Examining The Effect Of Feminist Self-Labeling And
Feminist Perspectives On Young Adults' Self-Efficacy
CHAPTER 1
INTRODUCTION
Recently, there have been several campaigns and initiatives to
increase awareness of feminist issues and educate people about what the “f
word” really means (e.g., Ban Bossy, LikeAGirl, HeForShe, 2015). While
women have been at the forefront of each wave of the feminist movement,
more men are voicing their support of feminism. Men and boys are even the
focus of the United Nations' HeForShe campaign, encouraging support and
activism from males in order to end gender inequality across the world.
Because celebrities and notable figures are often used as figureheads for
such campaigns, young adults may be more likely to be influenced by such
efforts (Austin, Vord, Pinkleton & Epstein, 2008; Jackson, 2005). As more
celebrities have been publicly identifying as feminists, more young adults
are becoming aware of feminist issues and what it means to be a feminist.
And as individuals who have come of age during “third wave” feminism
(see Chapter 2) continue to develop in their understanding of feminism, it is
also important to understand how changing views on feminism impact
feminist perspectives. And because identifying as a feminist has potential
mental health benefits related to self-esteem and self-efficacy (Eisele &
Stake, 2008; McNamara & Rickard, 1989), it is an important area of inquiry
for counselors and counselor educators. In fact, Eisele and Stake (2008)
suggested that feminist self-labeling may be more strongly related to
positive mental health than espousing feminist perspectives without
identifying as a feminist. Researchers have suggested that feminist
identification made women more likely to engage in activism, and bisexual
and lesbian women who identified as feminists exhibited more self-
acceptance (Szymanski, 2004). By better understanding the relationship
between self-efficacy, feminist identification, and feminist perspectives,
counselors and counselor educators can better support their supervisees
and/or clients.
Problem Statement
In general, women’s gender consciousness (i.e., the understanding that
one’s gender affects their experiences in the world) and collective action
efforts (i.e., unified efforts to improve a group’s position and achieve shared
goals) for women have typically been weaker than that of other groups
despite generally being more aware of gender issues than men (Aronson,
2003). Although several studies have investigated how such issues affect
White college-aged women, perspectives from ethnic and racial minorities
and men in the same age group have not been well represented. Previous
research has suggested that status as a minority may make individuals more
likely to support other minority groups (Hunter & Sellers, 1998; Kelly &
Breinlinger, 1995; Wiley et al., 2012), yet some researchers believe that this
may not be the case with feminism, as feminism may only align with the
issues and experiences of White, middle class women (Aronson, 2003;
Hunter & Sellers, 1998; Williams & Wittig, 1997; Zucker, 2004). Although
previous waves of feminism have included more singular views and
identities for group members, feminism today consists of more varied
opinions about what it means to be a feminist and how identification as a
feminist is expressed (Heywood & Drake, 1997; Aronson, 2003). Such
variance in beliefs can be confusing for individuals and may contribute to
them not identifying as feminist, despite sharing similar perspectives (Kelly,
2015). Researchers have suggested that feminist perspectives are linkedwith
higher perceived physical attractiveness, a more positive body image
(Kinsaul, Curtin, Bazzini, & Martz, 2014), and improvement in coping with
societal pressures (Rubin, Nemeroff, & Russo, 2004). Because the most
recent wave of feminism (i.e., third wave) looks quite different from how
society has previously defined what it means to be a feminist, becoming
more familiar with how young adults currently view feminism and how they
are incorporating the label into their identities is important for counselors
and counselor educators in their practice. Understanding how students or
clients incorporate the feminist label into their identity can potentially
provide a clearer direction for work with students or clients and encourage a
more trusting, collaborative relationship.
Social Significance
While men and women frequently report agreeing with feminist ideas
and tenets, they rarely self-label as feminists (McCabe, 2005; Roy, Weibust,
& Miller, 2007; Williams & Wittig, 1997; Zucker, 2004). When
investigating the reasons behind the hesitation to self-label as a feminist, one
of the most frequently expressed reasons from both men and women is the
fear that they will be perceived negatively and inaccurately by others
(Anderson, 2009; McCabe, 2005; Roy, Weibust, & Miller, 2007; Williams
& Wittig, 1997; Zucker, 2004). For example, feminist women are thought of
as being more intelligent, confident, and productive, they are also thought of
as being lesbians, less attractive, and volatile (McCabe, 2005; Roy, Weibust
& Miller, 2007; Suter & Toller, 2006; Twenge & Zucker, 1999). Feminist
men are often characterized as being gay, less masculine, and having more
stereotypically feminine qualities (e.g., emotional, submissive, physically
weaker) (McCabe, 2005; Roy, Weibust, & Miller, 2007; Twenge & Zucker,
1999). Individuals may also decline to identify as feminists due to a lack of
education about feminism. Kelly (2015) found that many individuals who
did not identify as feminists said they did not know about feminism,
although they supported feminist perspectives and beliefs. Despite the
potential negative consequences of self-labeling, research has suggested that
having positive perceptions of feminists and feeling connected to women
may buffer the fear of negative stereotypes and increase the likelihood that
someone will self-identify as a feminist (Myaskovsky & Wittig, 1997; Roy,
Weibust, & Miller, 2007; Wiley et al., 2012). Even brief exposure to
positive portrayals of feminists may encourage positive perceptions of
feminism, as well as increase the desire to participate in collective action for
women (Roy, Weibust, & Miller, 2007; Wiley et al., 2012). Additionally,
research has found that the act of self-identifying with a group or self-
labeling may increase the likelihood of activism on behalf of that group
(Leaper & Arias, 2011; Wiley et al., 2012; Yoder, Tobias, & Snell, 2011;
Zucker, 2004), which ultimately encourages social change. And while
identification as a feminist may encourage positive mental health (e.g., self-
efficacy), identifying as a feminist may not be as important as an individual
adopting feminist attitudes due to the negative stereotypes about feminists
(Eisele & Stake, 2008; Twenge & Zucker, 1999). For example, even
individuals who only privately identified as feminists (i.e., they do not label
themselves as feminists to others) have still reported being supportive of
feminist perspectives and beliefs (Kelly, 2015). This suggests that although
they have some concern over being perceived negatively by others, they are
not completely stigmatized by the label and therefore may be at a tipping
point in their feminist identity. In fact, Kelly (2015) found that feminist self-
labeling was more related to engagement in activism. However, feminist
self-labeling has been shown to mediate the relationship between feminist
perspectives and self-efficacy (Eisele & Stake, 2008). Higher selfefficacy,
defined as an individual’s belief in their ability to achieve a desired
outcome, has been linked to increased health, higher levels of achievement,
and better social skills (Bandura, 2002). Therefore, further investigation is
needed into the relationship among feminist self-labeling, feminist
perspectives, and self-efficacy.
By nature, feminism encourages deeper examination of social
imperatives (e.g., desire to form and to belong to groups) and societal norms
(i.e., rules used to define acceptable behavior in a group) (Ruben et al.,
2004). This may be particularly important for women, as such evaluation
may act as a buffer for certain mental health issues (e.g., disordered eating,
negative body image) (Rubin et al., 2004). However, self-efficacy has been
shown to be an even stronger moderator of such issues in women compared
to feminism and feminist beliefs (Kinsaul et al., 2014). Therefore, self-
efficacy may be the most vital component necessary for improving and
maintaining women’s mental health.
Professional Significance
Researchers identified a positive relationship between feminist
perspectives and self-efficacy (Eisele & Stake, 2008). This may be partially
explained by the positive relationship between nontraditional gender role
attitudes and self-esteem (Szymanski, 2004), which are components of
feminist perspectives and self-efficacy. If an individual feels more
empowered about an issue, they may feel more encouraged and more
capable (i.e., self-efficacy) to enact change on behalf of that issue (Eisele &
Stake, 2008; Zimmerman, 1995). This may illuminate important knowledge
regarding the relationship between self-efficacy and advocacy in general.
While advocacy is not a focus of this study, understanding how self-efficacy
impacts advocacy can provide counselors and counselor educators insight
into how this relationship might affect clients, students, and supervisees in
their desire to engage in advocacy for any group or issue. Additionally, the
findings may provide implications for how counselor educators may
structure their teachings about feminism to make them seem more relevant
to students of color and men. For example, if students of color and men
significantly vary from White students and women in their understanding
and support of feminist perspectives and the feminist label, counselor
educators may focus more on making feminism seem more relevant to their
lives. As previous researchers have suggested, seeing feminism as relevant
to one’s own life is a crucial piece to supporting feminist beliefs and
identifying as a feminist (Eisele & Stake, 2008; Hunter & Sellers, 1998;
Wiley et al., 2012).
For counselors and counselor educators, understanding the
relationship among such factors will not only help them better understand
the worldview of clients who do not ascribe to traditional gender roles, but it
will also encourage self-reflection on the impact that gender stereotyping
and sexism can have on interactions with and treatment of clients and
counselors-in-training (Goodman et al., 2004). By being unaware of how
gender stereotyping influences clinical decision making, counselors risk
providing biased treatment to clients based on their own values and
unintentionally supporting traditional gender concepts (Crethar, Rivera, &
Nash, 2008; DeVoe, 1990; Good, Gilbert, & Scher, 1990). This imposition
of values is addressed in the ACA Code of Ethics (2014), which states that
counselors must be aware of their own values, resist imposing those values,
and engage in additional training in areas that put them at risk of imposing
their values onto those with whom they work (A.7.b). Counselors also have
a responsibility to serve as advocates for their clients when necessary
(A.7.b), and counselor educators must infuse multicultural issues into all
coursework (F.7.b.). Furthermore, DeVoe (1990) found that participating in
advocacy efforts and feminist consciousness raising may make counselors
more aware of feminist issues, resulting in increased insight into power
differentials between men and women and awareness of how sexist values
can negatively impact relationships.
