Psychology of Women Quarterly, 34 (2010), 508–520. Wiley Periodicals, Inc. Printed in the USA.
Copyright C 2010 Division 35, American Psychological Association. 0361-6843/10
ASSOCIATIONS BETWEEN FEMININITY AND WOMEN’S
POLITICAL BEHAVIOR DURING MIDLIFE
Elizabeth R. Cole and Natalie J. Sabik
University of Michigan
The contention that femininity makes women unsuited for political participation has roots in feminist theory and
political science. This study investigated whether the desirable and undesirable dimensions of femininity, corresponding
to Feminine Interpersonal Relations (FIR: warmth, nurturance, and interpersonal appeal) and Feminine Self-Doubt
(FSD: submissiveness, self-doubt, anxiety, and passivity), have independent and interactive effects on Black and White
women’s political efficacy and participation. Using questionnaires administered to alumnae of the college classes of
1967–73 in 1992 and 2008, coders assessed femininity variables at Time 1 when participants were in their 40s using
items from the California Q-Set. Political variables were assessed at Time 1 and when the participants were in their
early 60s. In general, FIR was associated with greater participation and efficacy, both directly and in interaction with
low FSD, and FSD was associated with lower efficacy scores. Specifically, at Time 1, women rated high on FIR and low
in FSD were highest on political efficacy; those high in both types of femininity scored lowest. At Time 2, among women
high in FIR, low FSD was associated with enhanced levels of participation; however, among those low in FIR, FSD was
unrelated to participation. Results are discussed in light of women’s midlife development and Black women’s gender
socialization. Recognition of the role of feminine qualities such as warmth, social skill, and compassion in political work
could encourage women endorsing feminist beliefs to act politically.
The contention that femininity makes women unprepared
or unsuited for participation in the political realm has roots
in both feminist theory and in the field of political science.
Many second-wave feminists writing in the 1970s conceptualized
femininity in terms of passivity, submissiveness, and
dependence, and viewed the socialization of young women
into these traits as a form of internalized oppression that facilitated
women’s acceptance of patriarchy (Hollows, 2000;
see also Jackman, 1994). In an early and influential paper,
Bourque and Grossholtz (1974) suggested that political science
had reached a similar conclusion about the incompatibility
of femininity and politics. They argued the field
had long overlooked the political involvement of women by
uncritically assuming that two commonly held traditional
beliefs about women—(a) that women shared the political
attitudes of their husbands and (b) that women failed
Elizabeth R. Cole and Natalie J. Sabik, Women’s Studies Department
and Psychology Department, University of Michigan.
We are grateful to Nicola Curtin, Lauren Duncan, and Abigail
Stewart for helpful feedback on the manuscript.
Address correspondence and reprint requests to: Elizabeth
R. Cole, University of Michigan Women’s Studies Department,
204 S. State Street, Ann Arbor, MI 48109-1290. E-mail:
ecole@umich.edu
to vote due to the demands associated with their family
roles—were accurate. Concisely put, political scientists assumed
that women’s deference and nurturance precluded
their political participation (see also Sigel, 1996).
In this article we take up this assumption, investigating
whether there is in fact any association between
whether and how women enact normative femininity (Cole
& Zucker, 2007) and their political behavior. We present
research testing whether two dimensions of femininity were
associated with women’s political behavior, measured when
these women were in their mid 40s and 16 years later, when
they were in their early 60s. These data were collected from
Black and White women who graduated from college during
the late 1960s and early 1970s as members of a cohort
who were among the first to have access to the new educational
and occupational opportunities won by the civil
rights and women’s movements.
Femininity and Political Behavior
Despite the fact that American women’s political participation,
particularly in electoral politics, dramatically increased
subsequent to the peak of the second wave women’s movement,
many women still view political participation as either
unfeminine or incompatible with women’s roles. For example,
Romer (1990) asked high school activists to write about
their future lives. She found that although more than 80%
508
Femininity and Political Behavior During Midlife 509
of both boys and girls anticipated continuing their activism
in adulthood, girls, but not boys, expressed concerns about
combining political work with family life. In fact, few boys
wrote about family life at all. Several studies have found that
women tend to view women who claim a political identity
as feminists as unfeminine (Henderson-King & Stewart,
1994; Rudman & Fairchild, 2007). Although this attitude
may in part be due to stereotypes about feminists (Bullock
& Fernald, 2003), it may also follow from the belief that
feminists typically display traits associated with masculinity
such as aggressiveness (Rich, 2005).
Other studies suggest, just as the second wave feminists
warned, that women’s concern with femininity might pose
a limitation for their political behavior defined broadly.
In a diary study, Hyers (2007) asked women to record
their encounters with prejudice, how they responded, and
why. Those women who described the motivation for
their response as typical gender role concerns (e.g., conflict
avoidance) were more likely to do nothing or to respond
nonassertively. Indeed, Bullock and Fernald (2003)
found that even self-identified feminists rated a profeminist
speech as more persuasive when the speaker wore
feminine clothing, styled her hair, and used cosmetics than
when the same speech was delivered by the same speaker
without feminine attire and grooming. They speculated that
respondents’ ratings reflected their desire to distance themselves
from the stereotype of feminists as unfeminine. The
authors observed that although feminine self-presentation
may increase the persuasiveness of feminist messages, this
outcome comes at a cost for the speaker herself because
feminine traits are generally devalued (Carli, 2001; Ridgeway
& Bourg, 2004).
Role congruity theory provides another line of research
suggesting that laypeople view femininity and political participation
as being at odds. Eagly and Karau (2002) theorized
that due to gender segregation in social roles, people
expect women to display communal traits of concern
for others that are associated with their traditional nurturing,
caregiving roles. However, people expect leaders
to display agentic traits more commonly associated with
men, such as dominance, control, and ambition. Because of
the incongruity between the communal traits desirable for
women and the agentic traits desirable for leaders, women
are often judged as unsuited for leadership. This theory
has found empirical support using multiple methods (e.g.,
Johnson, Murphy, Zewdie, & Reichard, 2008; Ritter & Yoder,
2004).
Dimensions of Femininity
Gender attributes are multidimensional, including grooming
and appearance, leisure activities, traits, and beliefs
about work/family roles (Mahalik et al., 2005; Twenge,
1999). Psychological theories of gender have largely focused
on personality traits. In the 1970s, feminist psychologists
argued that masculinity and femininity corresponded
to two separate and independent constructs rather
than to opposite ends of one continuum (Constantinople,
1973). These constructs have been variously termed masculinity
and femininity (operationalized by the Bem Sex
Role Inventory [BSRI]; Bem, 1974), agency and communion
(Bakan, 1966), or instrumental/assertiveness and emotional/expressiveness
(operationalized as the Personal Attributes
Questionnaire [PAQ]; Spence & Helmreich, 1978).
Generally, these conceptualizations of femininity include
traits related to the care and comfort of others, such as
warm, sympathetic, giving, cheerful, and socially skilled.
Widely used measures of these constructs (e.g., Bem, 1974;
Spence & Helmreich, 1978) were derived from the responses
of predominantly White samples; however, more
recent studies have shown that Black and White women
rate themselves similarly on the femininity subscale of
the BSRI (Cole & Zucker, 2007; De Leon, 1993; Harris,
1996).
