (1987), 123–32CrossRefGoogle Scholar; and Studiar, Donley T. and Welch, Susan, “A Giant Step for Womankind? Women Candidates and the 1992 Election,” in Norris, Pippa, Denver, David, Broughton, David and Railings, Colin, eds., Elections, Parties and Public Opinion Yearbook 1993 (London: Harvester Wheatsheaf, 1993).Google Scholar

41 If the analysis is limited to those ridings which maintained the same borders (approximately two thirds), and where the party competitiveness variable can be estimated based on real election results, a re-analysis shows that party competitiveness is not statistically significant. The other variables, in this reduced sample equation, maintain approximately the same strength.

42 For a discussion of strategic voting among US and British voters see Abramson, Paul R., Aldrich, John H., Paolino, Phil and Rohde, David, “‘Sophisticated’ Voting in the 1988 Presidential Primaries,” American Political Science Review 86 (1992), 55–69CrossRefGoogle Scholar; and Niemi, Richard G., Whitten, Guy D. and Franklin, Mark N., “Constituency Characteristics, Individual Characteristics and Tactical Voting in the 1987 British General Election,” British Journal of Political Science 22 (1992), 229–40.CrossRefGoogle Scholar

43 Multicollinearity does not lead to biased estimates of the underlying relationship. High multicollinearity will, however, tend to increase the standard errors of the estimates and also lead to greater instability. The usual problem one needs to be concerned about is whether multicollinearity has caused a variable which is significant to appear statistically insignificant because of an inflated standard error. A less common problem is whether estimate instability might be the reason a variable, such as party competitiveness, appears to be significant. We are concerned here about the correlation between party competitiveness and the remaining variables, especially candidate incumbency. We tested whether party competitiveness is a linear combination of the other independent variables by regressing all the independent variables on party competitiveness. Transforming the R² into simple R, R is .81 in 1980, .73 in 1984 and .79 in 1988. These are high, and the strong correlations can affect the estimates, but they are not so high that it is impossible to get independent estimates of each variable's effect. There is also a strong correlation between women's labour force participation rates (WLFPR) and proportion of adult population with some university education. Regressing all the other independent variables onto WLFPR produces a simple R of .74 in 1980 and 1984 and .75 in 1988. Again these are high, but we feel confident that the independent estimates are reasonable.

44 Erickson, , “Women and Candidacies for the House of Commons,” 112.Google Scholar

45 The census did not report religion in its 1986 count of the Canadian people, therefore we were forced to use percentage of population with French as their mother tongue as a proxy. This variable correlated with percentage Catholic above the .90 level in the 1981 census.

46 We should mention a number of additional technical issues which were dealt with in developing this table. The table is based on the probits presented in Table 3. The percentage Catholic and level of unemployment variables were set equal to their means and included in the constant. These two variables have very weak effects and even changing them substantially from the levels they were set at would not have noticeably affected the probability estimates. The ridings are characterized by low, moderate or high levels of women's labour force participation and adults with university education. It is reasonable to couple these variables as they were strongly correlated in the data set (R = .63 and .60). For the estimates of low levels these two variables were set equal to their value at the 25th percentile mark. The moderate estimates were set equal to the value at the 50th percentile, and the high estimates were set equal to the value at the 75th percentile. For the 1980 and 1984 elections, the values for women's labour force participation were 45.33 per cent (25th percentile), 50.44 per cent (50th percentile) and 56.12 per cent (75th percentile). The values for percentage of adult population with some university education were 10.14 per cent (25th percentile), 12.87 per cent (50th percentile) and 18.16 per cent (75th percentile). For the 1988 election, the values for women's labour force participation were 50.51 per cent (25th percentile), 55.40 per cent (50th percentile) and 60.01 per cent (75th percentile). The values for percentage of adult population with some university education were 11.81 per cent (25th percentile), 14.68 per cent (50th percentile) and 21.78 percent (75th percentile).

47 Franks, , The Parliament of Canada, 74.Google Scholar

48 Williams, , “Candidate Selection,” 99–100Google Scholar; and Erickson, and Carty, , “Parties and Candidate Selection,” 342.Google Scholar

49 Running the other independent variables against party competitiveness produces a simple R of .73.

50 The coefficient is -.0077 with a standard error of .0074. The T-statistic drops from 2.37 to 1.04. While the sample does decrease quite a bit, there were still 135 cases left.

51 Lipset, Seymour Martin, Continental Divide (New York: Routledge, 1990)Google Scholar; and Noms, , “Through the Eye of a Needle.”Google Scholar

52 Hunter, Alfred A. and Denton, Margaret A., “Do Female Candidates ‘Lose Votes’?: The Experience of Female Candidates in the 1979 and 1980 Canadian General Elections,” Canadian Review of Sociology and Anthropology 21 (1984), 395–406.CrossRefGoogle Scholar

53 Erickson, and Carty, , “Parties and Candidate Selection,” 343–45.Google Scholar

54 Norris, Pippa, Carty, R. K., Erickson, Lynda, Lovenduski, Joni and Simms, Marian, “Party Selectorates in Australia, Britain, and Canada: Prolegomena for Research in the 1990s,” Journal of Commonwealth and Comparative-Politics 28 (1990), 219–45.CrossRefGoogle Scholar