Content uploaded by Susan Welch
Author content
All content in this area was uploaded by Susan Welch on Apr 12, 2016
Content may be subject to copyright.
University of Utah
The Opportunity Structure for Women's Candidacies and Electability in Britain and the
United States
Author(s): Susan Welch and Donley T. Studlar
Source:
Political Research Quarterly
, Vol. 49, No. 4 (Dec., 1996), pp. 861-874
Published by: Sage Publications, Inc. on behalf of the University of Utah
Stable URL: http://www.jstor.org/stable/449141
Accessed: 12-04-2016 01:52 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
University of Utah, Sage Publications, Inc.
are collaborating with JSTOR to digitize, preserve and
extend access to
Political Research Quarterly
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
The Opportunity Structure for
Women's Candidacies and
Electability in Britain and the
United States
SUSAN WELCH, PENNSYLVANIA STATE UNIVERSITY
DONLEY T. STUDLAR, WEST VIRGINIA UNIVERSITY
The influence of the social and political environment on candidacies and
electability is an important element of political recruitment structures. We
investigate several contextual factors that may affect the likelihood of women
becoming candidates to, and members of, the British House of Commons
and the U.S. House of Representatives. Using data from the 1992 British
General Election and the 1992 U.S. congressional elections, we find both
similarities and differences in the contexts promoting women's candidacies
in the two countries. Overall, there is not an especially distinctive pattern
in either country to women's candidacies or electability. Women have moved
beyond "sacrificial lamb" status, but incumbency remains the most formi-
dable electoral barrier in both countries.
Many political scientists have examined the question of why women's access
to political office in Western democracies has lagged far behind their access to
the vote. However, these studies have focused largely on the fate of women
who are already candidates rather than on how women become candidates.
Researchers into political recruitment have traditionally divided the re-
cruitment process into three stages: eligibility, candidate selection, and elec-
tion (Rule 1981). Recently, several scholars have developed more elaborate
NOTE: Earlier versions of this article were presented at the meeting of British Politics
Group-American Political Science Association, New York, September 1-4, 1994,
and the West Virginia Political Science Association, Morgantown, October 7-8,
1994. Thanks to Deborah Wituski, Bob Duval, Michael Goldcamp, Alex Jalso,
Susan Hunter, Kathy Fletcher, Henry Rieke, and Kent Woodson for technical as-
sistance, to the Senate House Library, University of London, for research facilities,
and to Rebekah Herrick, Neil Berch, Jim Forester, Karen Beckwith, Anthony
Messina, and anonymous PRQ reviewers for comments on earlier drafts.
861
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
models of the recruitment process (Lovenduski and Norris 1989). Consider-
able attention has focused on the concept of the opportunity structure, that is,
how myriad potential candidates who meet legal and practical eligibility crite-
ria are winnowed to the relative few who reach the formal selection process
(Schlesinger 1966; Carroll 1994: 158; Burrell 1994).
Studies of women candidates in the United States and cross-nationally
have shown that political opportunity structures such as type of electoral sys-
tem, district magnitude, incumbency, party, and level of the political office are
significant influences on women's access to legislatures, even in comparison
to individual social and political factors, which are also important. However,
all of these factors have more often been used to explain women winning
office rather than their becoming party candidates (Fowler 1993).'
This paper develops a more comparative perspective on the opportunity
structure for central-level legislatures by examining whether there are certain
kinds of constituencies where local party selectors (in Britain) or primary elec-
torates (in the United States) are more likely to nominate women, and by
assessing how women candidates in the two countries fare for election under
different social and political contexts. The two countries have single-member
district electoral systems, with the effective choice of candidates lying in local
hands (party organizations in Britain, electorates in the United States), in con-
trast to more centralized nominations in proportional representation systems
(Lovenduski and Norris 1993). This study analyzes all major party candi-
dates in elections to the British House of Commons in 1992 and to the U.S.
House of Representatives in 1992. In both, there was a substantial increase
in women's candidacies and election to the lower house.
PREVIOUS RESEARCH AND CURRENT EXPECTATIONS
Constituency socioeconomic and political culture characteristics, partisan fac-
tors, and the electoral context, particularly whether the seat is safe, competi-
tive, or hopeless for a candidate's party, may affect women's candidacies. Much
previous research has concentrated on the opportunity structure for women
at lower levels of government in the two countries because, until recently,
there were few women candidates and legislators at the central level.