There is evidence that trait masculinity and femininity
are in fact multidimensional. Commonly used versions
of both the BSRI and PAQ are composed of items developed
by asking respondents to rate traits in terms of
their social desirability for men and for women. Although
less often employed (McCreary & Korabik, 1994), Spence,
Helmreich, and Holahan (1979) also developed an expanded
version of the PAQ that included subscales measuring
positively valued masculinity (M+) and femininity
(F+) as well as negatively valued versions of these constructs
(M− and F−). They conceptualized F+ as including
traits relevant to interpersonal connection and emotional
expressiveness, such as kindness, helpfulness, and
warmth.
In contrast, they conceptualized F− as including both
unmitigated communion (e.g., spineless, subordinates self
to others) and passive aggressiveness (e.g., whiny, complaining).
Among women, F+ was uncorrelated with both
dimensions of F−; however, both F− dimensions were
associated with negative self-esteem, and passive aggressiveness
was also correlated with high neuroticism. Similarly,
Ricciardelli and Williams (1995) characterized negative
femininity in terms of doubts about personal competence,
describing the construct with traits such as weak,
dependent, worrying, timid, and self-critical. Their factor
analysis of a scale including 50 trait adjectives designed to
reflect both negative and positive aspects of masculinity and
femininity revealed the four theorized factors (Antill, Cunningham,
Russell, & Thompson, 1981). Among women, the
undesirable femininity factor included items such as needs
approval, anxious, and worrying, and scores on this scale
were correlated negatively with self-efficacy. Interestingly,
these two positive and negative dimensions of femininity
map onto Fiske, Cuddy, Glick, and Xu’s (2002) characterization
of stereotypes of women as including high warmth
and low competence. Additionally, they argue that this type
of mixed stereotype supports paternalistic forms of preju-
dice.
510 COLE AND SABIK
Femininity and Race
Although Black American women value and perform many
of the behaviors and traits associated with conventional
femininity (e.g., Haynes, 2000; Smith, Thompson, Raczynski,
& Hilner, 1999), there simultaneously exists within
Black communities a specific raced gender identity that
values characteristics discrepant with normative femininity,
including assertiveness, participation in the paid
labor force, community leadership (Gilkes, 2001), and
personal resilience (Beauboeuf-Lafontant, 2007; Settles,
Pratt-Hyatt, & Buchanan, 2008). Wyatt (2008) traced the
genealogy of this ideal as a survival mechanism that has endured
and adapted over time to resist the different forms
racial oppression has taken. The justification for this ideal
relies on the importance of Black women’s roles to community
service and racial advancement; for example, Harriet
Tubman and Sojourner Truth are cited as exemplars
(Beauboeuf-Lafontant, 2007). Note that although the traits
of the “strong Black woman”—perseverance, stoicism, confidence
(Beauboeuf-Lafontant, 2007)—are at odds with
the characteristics of negative femininity, they cannot be
equated with masculinity because they are valued specifically
when deployed in the communal service of families
and communities. Taking an intersectional approach that
considers how the experience of gender is shaped by race
(Cole, 2009; Greenwood, 2008), we expect that for Black
women, rejection of the traits associated with the negative
aspects of femininity may be particularly important for
political participation.
The Current Research
Our research is based on lengthy surveys, including many
open-ended items, that were administered to a sample of
Black and White alumae of a large Midwestern university in
1992, when they were in their mid to late 40s, and again in
2008, when most members of the sample were in their early
60s. Raters used the California Q-Set (CAQ; Block, 2008)
to describe participants’ 1992 responses, and these CAQ
profiles were used to create scores for desirable femininity,
which we term Feminine Interpersonal Relations (FIR—
including warmth, nurturance, and interpersonal appeal),
and undesirable femininity, which we term Feminine SelfDoubt
(FSD—including submissiveness, self-doubt, anxiety,
and passivity). We present analyses investigating the
association between these two aspects of femininity with
political participation and political efficacy, measured in
both 1992 and 2008. The unique features of this data set
allow us to explore (a) whether these associations vary by
race and (b) whether desirable and undesirable aspects of
femininity assessed at midlife are associated with political
behavior over time.
Although these data do not allow us to look at change
in femininity over time, they do provide an opportunity to
explore developmental aspects of femininity by looking at
whether the correlates of the two dimensions of femininity
vary from early to late in midlife. We know that women’s
feelings about their bodies, weight and appearance—
concepts often linked to femininity—may change with age
(Hurd, 2000; Tunaley, Walsh, & Nicolson, 1999). Women’s
scores on feminine traits may change as well. Feldman,
Biringen, and Nash (1981) found that tenderness and compassion,
factors extracted from the femininity subscale of
the BSRI and that are related to our construct of FIR, were
higher among grandparents than among men and women
at earlier stages in the parenting life cycle. Conversely,
Roberts, Helson, and Klohnen (2002) found that women’s
femininity (defined in terms of dependency and insecurity)
decreased between young adulthood and midlife. Similarly,
longitudinal research following college-educated women
(Stewart, Ostrove, & Helson, 2001; Zucker, Ostrove, &
Stewart, 2002) demonstrated that between their 30s and
60s, women increased their sense of confidence, power, and
authority; these traits appear to correspond to low scores
on FSD. If midlife is a developmental moment at which
women’s relationship to dimensions of femininity including
appearance, nurturance, and power are in fluctuation,
it is of interest to explore whether feminine traits at midlife
predict behavior later in life. Moreover, this unique data
set gives us an opportunity to explore these associations
within a racially diverse cohort that came of age during
the social movements of the late 1960s and early ’70s, social
movements that advanced the interests of women and
Black Americans. This cohort might be especially likely to
express their sense of power and authority in the political
realm (Muhlbauer, 2007).
As Spence et al. (1979) found, we expected that the
socially desirable and undesirable dimensions of femininity
would be uncorrelated. Based on the extant literature, we
hypothesized there would be no difference between Black
and White women on FIR. However, due to the emphasis
on strength, coping, and competency in Black women’s
gender socialization, we expected they would score lower
than White women on FSD. Cole and Stewart (1996) found
that Black women in this sample scored higher than White
women on political participation and political efficacy at
midlife; thus we expected to find similar differences on
these variables measured in 2008.
Because undesirable femininity, particularly the aspects
related to confidence, has been found to be associated with
low efficacy generally (Ricciardelli & Williams, 1995), we
expected that FSD would be associated with lower political
efficacy. Because self-efficacy in a particular domain
has been shown to predict behavior in that domain generally
(e.g., Holden, 1991; Stajkovic & Luthans, 1998) and
in particular for political efficacy (Zimmerman, 1989), we
similarly expected that FSD would be related to lower political
participation. Additionally, individuals who are anxious,
fearful, and concerned with their own adequacy might be
too focused on their inner lives to expend their resources on
political goals. Because of the link between the strong Black
woman ideal and community service and social change work
Femininity and Political Behavior During Midlife 511
(Beauboeuf-Lafontant, 2007; Gilkes, 2001), we expected
that for Black women, low scores on FSD would be particularly
important for political participation.
