Forthcoming Legislative Studies Quarterly, February, 1998

Richard E. Matland
Department of Political Science
University of Houston
Houston, TX 77204-3474

July 22, 1996

The author would like to thank NATO for an Advanced Research Fellowship under The Study of Democratic Institutions Program and the University of Houston for an LGIA grant which helped support this research. The author would also like to thank Christine Pintat of the Inter Parliamentary Union, Philip Michelbach and Deborah Orth for their assistance and comments on earlier versions of this paper.

WOMEN'S IN NATIONAL LEGISLATURES: A COMPARISON OF DEMOCRACIES IN DEVELOPED AND DEVELOPING COUNTRIES  LEGISLATIVE REPRESENTATION

Richard E. Matland

This article expands research on representation of women in national legislatures. Existing models are tested on newer data in advanced industrialized democracies, and these models are then applied to a sample of democracies in developing countries. There are striking differences across the two samples. While a proportional representation electoral system, women's labor force participation rates, a cultural standing of women variable, and level of development all have positive effects on female representation in OECD democracies, none of these variables have a statistically significant and positive effect in less developed countries. These findings strongly suggest the existence of a threshold. Only after that threshold is passed do proportional representation, labor force participation, and cultural standing exert positive influences on the representation of women.

Women are poorly represented in national legislatures throughout the world. The Inter- Parliamentary Union (1995) reports that as of June 30, 1995, only 11.3% of the representatives in the 176 existing national parliaments were women. Yet despite the general picture of significant underrepresentation, there are exceptions and considerable variation. Studies considering the causes of variations in representation levels for women in western democracies find fairly consistent results. Variations are explained by political factors, especially electoral institutions, by demographic and socio-economic factors, and by cultural variables (Darcy, Welch, and Clark 1994; Norris 1985; Rule 1981, 1987, 1994). This note retests this earlier work on later data to see whether factors that were important in the early 1980s retain their significance. It then moves on to whether theories based on results from OECD countries also hold in less developed countries (LDCs) with stable democracies. Earlier analyses concentrated exclusively on advanced industrialized democracies. This note provides a useful test of the robustness of these findings by testing the theories in less developed countries.

While comparative politics scholars regularly urge the inclusion of a broader range of countries in comparative research, attaining this goal often runs into a couple of serious technical problems. One is lack of data. For many countries and legislatures, data for desired variables are not available. I deal with this difficulty by using indirect measures of some relevant concepts. A second obstacle is defining the sample of democratic countries that should be included in the study. Countries that meet one standard may fail a different standard. I use an adjusted index of democracy developed from measures in the POLITY II data set compiled by Ted Gurr (Gurr, Jaggers, and Moore 1990). I adjusted the original index de-emphasizing its heavy weighting on chief executive powers and recruitment and placing greater weight on the breadth of political participation (see Matland (1994) for a detailed description of the sample selection process). Using this adjusted democracy index I identified the 24 advanced industrialized democracies and 16 democracies in LDCs shown in tables 1 and 2.

(TABLES 1 and 2)

The rest of the paper discusses the level of women's representation across the countries in the sample, the relevant independent variables, the regression results for industrialized and less developed countries, and implications of the findings.

Tables 1 and 2 show women's representation levels in 1980, 1990, and 1997 for the national legislatures of the 24 industrialized democracies and the 16 LDCs with democratic regimes. The difference of means between these two groups is statistically significant for all three time periods and indicates that the two samples differ in meaningful ways. Inspection of the data shows the gap between the industrialized democracies and the democratic LDCs has widened over time. From 1980 to 1997 women have made gains averaging 11.0% across the industrialized democracies, while the average gain across the LDCs has been a much more modest 4.5%. Among the industrialized countries the Scandinavian countries are all clustered at the top of the chart. The biggest gain over this time period was 24.9% for New Zealand; 8% of that gain (from 21.2% to 29.2%) occurred in October 1996 as New Zealand held their first election under a mixed electoral system. Among the LDCs all the gains are modest, with Argentina being a stunning exception. In Argentina a constitutional amendment requiring the nomination of women to at least one-third of the viable seats has had a dramatic effect on women's representation (Jones, 1996).

Independent variables found to be significant in previous research in advanced industrialized democracies fall into three categories: political/electoral institutions, cultural variables, and socio-economic variables.