Constituency Characteristics
Constituencies where women candidates are not welcome will obviously
diminish the number of potential women candidates entering the formal
1 Space precludes citing much of the now voluminous literature on women candidates
in the U.S. and Britain. Recent comprehensive reviews are found in Darcy et al. 1994;
Burrell 1994; Lovenduski and Norris 1993).
862
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
selection process. Two constituency characteristics seem especially wor-
thy of exploration: the social class of voters in the constituency and the
general cultural traditionalism of the area. In both countries, women are
slightly more likely to be elected in more middle-class rather than work-
ing-class areas (Bristow 1980; Welch and Studlar 1988; Studlar and Welch
1987; Nechemias 1987; but see Rule 1990). We expect better-off districts
to provide more women candidates and legislators at the central level be-
cause women in such areas may have occupations that allow them oppor-
tunities to run for office.
Women candidates and elected officials are also more common in less
traditional constituencies. Region is one crude measure of traditionalism. In
Britain, the more traditional areas include Scotland, Wales, and the North of
England. Women councillors form a larger proportion of the local councils in
England than in Scotland or Wales (Studlar and Welch 1987; Welch and Studlar
1990), while London is more conducive to the election of women than more
rural, traditional areas (Bristow 1980; Studlar and Welch 1991). Lovenduski
and Norris (1989) find that women parliamentary candidates are least likely
to be selected in Northern England, Yorkshire, Scotland and Wales (see also
Vallance 1984: 306-307; Rasmussen 1983).
In the U.S., the South is the primary source of traditional culture and the
West is least traditional, especially with regard to women's roles (Bennett and
Bennett 1992). Hence it is not surprising that, in the U.S., women state legis-
lators and members of the House have been more likely to come from states
outside the traditionalistic southern culture (see, most recently, Rule 1990;
Nechemias 1987; Burrell 1994).
Another indicator of a distinctive culture is the persistence of traditional
views of women's roles. We would expect more women candidates to emerge
in areas where views of women's roles are more open, for example, where
women are more likely to be employed outside the home. Earlier studies
found professional status and higher education qualifications to be character-
istic of British women MPs and candidates (see, for example, Lovenduski and
Norris 1989) and for women members of Congress (Burrell 1994). At the
aggregate level, the states in the U.S. with greatest proportion of women legis-
lators are also those with the greatest proportion of professional women in the
workforce (Rule 1990; Nechemias 1987).
Partisan Factors
Parties that encourage women candidates, or at least do not discriminate against
them, by definition offer a better opportunity structure for women. Cross-
national studies indicate that leftist parties often take the lead in nominating
women candidates (Norris 1987: 128; Rule 1987), although in some coun-
863
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
tries right-wing parties are as likely as left-wing ones to select women candi-
dates (Norris 1987; Lovenduski and Norris 1993).
In Britain and the U.S., neither parties of the left nor right can claim a
clear advantage in nominating women candidates. In Britain, the Labour party
is more ideologically open to gender equality, and in London Labour does
nominate women for more winnable local council seats (Studlar and Welch
1991). Labour is also more likely than the Conservatives to nominate women
candidates for Parliament (Rasmussen 1977; Hills 1981; Vallance 1984; Studlar
et al. 1988; Norris 1993). Women, however, are a sizable proportion of local
Conservative activists and have increased their proportions of nominations
for local elections (Bristow 1980). The Liberal and Nationalist parties have
been most receptive to women candidates, though most of them, like their
male counterparts, have been defeated (Norris 1993).
In the U.S., women traditionally were a smaller proportion of legislatures
controlled by Democrats than by the Republicans (Rule 1981), largely be-
cause the former were located predominantly in the traditional South. Re-
cently women still have been somewhat more likely to be elected in Republican
controlled legislatures (Nechemias 1987; Rule 1990), but in 1993, over 60
percent of women state legislators were Democrats. Women have usually
secured more congressional nominations in the Democratic party, but only
slightly greater chances of election (Burrell 1994). The recent Democratic
advantage in electing women candidates has been due largely to the fact that
the seats they contested were more Democratic in partisan orientation (Berch
1996).