Theoretical Foundation
Feminist theory is based on the following principles: (a) problems
originate in political and social contexts; (b) commitment to social change is
necessary; (c) acknowledging different ways of knowing gives voice to
women; (d) an egalitarian relationship is central to the therapeutic
relationship; and (e) political and social inequity negatively affect all people
(Corey, 2009). Feminist theory shares many common threads with
multicultural counseling theories (Crethar, Rivera, & Nash, 2008; Goodman
et al., 2004) that are currently taught in counselor education programs. And
because multicultural approaches are highly valued in today’s counselor
education programs, understanding feminist theory as well seems like a
logical next step for counselor educators. Feminist theory also lends itself to
this study because of its focus on empowerment and facilitating
consciousness raising (Corey, 2009).
Social cognitive theory can also be used to help understand the
constructs being examined in this study, particularly with the college-aged
population. Social cognitive theory (SCT) posits that individuals’ beliefs in
their abilities to influence the environment shape their actions in order to
produce desired outcomes (Stajkovic & Luthans, 1998). As societal views
on sexual mores, family structure, and gender roles continue to evolve,
stereotypes continue to be influenced primarily by culture, not by inherent
biological differences between men and women (Khajehpour, Ghazvini,
Memari, & Rhamani, 2011). Khajehpour et al. (2011) elaborate on this idea
by suggesting that modeling is the most powerful means of transmitting
cultural values, attitudes and behaviors, and thought patterns across
generations. While modeling can occur through direct observation of others,
media may also serve as a source from which individuals model behavior.
Because young adults tend to be more susceptible to the media’s influence
on their behavior and beliefs (Austin, Vord, Pinkleton & Epstein, 2008;
Jackson, 2005), using a social cognitive theory lens to explain the potential
power of this influence on feminist identification may be useful. Further,
tenets of social cognitive theory may help explain why individuals choose to
identify or not identify as feminists (e.g., learning as a cognitive process in a
social context, vicarious reinforcement). Social cognitive theory may also
help to explain the relationship between self-efficacy and advocacy, as those
with higher levels of self-efficacy may persist with action despite
unfavorable circumstances as long as they believe their efforts will produce
the expected results (Stajkovic & Luthans, 1998). Social cognitive theory
also helps to explain how observed behavior (i.e., modeling) can influence
values, attitudes, and thoughts, thereby affecting stereotypes and regulation
of gender roles that are typically associated with the feminist label
(Khajehpour et al., 2011). Thus, incorporating feminist theory and social
cognitive theory (SCT) provides both a political/social lens and a learned
behavior (i.e., modeling) lens through which to view the impact of feminist
labeling and feminist perspectives on self-efficacy in undergraduate and
graduate students.
Purpose of Study
The purpose of the study was to understand the relationship between
feminism and self-efficacy in college-aged students, by examining the
differential relationship (i.e., discrepancies between the relationship between
feminist self-labeling and self-efficacy, and between feminist perspectives
and self-efficacy) of feminist self-identification and feminist perspectives to
self-efficacy. As such, the study examined the relationships among variables
of interest (i.e., feminist attitudes, feminist perspectives, self-efficacy) and
differences among demographics (i.e., gender, race) for each variable of
interest. One potential implication for understanding the relationship among
feminist identification, self-efficacy, and feminist perspectives might be that
counselors and counselor educators gain more awareness of the impact these
constructs have on young adult clients, supervisees, and/or students.
Additionally, results might also encourage self-reflection in counselors and
counselor educators in order to examine how their own beliefs on feminism
and women may impact their work with clients, students, and/or supervisees.
Research Questions and Hypotheses
This study explored the relationships among demographic factors,
feminist perspectives, feminist identification, and self-efficacy for
undergraduate and graduate students enrolled at a large four year University
in the Southeast. The research questions and null hypotheses are presented
below.
Research Question 1
What relationship exists among feminist self-identification, feminist
perspectives, and self-efficacy? Specifically, can feminist self-identification,
as measured by the SelfIdentification as a Feminist Scale (SIF; Szymanski,
2004), and feminist perspectives, as measured by the Feminist Perspectives
Scale – Short Form (FPS3; Henley, Spalding, &
Kosta, 2000) predict self-efficacy, as measured by the General Self-Efficacy
Scale (GSE; Schwarzer & Jerusalem, 1995) of young adults?
Hypothesis 1A. Feminist identification ratings and feminist
perspectives ratings are positively correlated with self-efficacy ratings.
Hypothesis 1B. Higher ratings for feminist self-identification and
feminist perspectives will predict higher self-efficacy for young adults.
Research Question 2
What differences exist among various races (e.g., White, African
American, Hispanic) and gender (i.e., male, female), as measured by the
demographics questionnaire, between feminist self-identification, as
measured by the Self-Identification as a Feminist Scale (SFI; Szymanski,
2004), feminist perspectives, as measured by the Feminist Perspectives Scale
– Short Form (FPS3; Henley, Spalding, & Kosta, 2000), and self-efficacy, as
measured by the General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem,
1995)?
Hypothesis 2A. Female participants will have higher ratings than
male participants for feminist identification, feminist perspectives, and self-
efficacy.
Hypothesis 2B. White participants will have higher ratings than non-
White participants for feminist identification, feminist perspectives, and self-
efficacy.
Research Design
The current study utilized data collected from undergraduate and
graduate students at the University of South Carolina in Columbia, South
Carolina. This study consisted of a quantitative, correlational survey
research design that examined relationships among demographic factors
(i.e., gender, race), feminist identification, feminist perspectives, and self-
efficacy. Correlational research allows researchers to assess the relationship
between variables without manipulation, while also identifying strength and
direction of the relationship (Smith & Davis, 2007). However, correlational
research design does not allow the research to determine cause and effect
relationships among variables (Smith & Davis, 2007). Prior to beginning the
study, I obtained approval from the University of South Carolina’s
Institutional Review Board (IRB). Once I received approval, I began data
collection.
The current study included undergraduate and graduate students who
volunteered to participate. Students who agreed to participate in the study
completed either a paper version or an online version of the assessment
(Survey Monkey). Students began the survey by reading the study
information form (see Appendix B) and were then prompted to provide
consent to participate. If students chose not to continue with the assessments,
or wished to discontinue the survey at any time, they were allowed to do so
without penalty. Participants then completed: (a) a researcher designed
demographic form; (b) a scale assessing feminist self-identification; (c) a
scale assessing feminist perspectives; (d) a scale assessing self-efficacy.
Participants’ demographics are also included in subsequent chapters. A more
detailed discussion of the methodology for this study is provided in Chapter
3 of this paper.
Methodology
Prior to beginning this study, I received approval from the University’s
Institutional Review Board (IRB). Data was collected in accordance with
IRB guidelines.
Participants
Participants for the study were enrolled as undergraduate or graduate
students at the University of South Carolina. There are currently 24,864
undergraduate students enrolled at the University, with this population being
46% male and 54% female. There are 8,108 graduate students enrolled at
USC, comprised of 59% females and 41% males. Minorities comprise
21.6% of the undergraduate student population and 31.6% of the graduate
student population. I contacted programs such as University 101 (containing
approximately 4,000 students), FemCo (a feminist student organization), the
Counselor Education and Supervision undergraduate minor program
(containing approximately 250 students) and Education Specialist (EdS)
program, National Pan-Hellenic Council (NPHC; historically African
American or multicultural fraternities and sororities), and large
undergraduate introductory courses (e.g., Psychology, Technology, Public
Health) in order to recruit participants. I engaged in both active (e.g., face-
to-face) and passive (e.g., email, word of mouth) recruitment strategies.
Upon IRB approval, I obtained permission from course instructors and
program coordinators to briefly speak with their students about participating
in the study (i.e., active recruitment). Additionally, I sent emails about the
study to professors who taught large undergraduate classes and requested
they send study information to their students via email or Blackboard (i.e.,
passive recruitment). Yancey, Ortega, and Kumanyika (2006) noted that
active recruitment strategies are more effective with culturally diverse
participants, hence my intention to partner with programs for minority
students (i.e., NPHC, TRIO programs). However, I received few responses
in my attempts to contact organizations or groups which contained members
of color. Therefore, I did not get the opportunity to utilize active recruitment
strategies with some of these groups.
In order to be included in the study, participants had to be at least 18
years of age and enrolled as an undergraduate or graduate student at the
University of South Carolina. Nonprobability sampling (specifically,
convenience sampling) was utilized to obtain participants for this study.
Convenience sampling is less expensive, has fewer timeconstraints, and
participants can be recruited with relative ease (Rubin & Babbie, 2011).