Political Variables: The existence of a proportional representation (PR) electoral system has been found to have a positive and statistically significant effect on the percent of MPs who are women (Rule 1981, 1987, 1994; Norris 1985). The most important reason that women fare better under PR systems relates to party strategy in putting together a slate of candidates. In single member district systems parties only nominate a single candidate, in PR systems a party nominates several candidates. Because of the strictly zero sum nature of nominating decisions in single member districts, female candidates must compete against existing interests in the party that are represented by men. In PR systems the party is much more conscious of balancing their ticket to attract support from different constituencies. A woman candidate can be seen as a benefit to the ticket by attracting voters, without having the significant costs to intra-party peace of requiring powerful intra-party interests represented by men to step aside. This lower opportunity cost also makes it more likely that parties will react quickly to another party’s conscious promotion of women (Matland and Studlar 1996).

Rule (1987) found one other political variable, the proportion of seats held by right wing parties, also affected female representation. Right wing parties are expected to support more conservative and traditional values that discourage women's participation in politics and Rule's analysis of early 1980s data supports this assertion.

Socio-Economic Factors. Participation in the labor force has consistently been found to have a significant, positive effect on women's level of political activity (Anderson 1975; Welch 1977; Togeby 1994). Increased levels of activism and political consciousness could easily result in demands for greater representation of women. Rule (1987) and Norris (1985) both found women labor force participation rates had a positive affect on women's representation, although only Rule found this effect to be statistically significant.

Political activism, besides being positively related to work force participation, also increases with level of education. In addition, because members of the national legislature are disproportionately well educated (Putnam 1976), increasing levels of university education among women should expand the pool of possible women candidates (Darcy, Welch, and Clark 1994). Therefore, I expect a positive relationship between the proportion of women with some university education and women’s representation in national parliaments. Both Rule (1987) and Norris (1985) found such a relationship, but only Rule found the effect to be statistically significant.

Cultural Variables: While many researchers emphasize the importance of political culture, developing good measures of cultural differences is quite difficult. Norris used data from the 1977 Eurobarometer to develop a measure of political egalitarianism. She found political egalitarianism positively affected the proportion of women in a parliament. Since the present data set includes several non-European countries, Eurobarometer measures cannot be used. Instead, a cultural measure of women’s standing in society was developed by factor analyzing three measures that compare women's standing with that of men. The measures used were the ratio of women's literacy to men's literacy, the ratio of women's labor force participation to men's labor force participation, and the ratio of university educated women to university educated men. I assume that when women approach men in levels of literacy, work force participation, and university education, these are all indications that women's social standing is approaching that of men. The more equal women's social standing, the more likely they are also seen as men's equal in the political sphere and the greater the probability they are represented in equal numbers. The measure is constructed so that the more positive the number, the greater the equality between men and women.

To this set of previously identified variables I add one additional variable, level of development. Tables 1 and 2 show there are clear differences between industrialized democracies and LDCs with a democratic regime. Previous studies looking at OECD countries have not tested the level of development. Significant variation exists, nevertheless, across these countries in development level and this may help explain variations in representation. As countries become more developed, women are increasingly integrated into all spheres of public life; this should include representation in the national legislature (Christy 1987). Several processes that accompany development should increase women's political resources and decrease existing barriers to political activity. Development leads to weakening of traditional values, decreased fertility rates, increased urbanization, greater education and labor force participation for women, and attitudinal changes in perceptions of the appropriate roles for women. Confirmatory factor analysis is used to create a measure of development based on three variables: the literacy level in the country, an energy use measure (barrels of oil equivalent), and a 1990 estimate of GDP adjusted for purchasing power equivalence. Note that this developmental factor is distinct from the cultural factor. The development factor looks at the absolute level of development, while the cultural factor looks at women’s position relative to men’s. The correlation between the two measures is .58, indicating the two are related, but that they do vary independently.

The initial regression tests variables found statistically significant in previous studies plus level of development. It includes an electoral system variable (this equals 1 if a country has a PR system, 0 if the system is majoritarian), percent of parliamentary seats held by right wing parties, the cultural factor score described above, the development factor score described above, women's labor force participation rate, and the percent of women with some college education. Except for the cultural and development factors, I use the same measures as used previously. The equation presented below reports unstandardized regression coefficients and their standard errors.