Electoral Context
It is sometimes alleged that women candidates are disproportionately slated
for "hopeless" seats (Gertzog and Simard 1981; Carroll 1994), running against
incumbents or in districts where their party received small proportions of the
vote. For instance, in the 1982 and 1984 congressional elections, women
candidates were not much more likely to run against incumbents, but were
somewhat more likely to run in hopeless races, those where the incumbents
had received 60 percent or more of the vote (Burrell 1988). Between 1970
and 1992, nonincumbent male and female candidates were about equally likely
to run for open seats or against incumbents (Darcy et al. 1994).
In Britain, women's disadvantage in terms of competitiveness of constitu-
encies contested is small and has shrunk (Studlar et al. 1988). Although
Labour continues to be more likely to nominate women for winnable parlia-
mentary seats than other parties (Rasmussen 1977; Studlar et al. 1988), re-
cently Conservative women, although fewer in number, have been
proportionately more likely to emerge victorious (Norris 1993).
864
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
Expectations
In sum, based on these somewhat contradictory previous findings, we expect
that women are more likely to be candidates in constituencies of higher socio-
economic class standing; in nontraditional areas, where women's roles are less
fixed; in more left-wing parties; when challengers rather than incumbents;
and, as challengers, in seats where their party is less likely to be able to win
the seat.
We also expect these factors to influence women candidates' electoral suc-
cess. Carroll (1994) found that political opportunity structures were more
important in explaining women victors than women candidacies, although
this varied somewhat by level of office. Burrell's (1994) analysis found some
differences in the attributes of elected and unelected female U.S. congres-
sional candidates. We also compare the factors predicting success of women
and men candidates (see Burrell 1994).
DATA AND METHODS
The analysis is based on data from the 1992 British General Election and the
1992 election for the House of Representatives in the United States. The unit
of analysis is the candidate, and data were collected on the characteristics of
each candidate and on the constituency itself (data sources include the 1990
U.S. Census, 1991 British Census, Times Guide to the House of Commons 1992,
and Almanac of American Politics, 1993). The data include only candidates of
major parties. For Britain, we define "major party" as Labour, Conservative,
Liberal Democrat, and the Nationalist parties in Scotland and Wales. North-
ern Ireland is not included.
The two dependent variables include whether (= 1) or not (= 0) the can-
didate is a woman and whether (= 1) or not (= 0) the candidate won. This
analysis was done separately for men and women, thus allowing us to assess
what characteristics distinguish women winners from women losers and
whether these characteristics are also observable for men.
We use six sets of independent variables. The census measures of the
proportion of houses in the constituency that are owner occupied is a mea-
sure of social class.2 Region, as an indicator of traditionalism, is measured by
a set of dummy variables indicating major regional divisions in British society
(including national divisions): the Southeast, Southwest, Greater London, East
Anglia, West Midlands, North, Yorkshire and Humberside, Scotland, and Wales,
2 The proportion of the workforce who are professionals and managers, another possible
measure of the constituency's social class, is highly correlated with the proportion in
owner occupied housing (-.54).
865
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
with Scotland being the omitted category in the equations. In the U.S., we use
a fourfold categorization: South, West, Midwest, and East, with the West be-
ing the omitted category. We also measure the traditional status of women by
the proportion of women in the workforce in the constituency or district,
with a higher proportion an indication of a less traditional view of women's
role.3 The political variables include party, incumbency, and electoral com-
petitiveness of the district. Party is coded as a set of dummy variables. Two
dummy variables indicate whether the seat is "safe," won by the candidate's
party by a margin of at least 10 percent in the previous election, or "competi-
tive," where the candidate's party won or lost by a margin of 10 percent or
less. The omitted category is "hopeless," where the candidate's party lost by
more than 10 percent.4
Because the dependent variables are dichotomous, logistic regression analy-
sis is used (Aldrich and Nelson 1986). Logistic regression coefficients are
similar to regression coefficients. Dividing each by its standard error yields a
t value. However, logistic coefficients do not have the same straightforward
interpretation as unstandardized regression coefficients, where a one unit
change in the independent variable yields a change in the dependent variable
equal to the size of the coefficient.