We could find no extant literature to directly support a
hypothesis about the association between FIR and political
efficacy. Hyers’s (2007) findings suggest that women
who conform to feminine norms of conflict avoidance and
social poise might steer clear of some forms of political participation
(e.g., protest). On the other hand, recruitment
into political organizations could be facilitated by membership
in social networks (Baumeister, Dale, & Muraven,
2000; Cable, Walsh, & Warland, 1998; Schussman & Soule,
2005), and the traits associated with FIR may foster the
development of such networks. Consequently, these analyses
were exploratory. However, we hypothesized that there
would be an interaction between FIR and FSD such that
women whom observers rated high on both dimensions
would be least politically engaged, both in terms of efficacy
and participation. We speculated that women whom others
perceive as both deeply responsive to the needs of others
and doubting their own competence—that is, those who
most conform to paternalistic stereotypes of women (Fiske
et al., 2002)—would be least likely to try to have a political
impact.
METHOD
Participants
Analyses presented here are based on questionnaires administered
through the mail to two samples of University
of Michigan alumnae in 1992 and 2008. In 1992, the participants
were in their mid to late 40s; in 2008, they were
in their early 60s (Mage = 61.44). One sample comprised
participants in the Women’s Life Paths Study (N = 107), a
longitudinal study initiated in 1967 by Sandra Tangri (Tangri,
1972). Because the longitudinal sample did not include
any Black women, in 1992 we recruited 64 Black alumnae
from the same era. Because substantial numbers of
Black women did not matriculate to the University until
the early 1970s, women who graduated between 1967 and
1973 were included in this sample. In both waves of data
collection, participants completed lengthy questionnaires
including items pertaining to occupational and family history,
health and life satisfaction, and political attitudes and
behavior. Thus, the questionnaire was not obviously focused
on political attitudes and behavior (see Cole & Stewart,
1996, for a more detailed description of this study and
the sample). As in any longitudinal sample, it is important
to note any patterns of systematic sample attrition. Of the
White participants, 79% (n = 85) participated in both the
1992 and the 2008 data collections. For the Black participants,
61% (n = 42) responded to the survey in both waves
of data collection. Participants who did not complete the
survey in 2008 were compared to those who did on demographic
variables measured in 1992, separately by race.
There were no significant differences on the key variables
in this article, including measures of femininity (FSD and
FIR), internal political efficacy, and political participation.
Cole and Stewart (1996) reported that in 1992, Black
and White women were comparable on mean levels of education,
income, and number of children. White women
were more likely to be married or living with a partner and
were significantly older, although these differences were
not large and were not related to participant’s scores on political
variables. In 2008, Black and White women did not
differ in terms of level of education, marital/partner status,
or yearly income.
On average, the women in both samples reported high
levels of education (78% reported having earned a postgraduate
degree) and income (28% reported a yearly household
income of between 50,000 and 100,000 dollars, and
54% reported yearly income of greater than 100,000 dollars).
Additionally, 65% reported currently living with a
partner or spouse. As was the case in 1992, White participants
in the 2008 sample were significantly older (M =
62.18, SD = 3.50) than Black participants (M = 59.99, SD =
3.70), t(239) = −4.51, p < .001). This difference was a function
of recruitment strategy, because White women were
drawn from the class of 1967 and Black women from the
classes of 1967–73.
Measures
Femininity: FSD and FIR. Two dimensions of femininity,
FSD and FIR, were assessed using observers’ ratings
of participants on the CAQ (Block, 2008). Typically, femininity
is assessed through self-report (e.g., Mahalik et al.,
2005; Parent & Moradi, 2010). Block (2008) argued that observer
ratings, or O data, are particularly valuable for assessing
characteristics that are influenced by social desirability,
and the CAQ provides a way to systematize these observations.
Given the highly prescriptive nature of gender roles
(Mahalik et al., 2005), particularly femininity (Prentice &
Carranza, 2002), the use of O data allowed us to assess femininity
without the demand characteristics of self-report and
to include both aspects of femininity, and their interaction,
in multiple regression models to predict political variables
assessed in 1992 and 2008.
Each respondent’s 1992 questionnaire was redacted so
that the scales assessing this study’s dependent measures
were removed. Many of the remaining items were openended,
including questions about mentoring, career difficulties,
views on combining work with family, experiences
in past and present relationships, feelings about motherhood,
spirituality, reflection about their life course, and activities
each year since 1981. Three to four trained coders
read each redacted file and rated the participants using the
CAQ (Block, 2008). The CAQ is designed to capture a broad
description of personality by using 100 traits, which coders
place on a continuum ranging from 1 (extremely characteristic)
to 9 (extremely uncharacteristic) in a forced normal
512 COLE AND SABIK
distribution: 5 items can be placed at the most extreme
ratings and 18 are placed in the center (relatively unimportant)
rating. For each participant, the coders’ scores on
each CAQ item were averaged to compute the participant’s
score on that CAQ item. Average interrater agreement was
.78 (SD = .09).
The Femininity/Masculinity (F/M) scale from the California
Psychological Inventory (CPI) was selected as a criterion
in the development of our CAQ measure of FSD.
The F/M subscale was designed to place individuals on a
continuum between femininity and masculinity reflecting
folk notions of these constructs (Gough & Bradley, 1996). A
series of validation studies reported by Gough and Bradley
(1996) suggested that the aspect of femininity tapped by the
F/M subscale corresponds to our conceptualization of FSD.
Observers rated people scoring high on F/M as dependent,
submissive, and high strung; spouses of high F/M scorers
described their partners as avoiding self-assertion, selfpromotion,
and attention from others. Gough and Bradley
(1996) provided correlations between the full set of CAQ
items and the CPI F/M subscale. Based on the CAQ correlates,
Gough and Bradley (1996) observed that, “women
with high scores on F/M seem to somaticize their worries
and tensions, accept dominance from others, seek reassurance
from others and worry about their own adequacy”
(p. 156), further supporting the use of this subscale to operationalize
FSD. We measured FSD as the average of the
12 CAQ items with the highest correlation with F/M. These
items tap submissiveness, self-doubt, anxiety, and passivity
(see the Appendix for items). For each participant, raters’
scores on these items were averaged to create a scale score.
The coefficient alpha for the Black sample was .63; for the
White sample, .75.
Because there is no precedent for the construct of FIR in
the CPI, this measure was developed by asking 12 women
psychologists (Ph.D.s and doctoral students, six Black and
six White) whose research concerns gender to complete
an online CAQ sort. These judges were asked to complete
a sort of “a woman you would describe as very feminine.
Imagine that this woman is of the same race as you. Rather
than thinking about a specific woman you know, try to imagine
the ideal or prototype of a feminine woman.” Following
Peterson and Klohnen (1995) we then examined the 13
most highly rated items averaged across the 12 raters (corresponding
to those rated in the two “most characteristic”
categories). Although the judges were not asked to select
items reflecting only positive aspects of femininity, all 13
of these items tapped socially valued characteristics related
to warmth, nurturance, and interpersonal appeal (Eagly &
Mladinic, 1989). Comparison of the femininity sorts made
by Black and White judges revealed a great degree of similarity,
thus we decided to make a single profile. These items
reflected (a) interpersonal warmth and concern for others
and (b) social skills. Two items were removed due to low
item-total correlations. For each participant, raters’ scores
on these 11 items were averaged to create a scale score. The
alpha for the Black sample was .76; for the White sample,
.86.