Despite its benefits, this sampling strategy may lead to an inadequate
representation of various groups in a sample (Rubin & Babbie, 2011). This
limitation of the study will be further discussed in Chapter 5. However, since
little is known about the relationship between feminism and self-efficacy for
men and minorities, convenience sampling may provide insight into whether
or not a problem exists in a biased sample. Because convenience samples are
typically already biased (Rubin & Babbie, 2011), uncovering the
relationship between feminism and self-efficacy for men and minorities in a
biased sample may provide valuable information into how to proceed in
future studies with these two groups (Rubin & Babbie, 2011). In other
words, if no significant differences exist for men and minorities on the
outcome variables compared to women and non-minority participants in the
biased sample, it may be unlikely that significant differences would exist in
an unbiased sample (Rubin & Babbie, 2011).
I conducted an a priori analysis using G*Power 3.0.10 (Faul,
Erdfelder, Lang, & Butchnew, 2008) to determine sample size and adequate
power with each of the anticipated analyses. A priori power analyses helped
determine the sample size necessary for adequate power (Balkin & Sheperis,
2011; Cohen, 1992), with larger sample sizes leading to less likelihood of a
type II error, higher statistical power, and larger effects (Balkin & Sheperis,
2011). The a priori analysis conducted for the current dissertation utilized an
alpha level of .05, moderate effect size of .06 (Cohen, 1992), and a
recommended power of .80 (Cohen, 1992). G*Power indicated a sample of
107 in order to achieve adequate power for research question one. The
power analysis conducted for research question two indicated a sample of
265 participants for adequate power. Because a larger sample size (n=265)
was indicated for research question two, the goal is to recruit at least 300
participants in order to avoid committing a Type II error. A Type II error
occurs when the researcher fails to reject a false null hypothesis (Rubin &
Babbie, 2011), meaning that the researcher reports finding no significant
differences between groups when such differences may actually exist.
Instruments
The following instruments were administered to study participants: (a)
a researcher designed demographic form; (b) Feminist Perspectives Scale-
Short Form (FPS3; Henley, Spalding, & Kosta, 2000); (c) General Self-
Efficacy Scale (GSE; Schwarzer & Jerusalem, 1995); and (d) Self-
Identification as a Feminist Scale (SIF; Szymanski, 2004). In order to
determine the level of reliability of each measure, a reliability analysis was
performed for each instrument to determine Cronbach’s alpha (α).
Data for the each instrument’s reliability in this study is discussed in Chapter
4 of this dissertation. All instruments were completed by participants online
using Survey Monkey, and results were transferred to SPSS for analysis.
Demographics questionnaire. The researcher-developed
demographics questionnaire was collected basic demographic information
from study participants. The form included questions about gender, age,
years of education, year in school (e.g., freshman, sophomore, first year
Master’s), and race. The demographics questionnaire was administered at
the end of the survey. See Appendix C for a copy of the demographic form.
Feminist Perspectives Scale – Short Form (FSP3). The Feminist
Perspectives Scale – Short Form (FSP3; Henley, Spalding, & Kosta, 2000) is
a 36-item scale which assessed feminist attitudes and feminist behavior. Of
the 36 items on the instrument, 30 items measure feminist attitudes and 6
items measure feminist behavior. The 30
attitudinal items are comprised of 6 subscales: (a) Conservative; (b)
Liberal Feminist; (c) Radical Feminist; (d) Social Feminist; (e) Cultural
Feminist; (f) Woman of Color/Womanist). Responses were totaled for each
subscale to obtain a total attitudinal score for each of the six subscales.
These scores were then summed together to obtain a total score for feminist
attitudes (i.e., Femscore3). Responses for the behavioral items were summed
separately in order to produce a total score for feminist behavior (i.e.,
Fembehave3). The FSP3 has shown to have a high internal consistency for
Femscore (α = .85), although some of the subscales have shown alpha
reliability ≤ .70. However, Henley, Spalding, & Kosta (2000) suggested
using a larger sample size (a sample of 209 was used in the development of
the short version of this instrument) in order to increase reliability for the
subscales.
General Self-Efficacy Scale (GSE). The General Self-Efficacy Scale
(GSE; Scwarzer & Jerusalem, 1995) measured beliefs about general self-
efficacy. The GSE is a 10 item scale which assesses one’s belief in their
ability to respond to difficult situations and cope with associated obstacles.
The 4 point Likert scale measured the extent to which each item applied to
the participant, ranging from not at all true (1) to exactly true (4). Participant
total scores can range from 10 to 40, with higher scores indicating higher
general self-efficacy. The GSE has shown good internal consistency, ranging
from .75 to .91 (Scholz, Dona, Sud, & Schwarzer, 2002).
Self-Identification as a Feminist Scale (SIF). The Self-Identification
as a Feminist Scale (SIF; Szymanski, 2004) is a 4 item scale which assessed
explicit feminist identity and support of the goals and values of the feminist
movement. Participants rated items on a 5 point Likert scale based on their
level of agreement or disagreement with each item. A total SIF score was
obtained by summing all 4 items, with higher scores indicating stronger
identification as a feminist. Szymanski (2004) found an alpha reliability
of .93 for the SIF.
Data Analyses
I conducted a preliminary analysis of the data in order to identify any
outliers, missing data, and violations of assumptions. I used two statistical
analyses to explore the two research questions for this study. I utilized a
multiple regression to examine the relationships among the constructs of
feminist self-identification, feminist perspectives, and self-efficacy (research
question one) for all participants. For research question one, feminist self-
identification (as measured by the SIF total score) and feminist perspectives
(as measured by Femscore3 total score and Fembehave3 total score) served
as the independent variables, while self-efficacy (as measured by the GSE
total score) served as the dependent variable. I also tested for any violations
of the assumptions of normality, multicollinearity, and singularity. I utilized
a two-way factorial MANOVA to examine what differences exist for various
races (e.g., White, African American, Hispanic) and gender (i.e., male,
female) for feminist self-identification, feminist perspectives, and
selfefficacy (research question two). For research question two, the
independent variables were race and gender, while the dependent variables
were self-identification as a feminist (as measured by the SIF total score),
feminist perspectives (as measured by Femscore3 total score and
Fembehave3 total score), and self-efficacy (as measured by the total GSE
score). The Statistical Package for Social Sciences (SPSS), version 23.0, was
utilized for all data analyses.
In order to ensure protection of participants’ rights, data were only be
reported in aggregate form, with no identifying information connecting
participants to their responses. All data were contained on a password
protected laptop, and I was the only person with access to the data.
Definition of Terms
Following, I operationally defined terms or phrases for the purposes
of the current study:
Feminist: Researchers have been unable to agree on a singular
definition for what it means to be a feminist (Yoder, Tobias, & Snell, 2011).
While there are many interpretations of the term feminist, the term is largely
viewed as a combination of one’s willingness to self-label as a feminist,
espoused beliefs, and the connection between selflabeling and the
endorsement of feminist beliefs (Yoder, Tobias, & Snell, 2011). Some
researchers have also highlighted the importance of understanding the label
of feminist as being both an individual and collective identity (Kelly, 2015).
And although the term feminist is typically connected to cohesive political
ideas, some scholars underscore the connection of the feminist label to social
movements and contexts (Kelly, 2015; Reger, 2012). The term feminist, for
the purposes of this study, is defined as “a person who believes in the social,
political, and economic equality of the sexes” (Adichie, 2012).
Feminist self-labeling: Feminist self-labeling is the act of identifying as
a feminist. More specifically, self-labeling as a feminist is “a binary choice
that either links, or does not link, a woman to feminists as a social group”
(Yoder, Tobias, and Snell, 2011). This label may be adopted publicly (i.e.,
proclaiming to be feminist to others) or privately (i.e., considering oneself to
be a feminist without identifying as a feminist to other people) (Kelly, 2015;
Leaper & Arias, 2011; Myaskovsky & Witting, 1997). However, some
scholars view feminist self-labeling as part of a feminist identity continuum
(Aronson, 2003; Williams & Wittig, 1997; Zucker, 2004), which may be
more applicable to individuals who came of age during “third wave”
feminism (Kelly, 2015). As such, this study includes a measure that assesses
for feminist identity and self-labeling on a continuum (see Chapter 3).
Feminism: For the purposes of this study, feminism is defined as a
political and social movement focused on political, social, and economic
equality between men and women (Kelly, 2015). Reger (2012) further
explains the complexity of the term by saying that feminism can
simultaneously be both “everywhere” (i.e., influential on a person’s
worldviews, culture, and social norms) and “nowhere” (i.e., explicit feminist
activism is limited or unseen).
Self-efficacy: Self-efficacy one’s belief in their capacity to achieve a
desired outcome (Bandura, 1997). Self-efficacy is a component of self-
esteem that influences an individual’s perception of his or her ability to
obtain expected results (Eisele & Stake, 2008). An important aspect of social
cognitive theory, Bandura (1997) explains that selfefficacy affects an
individual’s feelings (e.g., anxiety, depression), thoughts (e.g., motivation,
decision making), and actions (e.g., effort, recovery from setbacks).
Additionally, Bandura (1977) suggests the following four factors
significantly impact one’s self-efficacy: (a) mastery experiences; (b) social
modeling; (c) social persuasion; and (d) physiological factors.