FINDINGS
Since the 1916 to 1938 period, the number of women's candidacies for the
U.S. House has grown slowly (Gertzog and Simard 1981). From 6 per elec-
tion in these years, to an average of 32 between 1964 and 1978, 59 in the
elections of the 1980s, and 69 in 1990 (Burrell 1994), the number rose to 106
major party women in 1992. Still, only 13 percent of all candidates were
women, even in this so-called "Year of the Woman." Similarly, the number of
candidates to the Commons has also grown slowly, from 5 percent of all major
party candidates in 1945 (Lovenduski and Norris 1989) to 18 percent of all
major party candidates in 1992 (Norris 1993).
Proportionally, then, women were nearly half again more likely to be can-
didates in Britain than in the United States. This may be because of the larger
Unfortunately problems of comparability in the two Censuses prevent us from using
other measures of social context, including income and urbanization. Another variable
often studied in the United States is campaign finance, but British campaign finance is
more egalitarian and centrally disbursed. Studies based on recent congressional elec-
tions in the United States (Darcy et al. 1994; Burrell 1994; Berch 1996) have sometimes
found these missing variables to be significant in the nomination and election of women,
even if only weakly so.
4 The 10 percent margin has been used in previous research (Lovenduski and Norris
1989; Studlar et al. 1988).
866
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
number of competitive parties involved. However, women are as likely to be
incumbents in the U.S. as in Britain (see Table 1). In both polities, as pre-
dicted, women candidates are much more likely to be candidates in the more
leftist party than in the more conservative one.
TABLE 1
PARTY AND GENDER AMONG CANDIDATES IN 1992 BRITISH AND U.S. ELECTIONS
% Candidates % Women % Men
who are women who won who won
a. British House of Commons
Total 19 11 25*
Conservative 9 35 56*
Labour 22 28 48*
Liberal Democrat 22 2 4
Plaid Cymru 17 0 14
SNP 21 7 4
Other 23 0 0
Incumbents 8 91 91
Non Incumbents 22 4 7*
b. U.S. House of Representatives
Total 13 44 52
Republican 9 33 43
Democrat 17** 50 63*
Incumbents 8 89 95
Non Incumbents 16** 30 22
*Gender differences significant at p .05
**Differences in categories, i.e., Republican vs. Democrat, significant at .05.
Notes: For Britain: N = 2847, including 841 'other' party candidates. For the U.S.: N =
845.
Predictors of Women's Candidacies
In neither country is women's candidacy well predicted by our independent
factors. The R2s of .05 indicate poorly fitting models (see also Burrell 1994,
for a similarly poorly fitting model of women's congressional candidacies).
Examining constituency factors first, we find our prediction of a positive
relationship between the social class of the district and its propensity to have
women candidates was not confirmed. The proportion of owner-occupied
housing in a constituency is unrelated to the likelihood of having women
candidates.
Predictions of regional differences in the presence of women candidates,
with the more traditional areas having fewer women candidates, were par-
tially confirmed. In Britain, women are most likely to be candidates in London,
867
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
TABLE 2
THE IMPACT OF CONSTITUENCY AND ELECTORAL FACTORS ON WOMEN'S CANDIDACIES
FOR LEGISLATURE, 1992
regression standard error
coefficient of estimate
a. British House of Commons
Political Factors
Labour .21 .14
Conservative -.55* .19
Incumbency -.75* .24
Safe Seat -.33 .26
Marginal Seat .15 .20
Region
Wales -.10 .31
York -.06 .26
West Midlands .17 .25
North -.55 .38
Southeast -.18 .21
East Anglia -.16 .40
London .42* .20
Northwest .12 .24
Southwest .25 .26
% women in labor force .02 .02
% owner occupied housing -.00 .01
Constant -3.23 1.14
b. U.S. House of Representatives
Political Factors
Republican -.90* .25
Incumbency -.71* .35
Safe seat -.18 .36
Marginal seat .23 .34
Region
East -.64 .34
Midwest -.47 .32
South -.61 .32
% women in labor force .04* .02
% owner occupied house -.01 .01
Constant -2.57 1.55
* p .05, one tailed test except for r egion
N = 2006 British, 798 U.S.
-2 log likelihood, British 1764, 1989 df; U.S., 531, 788 df
Goodness of fit, British, 2006, 1989 df; U.S., 846, 788 df
Estimated R2 = .05 British, .05 U.S.