In order to maintain the discriminant validity of the two
femininity measures, CAQ item 93, “If female: Behaves in
a feminine style or manner,” was omitted from both scales.
The correlations between this item and both FSD and FIR
were significant and positive, r = .52, p < .001; r = .48, p <
.001, respectively. The two femininity measures were not
significantly correlated for either racial group—for White
women, r(N = 105) = −.13 p = .21; for Black women,
r(N = 66) = −.05 p = .69—supporting the contention that
these are two separate dimensions of femininity.
Political orientation. In 1992, respondents rated political
orientation on a 5-point scale ranging from 1 (very
conservative) to 5 (radical). The mean was 3.36 (SD = .89).
In 2008, a different response scale was used, ranging from
1 (very conservative) to 7 (very liberal) (M = 4.99, SD =
1.53). Political orientation was assessed in 1992 and 2008,
and these ratings were highly correlated, r(N = 119) =
.71, p < .001. In each multiple regression, we controlled
for political orientation as measured concurrently with the
dependent variable.
Political efficacy. Political efficacy, the sense that one
can successfully affect the political system relative to the
ability of other individuals to do so, was measured using
a 5-item scale developed by Craig and Maggiotto (1982).
Participants were asked to rate their agreement with statements,
such as “I feel like I could do as good a job in public
office as most of the politicians we elect,” from 1 (strongly
disagree) to 7 (strongly agree). The internal consistency reliability
of the measure in 1992 was .76 (Cole & Stewart,
1996). In 2008, the same measure had an alpha of .67.
Political participation. Political participation was measured
as a continuous variable based on self-reports of the
frequency with which respondents engaged in any of 17 different
political behaviors in both 1992 and 2008 (Fendrich
& Lovoy, 1988). These items tapped protest, community
activism, party and campaign work, and political communication.
At both times, respondents were asked to indicate
how often they had engaged in each behavior during the
past 2 years. Responses were summed and divided by the
number of items. The alpha for political participation in
1992 was .87 (Cole & Stewart, 1996); .88 in 2008. Women’s
participation in their 40s was correlated with participation
in their 60s, r (N = 119) = .64 p < .001. Additionally, participation
in both their 40s, r (N = 162) = .40, p < .001,
and 60s, r (N = 117) = .38 p < .001, was significantly positively
correlated with a retrospective measure of college
activism administered in 1992, indicating political behavior
was fairly consistent in this sample over time.
Femininity and Political Behavior During Midlife 513
RESULTS
The analysis strategy was as follows. First, we used t tests
to compare Black and White women’s means scores on
key variables, including FSD and FIR, as well as political
variables. Next, we ran hierarchical multiple regression
analyses with political efficacy and political participation
measured in 1992 and 2008 as the dependent variables.
The control variables, race and political orientation, were
entered at the first step. At the second step, we entered
the two femininity variables, FSD and FIR. At the third
step the interaction between the two centered femininity
variables was entered (FSD × FIR), as well as the interactions
between each centered femininity variable and race
coded as a binary variable (White = 0, Black = 1). For each
dependent variable, we also ran a model with a fourth step,
including the three-way interaction of the two centered
femininity variables with race; because this interaction did
not significantly increase the R2
for any of the dependent
variables, these models are not presented. In all regressions,
Cooks d, a statistic measuring each individual’s influence on
the model, was calculated. Any participant whose Cooks d
was over .5 (indicating disproportionate influence over the
model) was dropped from the analysis (no more than one
participant was dropped from any analysis). Additionally, a
measure of effect size for multiple regression, Cohen’s f2
,
was calculated. Statistical standards indicate that an effect
size of .02 is considered small, .15 is considered medium,
and .35 is considered large (Cohen, 1988). Finally, significant
interactions that emerged in the regressions were
graphed and simple slopes were calculated.
Racial Comparisons
The analyses comparing Black and White women’s mean
scores on measures of femininity and political variables
revealed that, as hypothesized, White women were significantly
higher than Black women in their level of FSD, and
the racial groups did not differ on their average level of FIR
(see Table 1). On average, Black women were more liberal
Table 1
Comparisons Between Black and White Women on Femininity and Political Variables
Black White
M SD M SD t df p eta2
1992 Femininity Variables
Feminine Self-Doubt 3.37 .51 3.91 .82 4.78 169 <.001 .12
Feminine Interpersonal Relations 6.51 .63 6.55 .86 .33 169 .75 .00
1992 Dependent Variables
Political Orientation 3.56 .73 3.23 .95 2.42 164 .02 .03
Political Efficacy 3.90 .78 3.61 .90 −2.21 168 .03 .03
Political Participation 1.11 .49 .92 .49 −2.39 166 .02 .03
2008 Dependent Variables
Political Orientation 4.68 1.59 5.04 1.61 −1.15 122 .25 .01
Political Efficacy 4.94 .76 4.63 .84 −1.94 121 .06 .03
Political Participation 1.18 .48 .95 .43 −2.61 122 .01 .05
than White women in their political orientation in 1992.
Although this pattern was reversed in 2008, the mean difference
did not reach significance. There were significant
racial differences on all of the political variables, showing
that the Black women in the sample were more politically
engaged. As reported in Cole and Stewart (1996), compared
to White women, Black women reported higher levels
of political efficacy and political behavior in their 40s.
This pattern persisted in 2008: on average, Black women
reported high levels of political efficacy and behavior.
Our central research questions focused on the relationship
between the two dimensions of femininity and political
efficacy and participation. We hypothesized that FSD
would be related to lower political efficacy and political
participation. Additionally, we hypothesized that, consistent
with paternalistic stereotypes of women, there would
be an interaction between FIR and FSD such that women
who were high on both would show the lowest levels of
political participation and efficacy.
Associations Between Femininity and 1992
Political Variables
Table 2 presents the results of the political variables measured
in 1992. In 1992, when participants were in their 40s,
the interaction between FIR and FSD was significantly associated
with political efficacy. Simple slopes analyses were
conducted to probe the significant interaction using values
one standard deviation above and one standard deviation
below the mean for FIR (Holmbeck, 2002). As seen in Figure
1, women rated low in FIR were moderate in terms of
political efficacy, regardless of FSD, B = −.54, ß = −.08,
SE = 0.59, p = .36. However, for those high on FIR, political
efficacy varied as a function of level of FSD; those
high on FSD were lowest on political efficacy, and those
low on FSD were highest on political efficacy, B = −2.72,
ß = −.41, SE = 0.81, p < .01. In other words, when the
women were in their 40s, those who were interpersonally
warm and compassionate felt politically efficacious if they
514 COLE AND SABIK
Table 2
Hierarchical Multiple Regression Analysis Predicting 1992 Political Variables From Femininity, Race,
and Their Interactions
Political Efficacy 1992 (N = 160) Political Participation 1992 (N = 157)
Variable B SE B ß R2 (p) R2 (p) B SE B ß R2 (p) R2 (p)
Step 1 .033 (.07) .033 (.07) .092 (.00) .092 (.00)
Political Orientation .71 .47 .12 2.19 .74 .23∗∗
Race 1.22 .85 .11 2.78 1.36 .16∗
Step 2 .055 (.07) .022 (.17) .122 (.00) .030 (.08)
Political Orientation .62 .47 .11 2.28 .74 .24∗∗
Race .63 .90 .06 2.19 1.43 .13
Feminine Interpersonal −.18 .55 −.02 1.49 .88 .13
Relations
Feminine Self-Doubt −1.07 .57 −.16 −1.23 .90 −.11
Step 3 .108 (.01) .053 (.03) .132 (.00) .010 (.62)
Political Orientation .55 .46 .09 2.37 .75 .25∗∗
Race .70 .97 .07 1.76 1.56 .10
Feminine Interpersonal −.11 .69 −.02 2.32 1.13 .20∗
Relations
Feminine Self-Doubt −1.78 .66 −.27∗∗ −1.28 1.07 −.12
Feminine Interpersonal −1.32 .52 −.22∗ −.76 .84 −.08
Relations × Feminine
Self-Doubt
Feminine Interpersonal 1.32 1.23 .10 −1.91 2.00 −.09
Relations × race
Feminine Self-Doubt × .86 1.44 .06 −1.20 2.33 −.05
race
∗
p < .05. ∗∗
p < .01.