Limitations
One limitation of this study is that it is unknown if students who
choose to participate in the study are similar to those who do not choose to
participate on their identification as feminists, self-efficacy levels,
perspectives on feminism, and intentions to engage in activism. For
example, participants who chose to engage in this study may have done so
because they already have strong, polarized feelings or views (positive or
negative) towards feminism, which could skew data. Nonprobability
sampling (specifically, convenience sampling) was utilized to obtain
participants for this study. Although convenience sampling is less expensive
and has fewer time-constraints than other sampling methods, it may lead to
an inadequate representation of various groups in my sample (Rubin &
Babbie, 2011). However, the researcher can attempt to improve convenience
sampling by making efforts to control and to assess the representative nature
of the survey sample (Henry, 1990). For example, I recruited participants
from majors who are not typically included in research on feminism and
self-efficacy (e.g., Public Health, Exercise Science, Technology).
Additionally, I compared sample demographics to those of the USC student
population to assess representativeness (see Chapter 3).
This study was also delimited to students who were in undergraduate
or graduate study at one four year institution in the Southeast. This limitation
could also affect both the internal and external validity of the study, as
geographic location may have impacted participants’ views on feminism and
feminist perspectives. Additionally, using one university in the Southeast
affects the generalizability of results to the general population of young
adults (Rubin & Babbie, 2011).
An additional limitation was the norming populations for the
instruments utilized in this study. For example, the Feminist Perspectives
Scale (FPS3) was normed on predominantly White and Asian
undergraduates. Therefore, results may vary for members of other ethnic
groups and for graduate students. The General Self-Efficacy Scale (GSE)
was originally normed on German citizens; however, it has since been
normed on individuals from 25 different countries (including the United
States). The Self-Identification as a Feminist Scale (SIF) was normed on
predominantly White, gay or bisexual undergraduate women and therefore
may not reflect similar outcomes for students of color, males, or students
who do not identify as gay or bisexual.
Summary
Despite the potential positive benefits of identifying as a feminist,
negative stereotypes and beliefs about feminism continue to discourage men
and women from selflabeling. By examining the relationships among self-
identification as a feminist, feminist perspectives, and self-efficacy, findings
from this study may aid counselors and counselor educators in better
supporting and understanding how to work with young adult students or
clients. And as feminism continues to become a more popular issue or
identity being adopted, counselors and counselor educators have an ethical
responsibility to better understand its impact on clients, supervisees, and
students, as well as how their own beliefs about women and feminism can
affect their work with others. Subsequent chapters review current literature
on feminist labeling, feminist perspectives, and selfefficacy (Chapter 2),
discuss the methodology of the current study (Chapter 3), provide results
from the current study (Chapter 4), and discuss conclusions and suggestions
for future research on the proposed topic (Chapter 5).
CHAPTER 2
LITERATURE REVIEW
In order to find relevant articles for this topic, I utilized the Encore
search engine provided by The Thomas Cooper Library at the University of
South Carolina. I used “feminism,” “feminist label,” “feminist
identification,” “feminist self-labeling,” “feminist beliefs,” “feminist
perspectives,” “feminist attitudes,” “self-esteem,” “self-efficacy.” Upon
researching these terms, I noticed the Sex Roles and Psychology of Women
Quarterly were two journals that published the most articles related to my
topic. I then conducted a search of my terms within these specific journals.
In reading journal articles, I also used the references listed in previous
research to locate possible articles for the current study.
While women have been at the forefront of each wave of the feminist
movement, men have become the focus of campaigns like the United
Nation’s HeForShe (2015), which encourages support and activism from
males in order to end gender inequality across the world. And while men and
women frequently report agreeing with feminist ideas and tenets, they rarely
self-label as feminists due to fear that they will be perceived negatively and
inaccurately by others (Anderson, 2009; McCabe, 2005; Roy, Weibust, &
Miller, 2007; Williams & Wittig, 1997; Zucker, 2004). Because identifying
as a feminist can impact self-esteem, self-efficacy, and desire to engage in
advocacy (Elise & Stake, 2008; Leaper & Arias, 2011; Roy, Weibust, &
Miller, 2007; Twenge & Zucker, 1999; Wiley et al., 2012; Yoder, Tobias, &
Snell, 2011) it is an important area of inquiry for counselors and counselor
educators. For example, empowerment is an important component of
feminism, and individuals who feel more empowered about an issue may
feel more encouraged and more capable to enact change on behalf of that
issue (Eisele & Stake, 2008; Zimmerman, 1995). The proposed link between
empowerment and advocacy may illuminate important knowledge regarding
the relationship between selfefficacy and advocacy in general. However,
women’s gender consciousness and collective action efforts have typically
been weaker than that of other groups, despite women generally being more
aware of gender issues than men (Aronson, 2003). Additionally, research in
this area of inquiry has underrepresented perspectives of men, minorities, or
lower socioeconomic status participants. This chapter discusses previous
research regarding the evolution of feminism, factors affecting feminist
identification, and the relationships between and among feminist
identification, feminist perspectives, and self-efficacy. A discussion of the
theoretical framework and brief summary of this chapter are also included.
Evolution of Feminism
As society and cultural context has continued to change over time, so
have the goals and ideologies of feminists (Phillips & Cree, 2014). Although
feminism may have begun before the late 1840s, it was during this time
period that the first collective efforts for women’s rights began (Phillips &
Cree, 2014). Today, there is some debate as to whether we are in a third
wave of feminism, or if we have crossed into a fourth wave. While previous
waves of feminism have included more singular views and identities for
group members, feminism today consists of more varied opinions about
what it means to be a feminist and how identification as a feminist is
expressed (Heywood & Drake, 1997; Aronson, 2003). Information about
each wave, along with the current state of feminism, is discussed below.
First Wave of Feminism (Late 1840s-1920s)
With urban industrialization and a move towards more liberal politics in the
late 19th century came the first wave of feminism in the United States. The
Industrial Revolution resulted in more women finding full-time work outside
of the home, which also provided women with various opportunities to
engage in discussions about social and political issues. First wave feminists
focused on a variety of issues affecting women and children, such as: (a) the
right for women to own property; (b) women’s suffrage; (c) access to higher
education; (d) protecting women and children from prostitution; and (e)
raising the age of sexual consent for women (Cree, 1996). First wave
feminists also believed that women were morally superior to men and rooted
their goals in the idea that men and women “inhabited separate spheres,”
with the hope of bringing female influence into a male-dominated world
(Phillips & Cree, 2014). However, this wave mostly consisted of middle-
class, heterosexual White women and did not take into account perspectives
of other classes, races, or sexual orientations (Phillips & Cree, 2014).
Second Wave of Feminism (Early 1960s-Late 1980s)
Women’s work and family lives were transformed yet again both
during and after World War II. In the post-WWII world, political views
became more liberal and women’s roles both inside and outside the home
continued to evolve. By the 1960s, the civil rights movement and Vietnam
War were heavy influences on the increasingly radical political ideologies
and activist efforts. This context can be seen throughout the second wave of
feminism, when feminists began to see “individual, social and political
inequalities as inevitably interlinked,” (Phillips & Cree, 2014) as evidenced
by the introduction of the feminist slogan “the personal is political” during
this time frame. Second wave feminists continued their focus on some of the
same issues as first wave feminists (i.e., equality in educational access,
protection of women from prostitution) (Cree, 1996). However, the
introduction of the contraceptive pill influenced second wave feminists’
interest in reproductive rights for women (including abortion), equality in
the workplace, and rape and domestic violence against women (Phillips &
Cree, 2014). Additionally, all of the societal changes during this time helped
illuminate that differences do exist among women, and that feminists had
not been including the perspectives of women who were not White or
middle-class (Ramazanoglu, 1989).
Third and Fourth Waves of Feminism (Late 1980s-present)
The third wave of feminism embraced more ambiguity regarding the
definition of what it means to be a feminist. Third wave feminists accepted
the idea of different feminist ideologies, saw gender as an expression not a
biological condition, and encouraged the involvement of men in feminism
(Phillips & Cree, 2014). Feminists also began examining intersections of
gender and other forms of oppression (Wrye, 2009). Although feminist
ideology continues to evolve, there is currently some debate as to whether or
not we have entered into a fourth wave of feminism. Proponents of a fourth
wave argue that technology and media have vastly changed how we are
presently viewing and understanding what it means to be a feminist (Phillips
& Cree, 2014). With various campaigns aimed at increasing awareness of
feminist issues (e.g., BanBossy, LikeAGirl, HeForShe), social media and
endorsement of feminism by public figures has re-ignited conversations
about what the feminist label means and who can claim it. For young adults
who have come of age in the 21st century (i.e., Millennials, Generation Y),
technology and social media are viewed as a normal part of life, and
therefore exert powerful influence over this and younger generations.
However, Zucker (2004) argues that education, personal relationships, and
personal struggles are greater influences on feminist identity and perceptions
of feminism, whereas exposure to media creates barriers to feminist
identification. Young adults use sites like Twitter and Facebook to share
political thoughts and opinions with the world. And with the continued
evolution of feminism, the parameters of feminist characteristics regarding
sexuality, employment, and reproduction also continue to expand. The most
recent wave of feminism has also developed a “call-out culture” in which
social media is used to address all forms of oppression in an effort to include
and support minority groups, working to eliminate power differentials and
eradicate previous negative perceptions about feminists (Phillips & Cree,
2014).