Dependent Variable: Female Percent of Parliament

*** = sig. at .01 level (2-tailed test)
** = sig. at .05 level (2-tailed test)
* = sig. at .01 level (2-tailed test)

 Adjusted R Square .75

Standard Error 5.17

F = 12.29 Sig. F = .0000

X1 = Electoral System
X2 = Women's Labor Force Participation Rate
X3 = Women's Comparative Standing, Culturally [Factor Variable]
X4 = Level of Development [Factor Variable]
X5 = Percent of Adult Women with University Education
X6 = Proportion of Seats Held by Right Wing Parties

For most variables the regression analysis provides a clear picture. The electoral system has a powerful, statistically significant, effect. The model predicts that for industrialized democracies changing from a majoritarian to a PR system will result in a 15-16% jump in the female proportion of the national legislature. Women's labor force participation rates and the cultural factor also show noticeable positive effects on women’s representation. It is just as clear that the proportion of seats held by right wing parties is unrelated to levels of female representation. The last two variables considered produce some uncertainty. The variable measuring proportion of women with university education is close to being statistically significant (t=-1.60, p=.13), despite the small number of cases. On the other hand, the result is fairly perplexing. It is in the opposite direction of previous findings and seems to suggest that as the number of college educated women increases, representation decreases! Further inspection of the data reveals this result is largely caused by extreme values for two cases, and disappears when these cases are removed. Given the variable is not statistically significant, and what results there are depend on extreme cases, it seems reasonable to assert that the proportion of university women in the population is not directly related to female representation in either a positive or negative manner.

The regression also shows the direct effect of development is not significant. When the indirect effects via development's affect on women's labor force participation and on the cultural standing of women are considered, however, level of development does influence women’s representation. In addition, when development is included in the model without the cultural factor it has a strong statistically significant. Women’s cultural standing is obviously influenced by the level of development, and the two of them in combination deserve to be included in the model. Although all these countries are considered industrialized democracies, those countries with higher levels of development have greater proportions of women in their national legislatures.

This replication reconfirmed the effect of three previously identified factors: proportional representation, women's labor force participation rates, and cultural standing. One additional factor, level of development is significant in differentiating this sample of OECD countries.

The next step is to apply this model to the developing democracies.

Women’s participation in politics in LDCs has been studied rarely and when it has it has been largely through single country studies. An important exception is Nelson and Chowdhury’s (1994) Women and Politics Worldwide. They consider much more than representation in national legislatures, but they do discuss factors affecting women’s participation in formal political arenas. In drawing conclusions from the 43 countries included in their study they suggest political socialization as to women’s proper role in politics plays an important role, but they also argue that political parties and women’s organizations engaged in interest articulation determine the degree to which women participate in political life.

The variables considered for the OECD democracies includes several that should affect women’s representation in LDCs with democratic regimes. The development variable, the cultural variable, and women’s labor force participation rate all tap important elements that can affect perceptions of women’s proper role in society. While measuring how individual political parties react to a demand for greater access for women is not possible, it is possible to test whether the electoral system variable, which was seen as crucial because it changed how parties looked at nominating and electing women in OECD countries, acts in the same manner in LDCs. Results from running the model developed in the first part of the paper on 16 LDCs are presented below.

Dependent Variable: Female Percent of Parliament

Adjusted R Square .37
Standard Error 2.60
F = 3.21 Sig. F = .06

** significant at the .05 level.

X1 = Election System Dummy

X2 = Women's Labor Force Participation Rates
X3 = Women's Comparative Standing, Culturally [Factor Variable]
X4 = Level of Development [Factor Variable]

The results are surprising. The coefficients for development level and women's labor force participation are negative. The electoral system variable, which had such a substantial effect in the developed countries, has a coefficient ten times smaller and is non-significant. Only one variable appears to work as expected, that is women's cultural standing. This finding, however, is not robust. While cultural standing and level of development use different components, they are, nevertheless, correlated at a fairly high level for such a small sample (.64). There is some danger that they have split the variances between them with all the positive factors loading on the cultural variable and all the negative factors loading on the development variable. This suspicion is borne out by further analysis. When development is dropped from the equation the cultural variable plummets to less than half its previous size and is no longer significant (b= .85, s.e. of b=1.09). If the cultural variable is dropped from the equation, development also drops to half its former size and is no longer significant (b=-1.30, s.e. of b=.90). In both cases, once only one of the variables is in the equation, none of the variables are statistically significant and the F-test shows the model as a whole does not come close to explaining enough variance to be statistically significant.