Fig. 1. The interaction of Feminine Self-Doubt and Feminine
Interpersonal Relations predicting Black and White women’s Political
Efficacy, 1992.
were low on FSD; however, those high on FIR and high
on FSD felt least politically efficacious. These results were
independent of race.
Contrary to hypotheses, none of the femininity variables,
that is, FSD, FIR, and their interaction, accounted for a significant
proportion of the variance in political participation
measured in 1992. However, there was a significant main
effect of FSD on political efficacy in 2008, such that those
who were higher in FSD reported less political efficacy (see
Table 3). This latter finding supported our hypothesis that
FSD would be related to lower political efficacy.
Associations Between Femininity and 2008
Political Variables
Examining the pattern for 2008 political participation, when
the participants were in their 60s, there was a significant
main effect for FIR, as well as a significant interaction
between FIR and FSD. Simple slopes analyses were conducted
to probe the interaction using values one standard
deviation above and one standard deviation below the mean
for FIR. Figure 2 shows that among women high in FIR,
low FSD was associated with enhanced levels of participation,
B = −.20, ß = −.33, SE = 0.09, p < .05. In contrast,
among women who were low in FIR, FSD was not significantly
related to participation, B = .03, ß = .05, SE = 0.07,
p = .66.
Additionally, the interaction between FSD and race had
a significant effect on 2008 participation. Figure 3 revealed
that White women reported moderate levels of political
participation regardless of FSD, B = .01, ß = .02, SE =
0.06, p = .83. However, Black women high in FSD reported
the lowest level of participation, and those low in FSD the
Femininity and Political Behavior During Midlife 515
Table 3
Hierarchical Multiple Regression Analysis Predicting 2008 Political Variables From Femininity, Race,
and Their Interactions
Political Efficacy 2008 (N = 120) Political Participation 2008 (N = 119)
Variable B SE B ß R2 (p) R2 (p) B SE B ß R2 (p) R2 (p)
Step 1 .027 (.20) .027 (.20) .094 (.00) .094 (.00)
Political Orientation .00 .05 .00 −.04 .03 −.14
Race .29 .16 .16 .28 .09 .29∗∗
Step 2 .106 (.01) .079 (.01) .142 (.00) .048 (.05)
Political Orientation .00 .05 −.01 −.05 .03 −.17
Race .10 .17 .06 .26 .09 .26∗∗
Feminine Interpersonal
Relations .03 .09 .03 .17 .05 .21∗
Feminine Self-Doubt −.33 .10 −.30∗∗ −.06 .06 −.10
Step 3 .137 (.02) .031 (.26) .222 (.00) .080 (.01)
Political Orientation .00 .05 .00 −.05 .02 −.18∗
Race .07 .18 .04 .14 .10 .15
Feminine Interpersonal
Relations .07 .10 .06 .14 .05 .25∗∗
Feminine Self-Doubt −.37 .12 −.35∗∗ −.02 .06 −.03
Feminine Interpersonal
Relations × Feminine −.25 .13 −.17 −.14 .07 −.17∗
Self-Doubt
Feminine Interpersonal −.12 .23 −.05 −.11 .12 −.09
Relations × race
Feminine Self-Doubt × .01 .26 .01 −.34 .14 −.27∗
race
∗
p < .05. ∗∗
p < .01.
Fig. 2. The interaction of Feminine Self-Doubt and Feminine
Interpersonal Relations predicting Black and White women’s Political
Participation, 2008.
highest, B = −.34, ß = −.57, SE = 0.13, p < .01. Thus,
FSD was associated with lower political participation for
Black women but not for White women in our sample.
DISCUSSION
In this study, we conceptualized femininity as composed of
both desirable and undesirable elements, which correspond
to FIR (warmth, nurturance, and interpersonal appeal) and
Fig. 3. The interaction of Feminine Self-Doubt and race predicting
Black and White women’s Political Behavior, 2008.
FSD (submissiveness, self-doubt, anxiety, and passivity).
We hypothesized that these two aspects of femininity would
have both independent and interactive effects on women’s
political efficacy and behavior.
In 1992, when the women were in their mid 40s, femininity
was related to political efficacy for both Black and
White women. There was a significant interaction between
FSD and FIR such that women low in FIR were moderate
in political efficacy regardless of their level of FSD. However,
for those high in FIR, high FSD had a depressing
516 COLE AND SABIK
effect on their political efficacy, and low FSD was associated
with enhanced efficacy. Put another way, the desirable
aspects of femininity—warmth, nurturance and interpersonal
appeal—were related to feeling one can make a political
impact, but only for women who were unencumbered
by femininity’s undesirable aspects: self-doubt, anxiety, and
passiveness. In fact, women who were rated high in both
desirable and undesirable femininity reported the lowest
levels of political efficacy. This combination of traits most
closely reflects paternalistic stereotypes of women as warm
but not competent (Fiske et al., 2002). In contrast, in 1992,
when the participants were in their 40s, there was no association
between political participation and either aspect of
femininity. Contrary to our hypotheses, there were also no
race differences in the association of femininity with either
of the political variables at this time.
However, in 2008, when the women were in their early
60s, femininity was related to both political participation
and efficacy. For Black and White women alike, FSD had
a significant negative main effect on political efficacy; that
is, high FSD was related to low efficacy regardless of FIR.
By the time the women entered their 60s, the social aspects
of femininity no longer magnified the effect of self-doubt
on efficacy. This pattern suggests that aging may provide
women some freedom from the subtly oppressive aspects
of femininity that feminist theorists identified and that support
paternalistic stereotypes of women.
Moreover, when the women reached their 60s, both aspects
of femininity were related to political participation.
Women who combined high FIR with low FSD reported
the greatest political involvement, suggesting that women
must transcend self-doubt in order for their social and nurturing
traits to lead to political efficacy and participation.
This finding is particularly interesting in light of the fact
that feelings of confidence, power, and authority have been
found to increase among college-educated women between
their 30s and 50s (Stewart et al., 2001; Zucker et al., 2002).
Why did the same model account for so little of the variance
in political participation when the women were in their 40s?
Here the literature from political science may be helpful.
At points in the life course when family demands are high,
individuals participate in politics less, particularly in nonschool-related
matters (Jennings, 1979). Perhaps later in
the life course, when many women face fewer family constraints
on their time (Riley & Bowen, 2005), personality
variables may be more relevant to participation.