Factors Affecting Feminist Identification
While men and women frequently report agreeing with feminist ideas
and tenets, they rarely self-label as feminists (McCabe, 2005; Roy, Weibust,
& Miller, 2007; Williams & Wittig, 1997; Zucker, 2004). Despite potential
positive benefits to identifying as feminist, a variety of factors inhibit
individuals from self-labeling as a feminist. Negative stereotypes and beliefs
about feminism, gender socialization, minority status, and gender all impact
one’s choice regarding whether or not they identify as a feminist.
Negative Perceptions of Feminism and Feminists
One explanation for the negative beliefs about feminists and feminism
is the development and maintenance of negative stereotypes (Dottolo, 2011).
Even when women do label themselves as feminists, they believe that
“typical” feminists are different than them and more radical in their views
(Anderson, 2009; Twenge & Zucker, 1999). Negative portrayals of feminists
in the media contribute to these beliefs and may make people hesitant to
identify with a group that is not valued or seen in a positive light (Wiley et
al., 2012). Roy et al. (2007) found that college-aged women who were
exposed to positive portrayals of feminists were twice as likely to identify
with feminism compared to women exposed to negative or neutral
portrayals. When investigating the reasons behind the hesitation to self-label
as a feminist, one of the most frequently expressed reasons from both men
and women is the fear that they will be perceived negatively and
inaccurately by others (Anderson, 2009; McCabe, 2005; Roy, Weibust, &
Miller, 2007; Zucker, 2004). For example, research has shown that while
feminist women are thought of as being more intelligent, confident, and
productive, they are also thought of as being lesbians, less attractive, and
volatile (McCabe, 2005; Roy, Weibust, & Miller, 2007; Suter & Toller,
2006; Twenge & Zucker, 1999). Men who identify as feminists are typically
seen as being less masculine, more likely to be gay, and less attractive than
nonfeminist men (Anderson, 2009; Twenge & Zucker, 1999). However,
women rated feminist men as being more warm, affectionate, and kinder
than men in general (Anderson, 2009). Anderson (2009) posits that this more
negative stigma for feminist men may be due to the fact that the feminist
label is most commonly linked to women. So while feminist men may be
seen more positively than non-feminist men, they typically do not receive as
much respect (Wiley et al., 2011). Liss, Hoffner, and Crawford (2000) even
found that women who identify as feminists still consider other feminists to
be more radical in their thinking and behavior. Alexander and Ryan (1997)
found that only one out of thirty-six women self-identified as a feminist
without attempting to qualify the label with statements about her
background, sexuality, and specific feminist beliefs. Twenge and Zucker
(1999) also found that overall women viewed feminists as being “not like
me.” Further, women in this study felt that others possess extremely negative
stereotypes and views about feminists, even if the participants themselves
did not share the same views. These studies illuminate the deep impact that
negative stereotypes of feminism have on those who self-label as feminists.
Feminist men are often characterized as being gay, less masculine, and
having more stereotypically feminine qualities (e.g., emotional, submissive,
physically weaker) (McCabe, 2005; Roy, Weibust, & Miller, 2007; Twenge
& Zucker, 1999). Dottolo (2011) also asserted that feminism is often
demonized because in-groups react with fear and anger when out-groups
make attempts to gain more power; therefore, stereotypes are created and
maintained in an effort to preserve the status quo.
However, research has suggested that having positive perceptions of
feminists and feeling connected to women can buffer the fear of being
perceived negatively by others and increase the likelihood that someone will
self-identify as a feminist (Myaskovsky & Wittig, 1997; Roy, Weibust, &
Miller; 2007; Wiley et al., 2012). Even brief exposure to positive portrayals
of feminists has shown to positively influence perceptions of feminism, as
well as increase the desire to participate in collective action for women
(Roy, Weibust, & Miller, 2007; Wiley et al., 2012). Roy, Weibust, and
Miller (2007) tested this notion in a study of 414 undergraduate female
psychology students. Participants were randomly assigned to one of three
paragraph conditions: (1) positive stereotypes about feminists; (2) negative
stereotypes about feminists; or (3) control paragraph that discussed a general
topic unrelated to feminism. After reading the paragraph, participants
completed a feminist attitudes scale, measure of feminist identification,
gender identification scale, and performance self-esteem measure. Lastly,
participants completed an assessment to measure their perceived ability to
evaluate the paragraph they read. Roy, Weibust, and Miller (2007) found
that participants in the positive portrayal group were nearly twice as likely
(30.8%) to label themselves as feminists compared to the negative portrayal
group and control group, whose scores were not significantly different from
each other (18% and 16.7%, respectively). However, Roy et al. (2007)
indicated that women exposed to the positive portrayal condition did not
significantly differ from the negative portrayal or control groups on their
scores for endorsement of feminist attitudes. Women in the positive
portrayal group who identified as feminists had greater nontraditional gender
role attitudes and higher performance selfesteem (“expressing confidence
about one’s ability to meet challenges” [Roy, Weibust, & Miller, 2007]) than
participants in the other two groups. The sample in this study consisted of
predominantly white, heterosexual females, as is the case with most of the
studies mentioned in this chapter.
Gender Socialization and Gender Beliefs
Feminism often seems to be viewed as an oppositional, exclusive idea
from masculinity for men and from femininity for women. Twenge and
Zucker (1999) asked college students to develop a story based on one of two
prompts: (a) “Michelle calls herself a feminist,” or (b) “Michael calls
himself a feminist.” While participants wrote positive statements about both
Michael and Michelle, there were significantly more negative assertions
written about both the feminist man and feminist woman; however, more
negative statements were written about Michael compared to Michelle (e.g.,
“Michael is a cross-dresser by night”). Further, participants in the study
more often attributed assertive or masculine characteristics to Michelle and
weaker or more feminine qualities to Michael. However, Breen and
Karpinski (2008) found that when asked to comparatively rate non-feminist
and feminist women, feminist women were overall rated more positively
than non-feminist women. Conversely, participants in the study rated
feminist men much lower than non-feminist men. Breen and Karpinski
(2008) did not report a difference in how participants rated feminist men
compared to feminist women, Anderson (2009) performed a t test on the
available data from the original Breen and Karpinski (2008) study and found
that participants rated feminist men significantly less favorably than feminist
women. Gourley and Anderson (2007) had somewhat different findings
when they asked college-aged students to rate a female feminist speaker, a
male feminist speaker, a female non-feminist speaker, and a male non-
feminist speaker. Overall ratings for the feminist speakers (both male and
female) and for the non-feminist speakers (both male and female) did not
significantly differ; nevertheless, students were more likely to label the male
feminist speaker as being gay or bisexual compared to the other speakers.
Anderson (2009) also produced contradictory findings regarding ratings of
feminist men and feminist women, with participants rating the term
“feminist man” more favorably than “feminist woman.” And while women’s
ratings for “feminist woman” were not significantly different from their
ratings for “man” or woman,” men’s ratings for “feminist woman” were the
lowest of all of the other aforementioned terms. Further, Anderson (2009)
found that men and women in the study rated feminist men as being less
masculine and more likely to be gay. And while women in the study rated
feminist men more favorably than non-feminist men overall, they also rated
them as less sexually attractive than non-feminist men. Therefore, this study
will compare men’s and women’s feminist identity and perceptions on
feminism in order to understand existing similarities and differences.
Minorities, Men, and Feminism
While several studies have investigated how such issues affect White
collegeaged women, perspectives from ethnic and racial minorities and men
in the same age group have not been well represented. Previous research has
suggested that status as a minority may make individuals more likely to
engage in advocacy for other minority groups (Kelly & Breinlinger, 1995;
Wiley et al., 2012), some researchers believe that this may not be the case
with feminism, as it may only align with the issues and experiences of
White, middle class women (Williams & Wittig, 1997; Zucker, 2004). And
while some female feminists do not believe that men should be a target of
feminist campaigns, hooks (1984) presented the following argument: “since
men are the primary agents maintaining and supporting sexism and sexist
oppression, they can only be successfully eradicated if men are compelled to
assume responsibility for transforming their consciousness and the
consciousness of society as a whole.” Despite the common belief that
feminism is only for women, men are also directly affected by anti-feminist
perspectives; this is because, like women, men are not an “ahistorical,
universal, and foxed category of analysis” (Mohanty, 1988). These
antifeminist perspectives, sometimes referred to as “postfeminism” are
rooted in the belief that gender and sexual equality has been achieved, ruling
feminism as outdated and unnecessary (O’Neill, 2015).
In order to assess for gender differences in perceptions about
feminists, Anderson (2009) randomly assigned 404 college students to
complete semantic differential ratings (Pierce et al., 2003) for one of four
groups: (a) “man;” (b) “woman;” (c) “feminist man;” or (d) “feminist
woman.” Participants were then asked to complete a demographics
questionnaire and a feminist identification assessment (Myaskovsky &
Wittig, 1997). On the feminist identification assessment, participants were
asked to select one of the following statements regarding their stance on the
feminist label: (1) I do not consider myself a feminist at all, and I believe
that feminist are harmful to family life and undermine relationship between
men and women; (2) I do not consider myself a feminist; (3) I agree with
some of the objectives of the feminist movement but do not call myself a
feminist; (4) I agree with most of the objectives of the feminist movement
but do not call myself a feminist; (5) I privately consider myself a feminist
but do not call myself a feminist around others; (6) I call myself a feminist
around others; or (7) I call myself a feminist around others and am currently
active in the women’s movement. The corresponding numbers indicate the
level of feminist identification; therefore, a higher number indicates stronger
identification as a feminist and a lower number indicates weaker
identification as a feminist. The majority of men in the study (59.7%) did not
identify as feminists by selecting one of the first two statements, but 32.6%
of men said that they agreed with some feminist objectives but did not call
themselves feminists (which was the most popular choice of female
participants in the study [45.4%]). However, less than 1% of men identified
as feminist either publicly or privately compared to nearly 7% of women.