The initial evaluation must be that factors influencing women’s representation in advanced industrialized democracies do not work in the same fashion in LDCs with democratic regimes. On reflection, there are a couple reasons why these results are not surprising. First the results may accurately reflect that female representation is still so low in LDCs that representation may largely be determined by idiosyncratic conditions within that country and not by broad forces influencing all LDCs. Despite substantial variation on the independent variables, the dependent variable falls in a very narrow band. Only two countries in 1990 (Costa Rica and El Salvador) are above 10%, and over half the sample is clustered between 5.0% and 6.7%. These legislatures may not have enough women in them to exhibit any consistent identifiable patterns across countries.

A second reason for the lack of identifiable effects is that the independent variables may mean something entirely different in this new context. For example, in industrialized countries women who previously worked in the home and have now moved into paid work outside the home are the primary cause of the increase in women’s labor force participation (Togeby 1994). Moving into the paid labor force, often into low paying, or public sector unionized jobs, has a consciousness raising effect on women's political participation and their propensity to articulate political demands. While women's labor force participation rates are quite high in many LDCs, this is largely due to women's presence in subsistence level primary sector work. This work is quite unlikely to have the same empowering and consciousness raising effect. Labor force participation does not mean the same thing in these two worlds, and therefore it is not surprising that the effects are different.

The inability to find other variables that differentiate among these countries intensifies the suspicion that idiosyncratic factors unique to each country are at the heart of variations across these countries. Several plausible variables were tested and none exhibited a strong enough influence on representation to meet even weak tests of significance. Among the variables tested were percent of seats won by left wing (and right wing) parties, level of urbanization, labor force participation adjusted by subtracting primary sector jobs, party magnitude, number of years since women gained the vote or were first elected, and percent of population that is Catholic. While some variables show effects in the expected direction, all effects were quite weak.

While the regression failed to show any meaningful statistically significant results, it still provides important information. The failure of the electoral system variable to have an effect is especially interesting. That ticket balancing occurs in PR systems both makes logical sense and is shown to work to women’s favor in the industrial democracies. The failure of the same mechanism to work in women’s favor in LDCs suggests either demands for representation are not being fowarded by women, perhaps because they are politically inactive, or, within the parties, the perceived costs to nominating women are so great that parties prefer to run virtually all male slates rather than risk the wrath of the voters by nominating women. Detailed individual country studies are needed to investigate these dynamics.

The non-effect for the electoral system variable is an important example of a more general point. Comparative institutional analysis concentrates on analyzing how different institutional arrangements can provide an advantage or disadvantage to specific interests. These institutional effects, however, can only occur if the inputs (demands) exist to take advantage of the institutional arrangements. Otherwise, the institutional arrangement may not affect outcomes. An unspoken behavioral component is almost always part of any institutional argument. Here it is that if forces interested in women’s representation are effectively organized, they will be more successful in achieving representation when a PR electoral system is used rather than a majoritarian system.

These results provide two very distinct pictures. Factors driving variations in representation in the developed world are clearly understood. There is much less clarity in the developing world. In the developed world, the effects of economic, cultural, and political factors identified in previous research were confirmed. Perhaps most striking is the enormous effect a proportional representation electoral system has on representation. In the developing world, however, none of the variables found significant among advanced industrialized democracies, nor several other plausible variables, were found to have an effect.

Based on these results there appears to be a threshold, a minimum development level is needed to create the foundation for other variables to have an effect. Below that level, the variables that assist women in gaining representation in more developed countries simply have no effect. It appears that in most LDCs the forces aligned against female political activity are so great as to permit only token representation. As development increases, however, more women start to acquire the resources needed to become politically relevant, resources such as education, salaried labor force experience, and training in the professions that dominate politics (such as law). This leads to the formation of a critical mass. When the number of women with the necessary resources becomes substantial the opportunity for effective interest articulation exists. Development is a crucial part of this process.

The second part of the story is that different political systems provide different levels of success after the critical mass is reached. While having the resources is an essential condition, clearly the political system plays a significant role. For a variety of reasons, the electoral opportunity structure in proportional representation systems provide greater access to women than majoritarian systems. When development levels are low, these structures are not utilized or additional barriers block the effective use of these mechanisms. As development occurs, however, women are more likely to see significant increases in their representation in polities that use a proportional representation electoral system.