FSD was directly related to political participation for
Black women entering their 60s, but not for White women.
For Black women, low FSD was related to greater participation;
high FSD was related to depressed participation.
Given that Black women in the sample also were rated as
having lower levels of FSD than White women, this finding
supports the importance of Black women’s particular gender
socialization to the Black community’s political mobilization
(Gilkes, 2001). Much of the literature on the “strong
Black woman” ideal discusses the costs to Black women’s
mental and physical health of maintaining strength, or the
appearance of strength, in the face of continuing discrimination
and structural disadvantage (Beauboeuf-Lafontant,
2007; Harris-Lacewell, 2001). In contrast, the findings in
our paper suggest that to the extent the “strong Black
woman” ideal helps Black women resist socialization into
the self-doubt, anxiety and passivity that can be associated
with femininity, the ideal can be a resource for political action,
possibly an important form of resistance, at least later
in life. At the same time, fighting for other people through
political involvement is consistent with the emphasis of the
strong Black woman ideal on self-sacrifice for the benefit
of others.
These findings suggest the importance of this gendered
racial socialization for collective approaches to resisting
racial oppression. However, it should be noted that because
we controlled for political orientation, this association holds
whether Black women participate on the political right or
left. The fact that Black women’s participation was associated
with FSD in 2008, but not in 1992, is puzzling. Perhaps
the unprecedented level of support for the Barack
Obama presidential campaign among African Americans
(more than 90% rated him favorably [Bostis, 2008]) led
Black women in the sample to participate unless personality
traits such as those tapped by FSD held them back.
Unfortunately, participant age and historical era are confounded
in this study, as they are in any longitudinal study
of a single cohort.
In contrast, in their 60s, White women’s political participation
was not related to FSD. For this group, feminine
characteristics such as submissiveness, self-doubt, and
anxiety did not pose an obstacle to political participation.
Recall that White women were higher than Black women
on this trait. Perhaps by the time they reach late midlife,
many White women have learned how to take action if they
chose, regardless of any personal doubt or anxiety. However,
because we did not measure femininity in 2008, it
is also possible that, rather than compensating for higher
FSD in 2008, White women’s FSD may have decreased as
they aged.
There are several broader implications of these findings.
First, the findings support the contention that psychological
femininity is multidimensional and includes socially desirable
and undesirable elements, which are uncorrelated and
have separate and interactive effects on political participation
and efficacy. Moreover, high FIR and high FSD, and
the CAQ items that assess them, correspond to the mixed
stereotypes of women as warm but not competent theorized
by Fiske et al. (2002), stereotypes that support benevolent
sexism. Future research should further investigate the correlates
of these personality traits for women.
Second, the findings that FIR was positively associated
with political efficacy and participation at two different ages
is at odds with the literatures positing femininity as a deterrent
to political involvement. Our results suggest that
claims that femininity either relegates women’s interests to
Femininity and Political Behavior During Midlife 517
the domestic sphere, or facilitates women’s acceptance of
patriarchy, have perhaps focused on only the undesirable
aspects of femininity without defining it as such. These critiques
have also overlooked the ways that communal and
expressive aspects of femininity may play an important role
in political life, which is inherently social. A few studies
have investigated social networks as a recruitment mechanism
for political mobilization, particularly for women (Cable
et al., 1988; Cole, 1994; Crossley, 2008; Irons, 1998).
The traits associated with feminine interpersonal relations
might be expected to foster such networks.
The positive association between FIR and the political
variables suggests that studies of political participation
should consider the role that communal traits might play for
women. Arguably, this expansion may take different forms
for women on the political left and right. It is noteworthy
that in all of the models tested, political orientation was
controlled. Certainly a given political issue might be understood
in terms of communal values on both the left and
right (e.g., liberals might oppose a war based on concern for
human life, whereas conservatives might support the same
war as protecting the homeland). Researchers have found
that liberals value and make use of the moral framework
of harm/care more than conservatives (Graham, Haidt, &
Nosek, 2009; McAdams et al., 2008). Thus, it is possible
that the traits associated with feminine interpersonal relations
are consistent with moral values for women on the
left, whereas on the political right, women’s performance of
the normative feminine traits of care and nurturance might
be viewed as representing valued traditional gender roles.
Whether warmth, compassion, and care for the welfare of
others plays the same role in the political participation of
women on the political left and right remains a question for
future studies. Future research should explore the types
of behavior that FIR and FSD might predict beyond the
political realm, and whether these differ among diverse
groups of women. Considering the results in a lifespan developmental
context, it would be useful to know how these
aspects of femininity are related to women’s social project
commitments at midlife and beyond (Newton & Stewart,
2010).
Of course our study has strengths and limitations. Because
the femininity variables were assessed using observers’
ratings, these analyses minimize the possibility that
shared method variance produced spurious relationships.
The sample is one of a handful of longitudinal data sets of
college-educated women, and to our knowledge the only
one that includes an over-sample of Black women. By the
same token, our sample is highly educated and affluent,
and thus our results cannot be generalized to American
women in general. Plentiful research in political science
demonstrates that education and income are associated
with higher levels of political participation (e.g., Leighley,
1995). The sample is also distinctive by virtue of the cohort
it represents. For women who came of age during the social
movements of the late 1960s and early ’70s, political participation
may be an important expression of their identities
(Stewart & Healy, 1989) and may thus have associations
with their gender roles that cannot be generalized to other
cohorts. However, Duncan’s (2010) finding that feminist
self-labeling was a stronger predictor of attitudes towards
feminism than was the cohort suggests that this is an open
question that is ripe for future research. Finally, the analyses
were constrained by the fact that we did not have CAQ
ratings for women in 2008, which limited our ability to look
at the development of femininity and its association with
political variables over time.
Our findings suggest that femininity is not entirely the
“velvet glove” that masks the iron fist of patriarchy (Jackman,
1994). Understanding femininity as including both
desirable and undesirable elements could help clarify and
even reconcile the tension between feminist critiques of the
political implications of femininity and the value placed on
femininity by women outside the movement. This understanding
could be particularly important for the political
mobilization of women who endorse feminist beliefs but
hesitate to identify as feminists because, for many women,
the perception of feminists as “unfeminine” is an important
obstacle to feminist identification (Henderson-King &
Stewart, 1994; Reid & Purcell, 2004; Rudman & Fairchild,
2007). A feminist campaign emphasizing the important role
of feminine qualities such as warmth, social skill, and compassion
for others in political work could encourage women
who endorse feminist beliefs to put those convictions into
action in the political sphere.
Initial submission: June 2, 2009
Initial acceptance: January 26, 2010
Final acceptance: June 30, 2010
REFERENCES
Antill, J. K., Cunningham, J. D., Russell, G., & Thompson, N. L.
(1981). An Australian Sex-Role Scale. Australian Journal of
Psychology, 33, 169–183.
Bakan, D. (1966). The duality of human existence: An essay on
psychology and religion. Chicago: Rand McNally.
Baumeister, R., Dale, K., & Muraven, M. (2000). Volition and belongingness:
Social movements, volition, self-esteem, and
the need to belong. In S. Stryker, T. J. Owens, & W. White
(Eds.), Self, identity, and social movements (pp. 239–251).