Wiley et al. (2012) also sought to illuminate men’s perspectives on
feminism. In their study, they presented male participants with one of three
paragraphs: (1) positive portrayal of feminist men; (2) negative portrayal of
feminist men; (3) a history of feminism that did not mention feminist men
(control condition). Participants then completed a scale to measure feminist
solidarity and an assessment for collective action intentions. Both measures
asked items that were rated on a 7-point Likert scale, with scores being
averaged; higher scores indicated higher levels of the measure variables.
Wiley et al. (2012) found that the brief exposure to positive portrayals of
feminist men positively influenced participants’ score on the feminist
solidarity scale and the collective action scale, while negative portrayals did
not decrease scores on either measure. Therefore, providing permission for
men to be feminists and emphasizing positive characteristics of feminist men
may transform negative beliefs about feminist men. These results underscore
the importance of having positive portrayals of feminists, how such
portrayals influence self-labeling and advocacy intentions, and why the
impact of increased media exposure about feminism may be important to
understand. Wiley et al. (2012) also posited that it may not be the presence
of negative stereotypes that impact men’s beliefs about feminism and
feminist identification, but that it instead may be largely impacted by the
absence of overall positive regard for feminist men. This may also hold true
for minority groups, as positive portrayals of minority feminists may not be
as available.
Masculinity. Ideas of masculinity and manhood are just as rigid as
notions of femininity and womanhood, and when men exhibit characteristics
that are deemed more appropriate for women they are viewed as being less
manly (Ratele, 2013). For example, men are traditionally taught that
emotional expression (e.g., crying) indicates weakness or femininity
(Wallace, 2007). Mahalik and colleagues (2003) proposed the following
traditional masculinity norms: winning, emotional control, risk-taking,
power over women, violence, dominance, playboy (i.e., being emotional
uninvolved in sexual relationships), primacy of work, self-reliance, disdain
for homosexuals, and pursuit of status. When men ascribe to traditional
masculinity, they are more likely to experience greater psychological
distress (Mahalik et al., 2003), to engage in substance abuse
(Mahalik, Lagan, & Morrison, 2006), and to exhibit hostile behavior
(Jackupcak, Tull, & Roemer, 2005). Black men who exhibit traditionally
masculine behaviors may experience poorer mental health outcomes, such as
depression and low self-esteem (Mahalik, Pierre, & Wan, 2006). Further,
Ratele (2013) purports that Black men may feel even more pressure to fulfill
societal expectations of manhood, as race may be a competing force with
gender. For example, although a Black man may be in a dominant position
as a male, they may still feel subordinate and experience oppression because
of their race; therefore, they may feel the need to overcompensate with
traditional masculinity in order to counteract feeling subordinate due to their
race. However, some research has indicated that Black men define
masculinity differently, including concepts of responsibility, maturity,
sacrifice, and accountability in their definition (Mincey et al., 2014).
Men ascribing to traditional masculinity norms has also been shown to
negatively affect their female partners. For example, women reported lower
relationship satisfaction and self-worth (Burn & Ward 2005) and higher
levels of anxiety and depression (Rochlen & Mahalik, 2004) when their
male partners embodied traditional masculinity. Further, traditionally
masculine husbands in heterosexual, dual-career households were less likely
to share childcare and housekeeping responsibilities with their wives, despite
both partners having equitable income (Mintz & Mahalik, 1996). Research
has also suggested that traditionally masculine men are more violent in
general (Courtenay, 2000), and they are more likely to engage in relationship
violence (Mahalik, Aldarondo, Gilbert-Gokhale, & Shore, 2005) and sexual
assault (Locke & Mahalik, 2005). This demonstrates the feminist
understanding that the sociopolitical context of male privilege negatively
impacts the psychological, social, economic, and political development of
both men and women (Brady-Amoon, 2011).
Intersectionality. Intersectionality acknowledges that every person
has multiple identities (e.g., race, class, gender), which create interdependent
systems of discrimination (Love, 2016). As individuals occupy different
roles, they have different access to power and resources, acting as the
oppressed in some settings and the oppressor in others (Alinia, 2015). In
other words, a Black man may be privileged in some settings due to his
gender, but in other settings his is penalized (i.e., oppressed) due to his race.
Although individuals occupy multiple identities, those who have competing
identities may sometimes feel like they have to choose one identity over the
other. For example, like Black men, Black women may feel torn between
their race and gender when issues of inequality arise. No matter what
identity they choose, they will be “taking sides against the self” (Collins,
2000). Historically, Black women have chosen to take the side of race
instead of gender, possibly realizing that Black men are also still oppressed
(Alinia, 2015). Therefore, Black women may choose to unite against the
common enemy of racial inequality, “ignoring internal injustice” (i.e.,
injustice against women) (Collins, 2000), possibly decreasing the number of
Black women who engage in activism for issues other than racism.
While the struggle of each oppressed group is related to other social
justice issues, each group’s experience is also dependent of other social
justice problems (Alinia, 2015). In other words, oppressed groups are
created and defined in relation to each other; however, inequity of power and
resource access between those groups still remains. Collins (2000) argues
that self-reflexivity, dialogue across oppressed groups, and mutual support is
needed. Collins (2000) further suggests that in order to enact political
change, groups must stop identifying people as either the oppressed or the
oppressor, and instead recognize individual and group identities. Collins’
(2000) notion of mutual support across oppressed groups also strengthens
the argument that minority status may increase the likelihood that an
individual will engage in supporting minority groups outside of their own
(Hunter & Sellers, 1998; Kelly & Breinlinger, 1995; Wiley et al., 2012).
Black feminism. The experience of Black women in America is
unique and complex due to similar experiences with racism, sexism, and
stereotyping (Love, 2016). Black feminism focuses on this unique
experience, acknowledging the intersectionality of identities like race, class,
and gender. Additionally, Black feminism highlights the relationship
between power and knowledge, while also questioning the notion of
objective knowledge (Alinia, 2015). Another tenet of Black feminist thought
is that no one group can obtain power without oppressing other groups; this
is based on the belief that each group decides which form of oppression is
most important, thereby deeming others as less important (Alinia, 2015).
Black feminists also believe that power flows among one’s privileged
identities, providing varying levels of privilege and resources depending on
the setting (Collins, 2000). Additionally, Black feminism focuses on
activism and shared history, with the shared experiences of being a Black
women in America being essential for consciousness raising and mobilizing
resistance efforts (Collins, 2000). However, Black feminism also
acknowledges that the collective identity of “Black woman” contains
internal differences due to varying positions in sexual orientation, social
class, education, age, and religion. These internal differences are important
to note, as saliency of oppression type may impact an individual’s ability to
view other forms of oppression as equally important. Further, because
traditional feminism has previously focused primarily on the issues more
salient to White women, it is important to understand why individuals who
ascribe to Black feminism may not support some feminist issues.
Additionally, if an individual is unaware of what traditional feminism is due
to a lack of education about the topic (Kelly, 2015), it is also unlikely that
they would be informed about Black feminist. Consequently, a lack of
education about Black feminism, combined with a belief that traditional
feminism is only for White women, may reduce the likelihood of African
American women identifying as feminists.
Impact of geographic region. Because data for the current study was
collected solely in the southeastern United States, it is important to
understand the potential impact of geographic region on feminist
identification and perspectives. Traditionally, individuals living in the
northern United States and individuals living in the southern United States
are thought to have different, and often opposing, views on issues like
politics and gender role expectations. Southerners are expected to be more
religious and more traditional in their views on gender roles compared to
non-southerners (Hurlbert 1989; Rice and Coates, 1995; Twenge, 1997).
Additionally, research suggests regional differences for racial attitudes and
gender-role attitudes, with white southerners exhibited more racial prejudice
than northerners (Kulinski et al., 1997). While traditional gender attitudes
are encouraged for both southern men and women, expectations for
women’s behavior are more defined and more mandated by culture than
expectations for men’s behavior (Suitor & Carter, 1999). It is also important
to note the impact of religion and political affiliation on gender role attitudes
and feminist perspectives, as the south is a predominantly conservative,
Christian region. In a study conducted by Lottes and Kuriloff (1992),
individuals who were more politically liberal were less accepting of
traditional masculinity and negative attitudes about homosexuality, more
accepting of feminist attitudes, and less traditional in attitudes about female
sexuality. Further, Morgan (1987) found that religious devoutness was a
significant predictor of traditional gender role attitudes.