Omitted categories are Scotland (Britain), the West (U.S.), and hopeless seats
868
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
probably the least traditional area. In the U.S., women are least likely to be
candidates in the East and South. We attribute this to traditionalism of union
leaders and some socially conservative white ethnic groups who carry signifi-
cant weight in the Democratic nomination process in parts of the East, in
addition to well-delineated conservative attitudes toward women in the South.
Our predictions about traditionalism were also partially confirmed in the
relationship between women candidates and women in the workforce. In the
U.S., the likelihood of a woman candidate is higher in constituencies where
there are more women in the workforce. However, in Britain, this expected
relationship did not materialize.
Among factors related to the electoral context, in both countries the more
conservative party is significantly less likely to have women candidates than
the more leftist party Hence, Democrats and Labourites exceed Republicans
and Conservatives in women candidates, after controls for incumbency and
other factors.5
As expected, incumbency is a powerful predictor of women candidates in
both countries. In Britain, it is the best predictor of whether a candidate is a
woman or not; most women are challengers. In the U.S., the relationship is
the second best predictor after party
We had predicted that women were most likely to be candidates in hope-
less races and least likely in safe seats, with competitive districts in between.
In the U.S., the difference between the proportion in safe and hopeless seats is
significant. In Britain, neither of the coefficients achieved statistical signifi-
cance; women candidates are spread among different types of seats.
Predicting Winners and Losers
In addition to influences on whether a person becomes a candidate, we also
must look at factors aiding candidate success. We are particularly interested
in whether the same mechanisms promote both male and female success or
whether there are some special factors that affect men but not women, or the
reverse.
Table 1 indicates that both in the U.S. and Britain, men are more likely to
win their elections than women. The difference is significant among Conser-
vatives and Labourites in Britain and Democrats in the U.S. Male
5 Space limitations prevent us from detailing the actual probabilities of women's candida-
cies estimated from our equations. However, to illustrate, we can predict that among
districts of average proportions of women in the labor force and owner occupied hous-
ing, in London, and in marginal seats, 5 percent of Labour candidates are women com-
pared to 1 percent of Conservative; among non-incumbents, the figures are 19 and 5,
respectively
869
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
nonincumbents in Britain and incumbents in the U.S. are also more likely to
win than their female counterparts. Table 3 examines the independent effects
of these and other factors on candidate success. The fit of the model in pre-
dicting winning is much better than for predicting women's candidacies.6
TABLE 3
FACTORS PREDICTING WINNING AND LOSING
Women Candidates Men Candidates
coefficient (st. error) coefficient (st. error)
a. British House of Commons
Labour 4.80 (1.38) 2.69 (.37)
Conservative .77 (1.23) .46 (.37)
Incumbency 4.93 (1.19) 3.03 (.26)
Safe Seat 9.90 (1.91) 5.48 (.48)
Marginal Seat 3.45 (.86) 2.32 (.30)
% Owner occupied housing -.13 (.05) n.s.
% women in labor force .42 (.15) n.s.
Estimated R2 .42 .48
N (353) (1653)
-2 log likelihood 71.7, 336 df 608.2, 1636 df
Goodness of fit 2251.7, 336 df 1889.3, 1636 df
b. U.S. House of Representatives
Incumbency n.s. 2.69 (.34)
Safe seat 4.11 (1.17) 3.09 (.34)
Marginal seat n.s. 1.33 (.35)
Estimated R2 .42 .43
N (91) (707)
-2 log likelihood 59.5, 81 df 419.5, 697 df
Goodness of fit 101.5, 81 df 764.4, 697 df
Only those variables with significant (.05) relationships are included in the table. The
total equation includes all variables used in Table 2. R2 includes all variables in the equation.
Table 3 also suggests that the conditions facilitating election are similar
for both men and women. Political factors are clearly more crucial than de-
mographic ones. For men and women in both countries, running for a safe
seat and, in Britain, being an incumbent, are significant predictors of success.
In Britain, both men and women candidates do better if they are running on
6 In all four equations (men and women in both countries), R2 values using a standard
multiple regression formulation are between .42 and .48, indicating that these few char-
acteristics explain a lot of the difference in whether candidates win or lose.