Minneapolis, MN: University of Minnesota Press.
Beauboeuf-Lafontant, T. (2007). “You have to show strength”:
An exploration of gender, race, and depression. Gender &
Society, 21, 28–51.
Bem, S. L. (1974). The measurement of psychological androgyny.
Journal of Consulting and Clinical Psychology, 42, 155–
162.
Block, J. (2008). The Q-sort in character appraisal: Encoding subjective
impressions of persons quantitatively. Washington,
DC: American Psychological Association.
Bostis, D. A. (2008, October). National opinion poll. Washington,
DC: Joint Center for Political and Economic
Studies. Retrieved June 25, 2010, from http://www.
518 COLE AND SABIK
jointcenter.org/publications_recent_publications/national_
opinion_polls/2008_national_opinion_poll.
Bourque, S. C., & Grossholtz, J. (1974). Politics an unnatural practice:
Political science looks at female participation. Politics
& Society, 4, 225–266.
Bullock, H. E., & Fernald, J. L. (2003). “Feminism lite?”: Feminist
identification, speaker appearance and perceptions of
feminist ad antifeminist messengers. Psychology of Women
Quarterly, 27, 291–299.
Cable, S., Walsh, E., & Warland, R. (1988). Differential paths to
political activism: Comparisons of four mobilization processes
after the Three Mile Island accident. Social Forces,
66, 951–969.
Carli, L. (2001). Gender and social influence. Journal of Social
Issues, 57, 725–741.
Cohen, J. (1988). Statistical power analysis for the behavioral
sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Cole, E. (2009). Intersectionality and research in psychology.
American Psychologist, 64, 170–180.
Cole, E. R. (1994). A struggle that continues: Black women, community
and resistance. In C. E. Franz & A. J. Stewart (Eds.),
Women creating lives: Identities, resilience and resistance
(pp. 309–323). Boulder, CO: Westview.
Cole, E., & Stewart, A. (1996). Meanings of political participation
among Black and White women: Political identity and social
responsibility. Journal of Personality and Social Psychology,
71, 130–140.
Cole, E. R., & Zucker, A. N. (2007). Black and White women’s
perspectives on femininity. Cultural Diversity and Ethnic
Minority Psychology, 13, 1–9.
Constantinople, A. (1973). Masculinity-femininity: An exception
to a famous dictum? Psychological Bulletin, 80, 389–407.
Craig, S. C., & Maggiotto, M. A. (1982). Measuring political efficacy.
Political Methodology, 8, 85–109.
Crossley, N. (2008). Social networks and student activism: On
the politicising effect of campus connections. Sociological
Review, 56(1), 18–38.
De Leon, B. (1993). Sex role identity among college students:
A cross-cultural analysis. Hispanic Journal of Behavioral
Sciences, 15, 476–489.
Duncan, L. E. (2010). Women’s relationship to feminism: Effects
of generation and feminist self-labeling. Psychology
of Women Quarterly, 34, 498–507.
Eagly, A., & Karau, S. (2002). Role congruity theory of prejudice
toward female leaders. Psychological Review, 109, 573–
598.
Eagly, A., & Mladinic, A. (1989). Gender stereotypes and attitudes
toward women and men. Personality and Social Psychology
Bulletin, 15, 543–558.
Feldman, S., Biringen, Z., & Nash, S. (1981). Fluctuations of sexrelated
self-attributions as a function of stage of family life
cycle. Developmental Psychology, 17, 24–35.
Fendrich, J. M., & Lovoy, K. L. (1988). Back to the future: Adult
political behavior of former political activists. American Sociological
Review, 53, 780–784.
Fiske, S., Cuddy, A., Glick, P., & Xu, J. (2002). A model of (often
mixed) stereotype content: Competence and warmth respectively
follow from perceived status and competition.
Journal of Personality and Social Psychology, 82, 878–
902.
Gilkes, C. T. (2001). If it wasn’t for the women: Black women’s
experience and womanist culture in church and community.
Maryknoll, NY: Orbis Books.
Gough, H. G., & Bradley, P. (1996). CPI manual (3rd ed., 2nd
printing). Palo Alto, CA: Consulting Psychologists Press.
Graham, J., Haidt, J., & Nosek, B. (2009). Liberals and conservatives
rely on different sets of moral foundations. Journal of
Personality and Social Psychology, 96, 1029–1046.
Greenwood, R. M. (2008). Intersectional political consciousness:
Appreciation for intragroup differences and solidarity in
diverse groups. Psychology of Women Quarterly, 32, 36–
47.
Harris, A. (1996). African American and Anglo-American gender
identities: An empirical study. Journal of Black Psychology,
22, 182–194.
Harris-Lacewell, M. (2001). No place to rest: African American
political attitudes and the myth of Black women’s strength.
Women & Politics, 23, 1–33.
Haynes, F. E. (2000). Gender and family ideals: An exploratory
study of Black middle class Americans. Journal of Family
Issues, 21, 811–837.
Henderson-King, D., & Stewart, A. (1994). Women or feminists?
Assessing women’s group consciousness. Sex Roles,
31, 505–551.
Holden, G. (1991). The relationship of self-efficacy appraisals
to subsequent health related outcomes: A meta-analysis.
Social Work in Health Care, 16, 53–93.
Hollows, J. (2000). Feminism, femininity and popular culture.
Manchester, UK: Manchester University Press.
Holmbeck, G. N. (2002). Post-hoc probing of significant moderational
and mediational effects in studies of pediatric populations.
Journal of Pediatric Psychology, 27, 87–96.
Hurd, L. C. (2000). Older women’s body image and embodied
experience: An exploration. Journal of women and aging,
12, 77–97.
Hyers, L. L. (2007). Resisting prejudice every day: Exploring
women’s assertive responses to anti-Black racism, antiSemitism,
heterosexism, and sexism. Sex Roles, 56, 1–12.
Irons, J. (1998). The shaping of activist recruitment and participation:
A study of women in the Mississippi civil rights
movement. Gender & Society, 12, 692–709.
Jackman, M. (1994). The velvet glove: Paternalism and conflict in
gender, class, and race relations. Berkeley, CA: University
of California Press.
Jennings, M. K. (1979). Another look at the life cycle and political
participation. American Journal of Political Science, 23,
755–771.
Johnson, S., Murphy, S., Zewdie, S., & Reichard, R. (2008). The
strong, sensitive type: Effects of gender stereotypes and
leadership prototypes on the evaluation of male and female
leaders. Organizational Behavior and Human Decision
Processes, 106, 39–60.
Leighley, J. E. (1995). Attitudes, opportunities and incentives: A
review essay on political participation. Political Research
Quarterly, 48, 181–209.
Mahalik, J., Morray, E., Coonerty-Femiano, A., Ludlow, L., Slattery,
S., & Smiler, A. (2005). Development of the Conformity
to Feminine Norms Inventory. Sex Roles, 52, 417–435.
McAdams, D., Albaugh, M., Farber, E., Daniels, J., Logan, R., &
Olson, B. (2008). Family metaphors and moral intuitions:
How conservatives and liberals narrate their lives. Journal
of Personality and Social Psychology, 95, 978–990.