Considerable work remains to be done. For developing countries, case studies, especially of countries where representation has reached moderately high levels, would be valuable tools in uncovering relevant variables that affect representation across LDCs. Studying the strength and actions of women’s organizations involved in interest articulation in these countries would be especially useful, to see if they hold the key to variations in representation. For advanced industrialized countries, individual country studies may help us discover how political variables interact with environmental factors to increase representation. As time passes existing countries will develop and changes in representation will occur. The number of new democracies has shown a promising upward trend. As new data becomes available they will provide opportunities to retest and expand upon these findings.

 REFERENCES

Anderson, Kristi (1975) "Working women and political participation, 1952-1972", American Journal of Political Science 19: 439-453.

Anderson, Kristi and Stuart Thorson. (1984) "Some Structural barriers to the Election of Women to Congress: A Simulation", Western Political Quarterly 37:143-156.

C.I.A. (1992) CIA World Factbook, 1992 Washington, D.C.: U.S. Govt. Press.

C.I.A. (1993) CIA World Factbook, 1993 Washington, D.C.: U.S. Govt. Press.

Christy, Carol (1987) Sex Differences in Political Participation. New York: Praeger.

Darcy, R., and Karen Beckwith (1991) "Political Disaster, Political Triumph: The Election of Women to National Parliaments". Paper presented at the annual meeting of the American Political Science Association, Washington, D.C. August 29-September 1, 1991.

Darcy, R., Susan Welch, and Janet Clark (1994) Women, Elections, and Representation. 2nd ed. Lincoln, Neb: Nebraska University Press.

Gurr, Ted Robert; Keith Jaggers, and Will H. Moore (1990) "The Transformation of the Western State: The Growth of Democracy, Autocracy, and State Power Since 1800." Studies in Comparative International Development 25:73-108.

Huber, John and Ron Inglehart (1995) "Expert Interpretations of Party Space and Party Locations in 42 Societies" Party Politics 1:73-111.

Inter-Parliamentary Union (1993) Electoral Systems: A Worldwide Comparative Study
Geneva, Switzerland: IPU.

Inter-Parliamentary Union (1995) "Women in Parliaments: 1945--1995 A World Statistical Survey" Series "Reports and Documents", No. 23, Geneva, Switzerland: IPU.

International Labour Organisation (1993) Yearbook of Labour Statistics 1993 Geneva, Switzerland: ILO Publications.

International Labour Organisation (1994) Yearbook of Labour Statistics 1994Geneva, Switzerland: ILO Publications.

International Labour Organisation (1995) Yearbook of Labour Statistics 1995 Geneva, Switzerland: ILO Publications.

Jones, Mark A. (1996) "Increasing Women's Representation Via Gender Quotas: The Argentine Ley de Cupos". Woman and Politics, 16(4): 75-98.

Klingemann, Hans Dieter, Richard I. Hofferbert, and Ian Budge (1994) Parties, Policies, and Democracy Oxford: Oxford University Press.

Laver, Michael and W. Ben Hunt (1992) Policy and Party Competition New York: Routledge.

Mackie, Thomas T. (1988) "General Elections in Western Nations During 1987" European Journal of Political Research 16: 573-584.

Mackie, Thomas T. (1989) "General Elections in Western Nations During 1988" European Journal of Political Research 17:747-752.

Mackie, Thomas T. (1991) "General Elections in Western Nations During 1989" European Journal of Political Research 19:157-162.

Mackie, Thomas T. (1992) "General Elections in Western Nations During 1990" European Journal of Political Research 21:317-332.

Matland, Richard E. (1993) "Institutional Variables Affecting Female Representation in National Legislatures: The Case of Norway", Journal of Politics 55: 737-755.

Matland, Richard E. (1994) "Comparing Legislative Represenation of Women in Developing and Developed Countries." paper presented at the American Political Science Association Meetings, New York, NY, Sept. 1-4, 1994.

Matland, Richard E. (1995) "How The Electoral System has Helped Women Close the Representation Gap in Norway", in Lauri Karvonen and Per Selle Closing the Gap: Women in Nordic Politics, Dartmouth Press: London.

Matland, Richard E. and Donley T. Studlar (1996) "The Contagion of Women Candidates in Single Member and Multi-Member Districts" Journal of Politics 58(3): 707-733.