870
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
the Labour or Conservative rather than as Liberal Democrat or Nationalist
candidate. In Britain and the U.S., male candidates have a better chance if
they are incumbents or running for competitive rather than hopeless seats,
but this is not true for women candidates in the U.S. Owner occupied hous-
ing and women in the labor force are systematically related to British women's
election chances. Women are more likely to win in constituencies with fewer
owner occupied houses, likely poorer and more urban areas, and in constitu-
encies where women are a larger part of the labor force. These factors have no
impact on British male candidates, nor do they affect men or women candi-
dates in the U.S. Region is not related to winning in either country.
CONCLUSIONS
This study is the first attempt at a focused, cross-national comparison of so-
cial and political opportunity structures for women candidates at the national
level in Britain and the United States. The constituency, electoral, and parti-
san factors we have examined explain only a small amount of the variation in
the presence of women candidates in these elections. Once candidates, women's
chances of election appear to be enhanced by the same factors that promote
men's chances; in other words the social and political opportunity structure is
relatively constant across districts. The "masculine candidate model"
(Lovenduski and Norris 1989) still obtains on the aggregate level.
In general, incumbency in a single-member district electoral system such
as those in Britain and the United States remains the most formidable barrier
for women to overcome in gaining equal access.7 Therefore variables which
weaken the power of incumbency, such as legal limitations on terms, large-
scale redistributions of seats (Pritchard 1992), and massive party shifts in
electoral results (Darcy 1992) may disproportionately help women's electability
if they are in positions to benefit as nominated candidates. Whether women
are so situated depends on the willingness of the parties to encourage and
allow women to be chosen as party candidates.
Although a two-country, two-election study is hardly definitive, and we
have examined only one recent election in each country, our findings do indi-
cate important similarities.8 These similarities are evident both in relation-
Incumbency depresses the number of women's candidacies (significantly in Britain);
the electoral success of incumbents is a barrier to challengers of either sex gaining
office. Since most incumbents are male, this barrier disproportionately affects women.
8 We have done a similar analysis for British candidates in the 1987 General Election and
U.S. candidates in the 1988 House elections although with slightly different social con-
text measures based on earlier censuses. The finding indicates considerable continuity
from the late 1980s to 1992, despite the large increase in women's candidates in the
871
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
ships that did obtain (such as the major role of incumbency and the positive
association between measures of nontraditional women's roles and women's
candidacies) and those not appearing (such as the influence of social class,
other things being equal). In comparison to previous studies, electoral com-
petitiveness of the constituency appears to have declined in both countries as
an influence on women's recruitment, while in the U.S., the Democrats have
now surpassed the Republicans in nominating and electing women to Con-
gress. Women are no longer sacrificial lambs in either country
There is obviously a need for greater integration of the individual and
aggregate characteristics which affect women's participation and electability
at different levels of the political recruitment process (see Fowler 1993) as
well as consideration of how similar objective circumstances may be viewed
differently by men and women (Bledsoe and Herring 1990). We have taken a
small step toward the first of these by showing, on a comparative basis, that
the social and political opportunity structure for women as party candidates
and electoral victors in legislative races remains important even as some of its
dimensions change.
REFERENCES
Aldrich, John, and Forrest Nelson. 1986. "Logit and Probit Models for Multi-
variate Analysis with Qualitative Dependent Variables." In William D. Berry
and Michael S. Lewis-Beck, eds., New Tools for Social Scientists. Newbury
Park, CA: Sage.
Barone, Michael, ed. 1993. The Almanac of American Politics 1994. Washing-
ton, DC: National Journal.
Bennett, Stephen Earl, and Linda L. M. Bennett. 1992. "From Traditional to
Modern Conceptions of Gender Equality in Politics: Gradual Change and
Lingering Doubts." Western Political Quarterly 45: 93-111.
Berch, Neil. 1996. "The 'Year of the Woman' in Context: A Test of Six Expla-
nations." American Politics Quarterly 24 (forthcoming).
Bledsoe, Timothy, and Mary Herring. 1990. "Victims of Circumstances: Women
in Pursuit of Political Office." American Political Science Review 84: 213-23.
Bristow, Stephen. 1980. "Women Councillors: An Explanation of the
Underrepresentation of Women in Local Government." Local Government
Studies May/June: 73-90.
Burrell, Barbara C. 1988. "The Political Opportunity of Women Candidates
for the U.S. House of Representatives in 1984." Women and Politics 8: 51-68.
latter year. The major difference is that, in both countries, region was more important
in 1992 than in the immediately previous election.