Relationship between Feminism and Self-Efficacy
Bandura (1997) defined self-efficacy as one’s belief in their capacity
to achieve a desired outcome. An important component of social cognitive
theory, self-efficacy affects an individual’s feelings (e.g., depression,
anxiety, depression), thoughts (e.g., motivation, decision making, academic
achievement), and actions (e.g., anticipating outcome scenarios, exertion of
effort, recovery from setbacks) (Bandura, 1997). Further, one’s belief in
their self-efficacy determines the initiation of coping behaviors, the amount
of effort presented, and the amount of time someone will continue to exert
effort when they encounter obstacles (Bandura, 1997). This seems
particularly important for individuals who publically identify as feminists, as
higher self-efficacy may act as a buffer when they face adverse experiences
related to the feminist label. Additionally, because individuals with high
self-efficacy are more likely to select challenging environments, they may
already be prepared for obstacles that arise out of the declaration of their
feminist identity. Bandura (1977) also posited that self-efficacy beliefs are
derived from the following four sources: (a) performance accomplishments;
(b) vicarious experiences; (c) verbal persuasion; and (d) physiological states.
Performance accomplishments are an individual’s experiences with mastery
and are the most influential factor in determining self-efficacy (Bandura,
1977). When an individual successfully executes a task or a skill, that
success serves as evidence of his or her ability to accomplish that goal,
thereby increasing their self-efficacy. Conversely, failure will likely
decrease an individual’s selfefficacy. Vicarious experiences (i.e., modeling)
provide individuals with external examples of a target goal being obtained.
In other words, if individuals see other people succeeding at a task, they may
be more likely to believe that they can be successful at accomplishing that
same task. While this factor is not as influential as mastery experiences in
increasing self-efficacy, it may be particularly useful for individuals who
have low levels of self-efficacy (Bandura, 1977). Verbal persuasion, or
social persuasion, is direct encouragement or discouragement from another
person that has the ability to impact an individual’s perceived self-efficacy.
Finally, self-efficacy can be impacted by physiological states. When an
individual experiences emotional or physical responses to a stressor, it is his
or her interpretation of those responses that can impact levels of selfefficacy
(Bandura, 1977). In other words, if a person experiences heart palpitations
and a fluttering sensation in their stomach before giving a speech, his or her
self-efficacy may be negatively impacted if he or she perceives those
responses as an indicator of unpreparedness. However, individuals with
higher self-efficacy are more likely to identify such sensations as normal
psychological responses to stress (Bandura, 1977). By adjusting any one of
these four factors, an individual is thereby impacting their selfefficacy
beliefs.
Roy, Weibust, and Miller (2007) found that women exposed to
positive portrayals of feminists not only had greater nontraditional gender
role attitudes, but also had higher performance self-esteem, which the
researchers defined as “expressing confidence about one’s ability to meet
challenges.” (p. 154). This definition of performance self-esteem is almost
identical to the definition of self-efficacy used in the current study. Adoption
of feminist beliefs has also been linked with higher body satisfaction, feeling
more attractive, and having an increased ability to cope with societal
pressures about appearance expectations (Kinsaul et al., 2014). Further,
Kinsaul et al. (2014) found that self-efficacy was a significant predictor of
such factors for college-aged women. In the same study, self-efficacy
explained more variance in beliefs about one’s body than feminism itself,
suggesting that self-efficacy is a crucial piece in understanding the mental
health of college-aged women. And while identification as a feminist may
encourage positive mental health, identifying as a feminist may not be as
important as an individual adopting feminist attitudes due to the negative
stereotypes about feminists (Eisele & Stake, 2008; Twenge & Zucker, 1999).
In other words, feminist identification may not be necessary for positive
mental health outcomes because of the heavy stigmatization of the feminist
label.
Zucker (2004) examined the effect of feminist identity on feminist
activism in college-aged women. Participants completed a feminist
identification measure, a feminist consciousness assessment, a questionnaire
inquiring about favorable conditions for adopting feminist identity (i.e,
exposure to feminism through education, personal relationships, and
personal struggles), a questionnaire about barriers to feminist identification,
two measures to assess feminist activism (i.e., Feminist Identity
Development Scale [Bargad & Hyde, 1991]), and a behavioral index created
by the researchers). Results suggested that being exposed to feminism in
various contexts influences feminist identification; in particular, participants
who identified as feminists had more exposure to feminism in the favorable
conditions categories. Further, the feminists in the study rated higher on
feminist consciousness and experiences of sexism compared to non-
feminists and egalitarians (i.e., participants who agreed with feminist ideas
but did not label themselves as feminists). Lastly, participants who identified
as feminists were more likely to engage in feminist activism, regardless of
favorable conditions or barriers to feminist identification. This finding
illuminates the link between feminism and advocacy, suggesting that
feminist identification is a better predictor of social justice participation,
even if individuals are faced adverse conditions or possible negative
consequences.
Theoretical Foundation
Feminist Theory
Feminist theory shares many common threads with multicultural
counseling theories (Crethar, Rivera, & Nash, 2008; Goodman et al., 2004)
that are currently taught in counselor education programs. Such similarities
between feminist and multicultural principles include: (a) identifying social
oppression as a contributor to mental health issues; (b) the belief that mental
health symptoms are often the result of oppressive conditions, not of
pathology; and (c) the importance of clients learning ways to cope with
oppression in their everyday lives (Goodman et al., 2004).
Another core component of feminist theory is the call for on-going
selfevaluation. Without being aware of deeply rooted, automatic biases and
stereotypes, counselors and counselor educators are unaware of how racial
dynamics and their conceptualization of treatment and pathology interfere
with their thoughts, behaviors, and reactions to clients (Helms & Cook,
1999). Similarly, awareness of one’s biases and prejudices allows counselors
and counselor educators to be aware of their inability to be value-free and to
clarify such values in a transparent manner with clients, students, or
supervisees (Enns, 1997). Because multicultural approaches are highly
valued in today’s counselor education programs, understanding feminist
theory as well seems like a logical next step for counselor educators. For
example, both feminist and multicultural approaches place a heavy emphasis
on social justice, acknowledging that “social justice work is the social
context in addition to or instead of the individual” (Goodman et al., 2004, p.
795). And while feminism has been criticized for originating from a place of
White privilege and power (Dill, 1983), evolving feminist perspectives are
more inclusive and aware of the experiences and worldviews of non-White
women (Goodman et al., 2004). Feminist theory also lends itself to this
study because of its focus on empowerment and facilitating consciousness
raising (Corey, 2009). Empowerment consists of one’s perceived personal
power, as well as one’s general sense of positive self-regard, which includes
self-esteem and self-efficacy (Kinsaul et al., 2014). Group consciousness is a
concept which consists of both group identification (e.g., feminist
selflabeling) and awareness of existing inequities, with the intention to take
action on behalf of the group (Kinsaul et al., 2014).
Social Cognitive Theory
Social cognitive theory can also be used to help understand the
constructs being examined in this study, particularly with the college-aged
population. Social cognitive theory (SCT) posits that individuals’ beliefs in
their abilities to influence the environment shape their actions in order to
produce desired outcomes (Stajkovic & Luthans, 1998). As societal views
on sexual mores, family structure, and gender roles continue to evolve,
stereotypes continue to be influenced primarily by culture, not by inherent
biological differences between men and women (Khajehpour, Ghazvini,
Memari, & Rhamani, 2011). Khajehpour et al. (2011) elaborate on this idea
by suggesting that modeling is the most powerful means of transmitting
cultural values, attitudes and behaviors, and thought patterns across
generations. While modeling can occur through direct observation of others,
media may also serve as a source from which individuals model behavior.
When modeling occurs, the observer extracts underlying rules of behaviors
and goes slightly beyond what they have observed, generating new patterns
of behavior (Khajehpour, 2011). However, emotional state and
preconceptions the observer possesses serve as prejudicial influences. This
underscores the important roles of stereotypes and portrayals of feminists
and feminism in media. However, Khajehpour et al. (2011) found that
selfefficacy beliefs are vital in the attainment and maintenance of gender
stereotypes and beliefs about appropriate behavior. Because young adults
tend to be more susceptible to the media’s influence on their behavior and
beliefs (Austin, Vord, Pinkleton & Epstein, 2008; Jackson, 2005), using a
social cognitive theory lens to explain the potential power of this influence
on feminist identification may be useful. Further, tenets of social cognitive
theory may help explain why individuals choose to identify or not identify as
feminists (e.g., learning as a cognitive process in a social context, vicarious
reinforcement). Social cognitive theory may also help to explain the
relationship between self-efficacy and advocacy, as those with higher levels
of self-efficacy may persist with action despite unfavorable circumstances as
long as they believe their efforts will produce the expected results (Stajkovic
& Luthans, 1998). Social cognitive theory also helps explain how observed
behavior (i.e., modeling) can influence values, attitudes, and thoughts,
thereby affecting stereotypes and regulation of gender roles that are typically
associated with the feminist label (Khajehpour et al., 2011). Thus,
incorporating feminist theory and social cognitive theory (SCT) provides
both a political/social lens and a learned behavior (i.e., modeling) lens
through which to view the impact of feminist labeling and feminist
perspectives on self-efficacy in undergraduate and graduate students.
Summary
While previous studies have examined variables related to feminist
identification, feminist perspectives, and self-efficacy, there have been gaps
in the research. One of the biggest gaps is related to diversity among
samples. The majority of previous studies have investigated how such issues
affect primarily White, college-aged, middle-class women, with perspectives
from ethnic and racial minorities and men in the same age group being
underrepresented. Subsequent chapters discuss the methodology of the
current study (Chapter 3), provide results from the current study (Chapter 4),
and discuss conclusions and suggestions for future research on the proposed
topic (Chapter 5).