Matland, Richard E. and Michelle A. Taylor (1997) "Electoral System Effect on Women's Representation: Theoretical Arguments and Evidence from Costa Rica." Comparative Political Studies, 30(2): 186-210.

Nelson, Barbara and Naijma Chowdhury (1994) Women and Politics Worldwide, New Haven: Yale University Press.

Norris, Pippa (1985) "Women's Legislative Participation in Western Europe." Western European Politics 8: 90-101.

Putnam, Robert (1976). The Comparative Study of Political Elites Englewood Cliff, NJ: Prentice-Hall.

Rule, Wilma (1981) "Why Women Don't Run: The Critical Factors in Women's Legislative Recruitment", Western Political Quarterly 34:60-77.

Rule, Wilma (1987) "Electoral Systems, Contextual Factors, and Women's Opportunity for Election to Parliament in Twenty-Three Democracies", Western Political Quarterly 40: 477-498.

Rule, Wilma (1994) "Parliaments of, by, and for the People: Except for Women?" In Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities, ed. Wilma Rule and Joseph F. Zimmerman, Westport, CT: Greenwood

Togeby, Lise (1994) "Political Implications of Increasing Numbers of Women in the Labor Force", Comparative Political Studies 27:211-240.

UNESCO (1992) Statistical Yearbook 1992 Paris: UNESCO Press.

UNESCO (1994) Statistical Yearbook 1994 Paris: UNESCO Press.

UNESCO (1995) Statistical Yearbook 1995 Paris: UNESCO Press.

Wagshal, Uwe (1996) "Parties, Party Systems and Tax Smoothing" Paper presented at the ECPR Joint Sessions Oslo, Norway, March 29-April 3, 1996.

Welch, Susan (1977) "Women as Political Animals? A test of some explanation for male-female political participation differences." American Journal of Political Science 21:711-730.

The POLITY II data set includes measures of autocracy, democracy, state power, and several sub-elements of each of these scales for 155 countries from 1800 to 1986. Four separate elements make up the original measure of democracy reported in Polity II: competitiveness of political participation, competitiveness of executive recruitment, openness of executive recruitment, and constraints on the chief executive. Seven of the ten points in the original scale are tied to the chief executive. For this study, however, the chief executive is of less interest and the general stability and legitimacy of political competition is more relevant. Therefore, I developed a modified democracy scale that decreases the weight on each of the chief executive components and includes an additional measure from the Polity II data set, regulation of participation. The regulation of participation variable measures the degree to which political competition is institutionalized and rules ensure participation of all relevant perspectives. For each element of the scale, data from the most recent year in the data set are used. Listed below are both the adjusted democracy scores and the original scores reported in the POLITY II data set for the 24 advanced industrialized democracies included in the study. My adjusted democracy score, which places less emphasis on chief executive powers, provides France, Greece, and Spain with higher scores than the democracy scale used in the Polity II data set (France and Greece still fail to score 10s because of powerful chief executive power).

COUNTRY ADJUSTED POLITY II

Listed below are developing countries that have been considered examples of third world democracies. This study includes only the sixteen third world countries with a score of 8.0 or higher on the adjusted democracy score. (Malta is included as one of these sixteen countries. While Malta was not included in the Polity II data set, it is by all accounts a stable democracy.) Looking at the countries included in the sample, the list seems to have a fair amount of face validity and greater face validity than the Polity II scores. If we use the admittedly arbitrary cutoff point of scoring greater than 8.0 on the democracy scale, the switch from the Polity II measure to the adjusted democracy measure leads to the inclusion of France, Uruguay, Argentina, Brazil, and India, and the dropping of Guatemala, Thailand, Trinidad & Tobago, and Zimbabwe. Taken as a whole, these seem to be steps in the right direction. The adjusted score also appears to have greater discriminant validity than the original measure. There is a sizeable gap in the new scale with no countries scoring between 6.0 and 8.0, providing an easy cut point between democratic and non-democratic regimes.

 TABLE 1: FEMALE REPRESENTATION IN THE NATIONAL LEGISLATURES OF ADVANCED INDUSTRIAL DEMOCRACIES, IN 1980, 1990, and 1997.**

** Data taken from Inter-Parliamentary Union Report #23, "Women in Parliaments: 1945-1995, A World Statistical Survey" Geneva, Switzerland: IPU, 1995 and the IPU’s world wide web site: http//www.ipu.org

Return to Home Page