Femininity and Political Behavior During Midlife 519
McCreary, D., & Korabik, K. (1994). Examining the relationships
between the socially desirable and undesirable aspects of
agency and communion. Sex Roles, 31, 637–651.
Muhlbauer, V. (2007). The well-being and quality of life of women
over 50: A gendered-age perspective. In V. Muhlbauer & J.
C. Chrisler (Eds.), Women over 50: Psychological perspectives
(pp. 95–111). New York: Springer.
Newton, N., & Stewart, A. (2010). The middle ages: Change
in women’s personalities and social roles. Psychology of
Women Quarterly, 34, 75–84.
Parent, M., & Moradi, B. (2010). Confirmatory factor analysis of
the Conformity to Feminine Norms Inventory and development
of an abbreviated version: The CFNI-45. Psychology
of Women Quarterly, 34, 97–109.
Peterson, B., & Klohnen, E. (1995). Realization of generativity in
two samples of women at midlife. Psychology and Aging,
10, 20–29.
Prentice, D., & Carranza, E. (2002). What women should be,
shouldn’t be, are allowed to be, and don’t have to be: The
contents of prescriptive gender stereotypes. Psychology of
Women Quarterly, 26, 269–281.
Reid, A., & Purcell, N. (2004). Pathways to feminist identification.
Sex Roles, 50, 759–769.
Ricciardelli, L., & Williams, R. (1995). Desirable and undesirable
gender traits in three behavioral domains. Sex Roles, 33,
637–655.
Rich, E. (2005). Young women, feminist identities and neoliberalism.
Women’s Studies International Forum, 28, 495–
508.
Ridgeway, C., & Bourg, C. (2004). Gender as status: An expectation
states theory approach. In A. H. Eagly, A. E. Beall, &
R. J. Sternberg (Eds.), The psychology of gender (2nd ed.,
pp. 217–241). New York: Guilford.
Riley, L., & Bowen, C. (2005). The sandwich generation: Challenges
and coping strategies of multigenerational families.
The Family Journal, 13, 52–58.
Ritter, B., & Yoder, J. (2004). Gender differences in leader emergence
persist even for dominant women: An updated confirmation
of role congruity theory. Psychology of Women
Quarterly, 28, 187–193.
Roberts, B. W., Helson, R., & Klohnen, E. C. (2002). Personality
development and growth in women across 30 years: Three
perspectives. Journal of Personality, 70, 79–102.
Romer, N. (1990). Is political activism still a “masculine” endeavor?
Gender comparisons among high school political
activists. Psychology of Women Quarterly, 14, 229–243.
Rudman, L., & Fairchild, K. (2007). The F word: Is feminism
incompatible with beauty and romance? Psychology of
Women Quarterly, 31, 125–136.
Schussman, A., & Soule, A. (2005). Process and protest: Accounting
for individual protest participation. Social Forces, 84,
1083–1108.
Settles, I., Pratt-Hyatt, J., & Buchanan, N. (2008). Through the
lens of race: Black and White women’s perceptions of womanhood.
Psychology of Women Quarterly, 32, 454–468.
Sigel, R. S. (1996). Ambition and accommodation: How women
view gender relations. Chicago: University of Chicago
Press.
Smith, D. E., Thompson, J. K., Raczynski, J. M., & Hilner, J. E.
(1999). Body image among men and women in a biracial
cohort: The CARDIA study. International Journal of Eating
Disorders, 25, 71–82.
Spence, J. T., & Helmreich, R. L. (1978). Masculinity and femininity:
Their psychological dimensions, correlates and antecedents.
Austin, TX: University of Texas Press.
Spence, J. T., Helmreich, R. L., & Holahan, C. K. (1979). Negative
and positive components of psychological masculinity and
femininity and their relationships to self-reports of neurotic
and acting out behaviors. Journal of Personality and Social
Psychology, 37, 1673–1682.
Stajkovic, A., & Luthans, F. (1998). Self-efficacy and work-related
performance: A meta-analysis. Psychological Bulletin, 124,
240–261.
Stewart, A. J., & Healy, J. M. (1989). Linking individual development
and social changes. American Psychologist, 44, 30–42.
Stewart, A., Ostrove, J., & Helson, R. (2001). Middle aging in
women: Patterns of personality change from the 30s to the
50s. Journal of Adult Development, 8, 23–37.
Tangri, S. (1972). Determinants of occupational role innovation
among college women. Journal of Social Issues, 28, 177–
199.
Tunaley, J. R., Walsh, S., & Nicolson, P. (1999). “I’m not bad for
my age”: The meaning of body size and eating in the lives
of older women. Aging and Society, 19, 741–759.
Twenge, J. M. (1999). Mapping gender: the multifactoral approach
and the organization of gender-related attributes.
Psychology of Women Quarterly, 23, 485–502.
Wyatt, J. (2008). Patricia Hill Collins’s Black sexual politics and the
genealogy of the Strong Black Woman. Studies in Gender
and Sexuality, 9, 52–67.
Zimmerman, M. (1989). The relationship between political efficacy
and citizen participation: Construct validation studies.
Journal of Personality Assessment, 53, 554–566.
Zucker, A., Ostrove, J., & Stewart, A. (2002). College-educated
women’s personality development in adulthood: Perceptions
and age differences. Psychology and Aging, 17, 236–
244.
APPENDIX: ITEMS FROM THE CALIFORNIA Q-SET
USED TO CALCULATE FEMININITY SCORES
Interpersonal Relations
35. Has warmth; has the capacity for close relationships; com-
passionate.
92. Has social poise and presence; appears socially at ease.
17. Behaves in a sympathetic or considerate manner.
5. Behaves in a giving way towards others. (N.B. Regardless
of the motivation involved.)
11. Is protective of those close to her. (N.B. Placement of
this item expresses behavior ranging from overprotection
through appropriate nurturance to a laissez-faire, underprotective
manner.)
64. Is socially perceptive of a wide range of interpersonal
cues.
80. Is interested in members of the opposite sex. (N.B. At
opposite end, item implies absence of such interest.)
29. Is turned to for advice and reassurance.
520 COLE AND SABIK
84. Is cheerful. (N.B. Extreme placement toward uncharacteristic
end of continuum implies unhappiness or depres-
sion.)
88. Is personally charming.
Self-Doubt
14. Genuinely submissive; accepts domination comfortably.
10. Anxiety and tension find outlet in bodily symptoms. (N.B.
If placed high, implies bodily dysfunction; if placed low,
implies absence of autonomic arousal.)
19. Seeks reassurance from others.
72. Concerned with own adequacy as a person, either at conscious
or unconscious levels. (N.B. A clinical judgment is
required here.)
30. Gives up and withdraws where possible in the face of
frustration and adversity. (N.B. If placed high, implies
generally defeatist; if placed low, implies counteractive.)
59. Is concerned with own body and the adequacy of its physiological
functioning.
45. Has a brittle ego-defense system; has a small reserve of
integration; would be disorganized and maladaptive when
under stress or trauma.
47. Has a readiness to feel guilt. (N.B. Regardless of whether
verbalized or not.)
68. Is basically anxious.
40. Is vulnerable to real or fancied threat, generally fearful.
42. Reluctant to commit self to any definite course of action;
tends to delay or avoid action.
78. Feels cheated and victimized by life; self-pitying.
Note. Each number refers to item number in the California Q-Set.