872
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Opportunity Structure for Women's Candidacies and Electability
. 1994. A Woman's Place Is in the House. Ann Arbor: University of
Michigan Press.
Carroll, Susan J. 1994. Women as Candidates in American Politics, 2nd ed.
Bloomington: Indiana University Press.
Darcy, R. 1992. "Electoral Barriers to Women." In Wilma Rule and Joseph
Zimmerman, eds., The Impact of Electoral Systems on Minorities and Women
in Comparative Perspective. Westport, CT: Greenwood Press.
Darcy, R., Susan Welch, and Janet Clark. 1994. Women, Elections, and Repre-
sentation, 2nd ed. Lincoln: University of Nebraska Press.
Fowler, Linda. 1993. Candidates, Congress and the American Democracy. Ann
Arbor: University of Michigan Press.
Gertzog, Irwin, and M. Michele Simard. 1981. "Women and 'Hopeless' Con-
gressional Candidacies." American Politics Quarterly 9: 449-61.
Hills, Jill. 1981. "Candidates: The Impact of Gender." Parliamentary Affairs
34: 221-28.
Lovenduski, Joni, and Pippa Norris. 1989. "Political Recruitment and the
Masculine Candidate Model in the British General Election of 1987." Eu-
ropean Journal of Political Research 17: 533-62.
, eds. 1993. Gender and Party Politics. London: Sage.
Nechemias, Carol. 1987. "Changes in the Election of Women to U.S. State
Legislative Seats." Legislative Studies Quarterly 12: 125-42.
Norris, Pippa. 1987. Politics and Sexual Equality. Boulder, CO: Rienner.
. 1993. "Britain." In Joni Lovenduski and Pippa Norris, eds., Gender
and Party Politics. London: Sage.
Pritchard, Anita. 1992. "Changes in Electoral Structures and the Success of
Women Candidates: The Case of Florida." Social Science Quarterly 73: 62-70.
Rasmussen, Jorgen. 1977. "The Role of Women in British Parliamentary Elec-
tions." Journal of Politics 39: 1044-54.
. 1983. "Women's Role in Contemporary British Politics: Impediments
to Parliamentary Candidature." Parliamentary Affairs 36: 300-315.
Rule, Wilma. 1981. "Why Women Don't Run: The Critical Contextual Factors in
Women's Legislative Recruitment." Western Political Quarterly 34: 60-77.
. 1987. "Electoral Systems, Contextual Factors, and Women's Oppor-
tunity for Election to Parliament in Twenty-Three Democracies." Western
Political Quarterly 40: 477-98.
. 1990. "Why More Women Are State Legislators: A Research Note."
Western Political Quarterly 42: 437-48.
Schlesinger, Joseph R. 1966. Political Ambition. Chicago: Rand McNally
Studlar, Donley T., Ian McAllister, and Alvaro Ascui. 1988. "Electing Women
to the British Commons: Breakout from the Beleaguered Beachhead?" Leg-
islative Studies Quarterly 13: 515-28.
873
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
Political Research Quarterly
Studlar, Donley T., and Susan Welch. 1987. "Understanding the Iron Law of
Andrarchy: The Effects of Candidate Gender on Voting for Local Office in
Scotland." Comparative Political Studies 20: 174-91.
. 1991. "Does District Magnitude Matter? Women Candidates in Lon-
don Local Elections." Western Political Quarterly 44: 457-66.
The Times. 1992. Guide to the House of Commons 1992. London: Times Books
Ltd.
Vallance, Elizabeth. 1984. "Women Candidates in the 1983 General Election."
Parliamentary Affairs 37: 301-309.
Welch, Susan, and Donley T. Studlar. 1988. "The Effects of Candidate Gen-
der on Voting for Local Office in England.' British Journal of Political Sci-
ence 18: 271-81.
. 1990. "Multi-Member Districts and the Representation of Women:
Evidence from Britain and the United States." Journal of Politics 52: 391-
412.
Received: May 17, 1995
Accepted for Publication: December 21, 1995
Political Research Quarterly, Vol. 49, No. 4 (December 1996): pp. 861-874
874
This content downloaded from 132.174.254.159 on Tue, 12 Apr 2016 01:52:45 UTC
All use subject to http://about.jstor.org/terms