CHAPTER 3
METHODOLOGY
Despite women being more aware of gender-related issues than men,
women’s gender consciousness and collective action efforts have been
weaker than that of other groups (Aronson, 2003). Previous research on such
issues have primarily examined the impact on White, college-aged women,
either excluding or underrepresenting male and minority perspectives (Eisele
& Stake, 2008; Kinsaul et al., 2014; Roy, Weibust, & Miller, 2007; Twenge
& Zucker, 1999; Zucker, 2004). While some researchers believe that
minority status makes an individual more likely to support minority groups
outside of their own (Hunter & Sellers, 1998; Kelly & Breinlinger, 1995;
Wiley et al., 2012), other researchers speculate that feminism may be an
exception due its origins in White, middle-class culture (Aronson, 2003;
Hunter & Sellers, 1998; Williams & Wittig, 1997; Zucker, 2004). In an
effort to be more inclusive, third wave feminism consists of more diverse
views on feminist identity and expression of that identity, a contrast to the
more singular views of previous waves of feminism (Aronson, 2003;
Heywood & Drake, 1997). Such variance in beliefs may cause confusion,
possibly contributing to the reluctance to self-label as a feminist while still
aligning with feminist perspectives (Kelly, 2015). Researchers have
suggested that feminist perspectives are linked with positive mental health
outcomes such as higher perceived physical attractiveness, a more positive
body image (Kinsaul, Curtin, Bazzini, & Martz, 2014), and improvement in
coping with societal pressures (Rubin, Nemeroff, & Russo, 2004). Due to
the potential benefits of feminist identification, as well as the continuous
evolution of what it means to be a feminist, examining how young adults
view feminism and how they incorporate the label and perspectives into their
identities is important for counselors and counselor educators. By
understanding how students or clients incorporate the feminist label and
feminist perspectives into their identity can potentially provide guidance for
their work with students or clients while also encouraging a more trusting,
collaborative relationship.
Thus, the current dissertation aimed to (a) examine relationships
among feminist self-identification, feminist perspectives, and self-efficacy;
and (b) explore existing differences for race and gender on feminist self-
identification, feminist perspectives, and self-efficacy in undergraduate and
graduate students.
Research Questions & Hypotheses
This study explored the relationships among demographic variables
(e.g., race, gender), feminist perspectives, feminist identification, and self-
efficacy in young adults. As such, the following research questions were
examined.
Research Question 1
What relationship exists among feminist self-identification, feminist
perspectives, and self-efficacy? Specifically, can feminist self-identification,
as measured by the SelfIdentification as a Feminist Scale (SIF, Szymanski,
2004), and feminist perspectives, as measured by the Feminist Perspectives
Scale – Short Form (FPS3; Henley, Spalding, & Kosta, 2000) predict self-
efficacy, as measured by the General Self-Efficacy Scale (GSE; Schwarzer
& Jerusalem, 1995)?
Hypothesis 1A. Feminist identification ratings and feminist
perspectives ratings are positively correlated with self-efficacy ratings.
Hypothesis 1B. Higher ratings for feminist self-identification and
feminist perspectives will predict higher self-efficacy for young adults.
Research Question 2
What differences exist among various races (e.g., White, African
American, Hispanic) and gender (i.e., male, female), as measured by the
demographics questionnaire, between feminist self-identification, as
measured by the Self-Identification as a Feminist Scale (SFI; Szymanski,
2004), feminist perspectives, as measured by the Feminist Perspectives Scale
– Short Form (FPS3; Henley, Spalding, & Kosta, 2000), and self-efficacy, as
measured by the General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem,
1995)?
Hypothesis 2A. Female participants will have higher ratings than
male participants for feminist identification, feminist perspectives, and self-
efficacy.
Hypothesis 2B. White participants will have higher ratings than other
ethnicities for feminist identification, feminist perspectives, and self-
efficacy.
Research Design
Prior to beginning the study, I sought approval from the University of
South Carolina’s Institutional Review Board (IRB). This study consisted of a
quantitative, correlational survey research design and examined the
relationships among demographic factors, feminist perspectives, feminist
self-labeling, and self-efficacy in undergraduate and graduate students. Once
I received IRB approval, I began data collection. This study utilized data
collected from undergraduate and graduate students from the University of
South Carolina who were at least 18 years old. After providing consent to
participate in the study, participants completed (a) a researcher-developed
demographics questionnaire; (b) the Self-Identification as a feminist scale
(SIF; Szymanski, 2004); (c) the Feminist Perspectives Scale – Short Form
(FPS3; Henley, Spalding, & Kosta, 2000); and (d) the General Self-Efficacy
Scale (GSE; Schwarzer & Jerusalem, 1995).
Participants
Participants in this study included undergraduate and graduate
students who were currently enrolled at the University of South Carolina.
There are currently 24,864 undergraduate students enrolled at the
University, with this population being 54% female and 46% male. There are
8,108 graduate students enrolled at USC, comprised of 59% females and
41% males. Minorities comprise 21.6% of the undergraduate student
population and 31.6% of the graduate student population. I contacted
programs such as University 101 (containing approximately 4,000 students),
FemCo (a feminist student organization), the Counselor Education and
Supervision undergraduate minor program (containing approximately 250
students) and Education Specialist (EdS) program, National Pan-Hellenic
Council (NPHC; historically African American or multicultural fraternities
and sororities), and large undergraduate introductory courses (e.g.,
Psychology, Technology, Public Health) in order to recruit participants. I
engaged in both active (e.g., face-to-face) and passive (e.g., email, word of
mouth) recruitment strategies. Upon IRB approval, I obtained permission
from course instructors and program coordinators to briefly speak with their
students about participating in the study (i.e., active recruitment).
Additionally, I sent emails about the study to professors who taught large
undergraduate classes and requested they send study information to their
students via email or Blackboard (i.e., passive recruitment). Yancey, Ortega,
and Kumanyika (2006) noted that active recruitment strategies are more
effective with culturally diverse participants, hence my intention to partner
with programs for minority students (i.e., Men of Color Initiative, TRIO
programs).
In order to be included in the study, participants had to be at least 18
years of age and enrolled as an undergraduate or graduate student at the
University of South Carolina. Nonprobability sampling (specifically,
convenience sampling) was utilized to obtain participants for this study.
Researchers may employ convenience sampling because compared to other
sampling methods (a) it is more cost effective; (b) it has less restrictions for
obtaining participants; and (c) it may be more feasible for a particular
population (Rubin & Babbie, 2011). For the current study, White students
and women comprise the majority of both undergraduate and graduate
students (see Table 3.1), with an even larger gap existing between the
percentage of White students and racial minority students. Therefore,
attempting to obtain equal numbers for the smaller groups (i.e., males and
minorities) for this study was less feasible considering the population
demographics.
Table 3.1
Comparing Sample and Population (USC) demographics
USC Sample
Category
Percentage
Percentage
Gender
Males
45%
26%
Females 55% 74%
Race/Ethnicity
White
74.3%
74.6%
African American 10.5% 14.8%
Native American 0.2% 0%
Asian 2.3% 1.3%
Hispanic 3.7% 2.3%
Pacific Islander 0.1% 0.3%
Other 6.9% 6.4%
No Response 2.0% 0%
Racial Minority Status
White
74.3%
74.6%
Minority 23.7% 25.4%
However, convenience sampling may lead to an inadequate representation of
various groups in a sample (Rubin & Babbie, 2011). This limitation of the
study will be further discussed in Chapter 5. Since little is known about the
relationship between feminism and self-efficacy in men and minorities,
convenience sampling may provide insight into whether or not a problem
exists in a biased sample. Because convenience samples are typically
already biased (Rubin & Babbie, 2011), examining the relationship between
feminism and self-efficacy in men and minorities in a biased sample can
provide valuable information into how to proceed in future studies with
these two groups (Rubin & Babbie, 2011). In other words, if no significant
differences exist for men and minorities on the outcome variables compared
to women and non-minority participants in the biased sample, it may be
unlikely that significant differences would exist in an unbiased sample
(Rubin & Babbie, 2011). In attempt to counteract some of the issues related
to convenience sampling (e.g., biased results, unrepresentative sample), I
recruited participants from majors outside of liberal arts fields (e.g., Public
Health, Technology, Business), as these majors may have less exposure to
feminism and feminist perspectives through coursework.
I conducted an a priori analysis using G*Power 3.0.10 (Faul,
Erdfelder, Lang, & Butchnew, 2008) to determine sample size and adequate
power with each of the anticipated analyses. Statistical power is the
probability that a false null hypothesis will be rejected, thereby not
committing a Type II error (Fink, 2013). In other words, power is the
capacity to detect the effect of a test if that effect does in fact exist. A priori
power analyses helped determine the sample size necessary for adequate
power (Balkin & Sheperis, 2011; Cohen, 1992), with larger sample sizes
leading to less likelihood of a Type II error, higher statistical power, and a
larger effect size (Balkin & Sheperis, 2011). In other words, adequate power,
a smaller likelihood of error, and larger effects, increase the generalizability
and trustworthiness of the results. However, as the likelihood of a Type II
error decreases, the likelihood of committing a Type I error (reporting
significant results when results were not significant; i.e., a false positive)
increases (Fink, 2013). The a priori analysis conducted for the current
dissertation utilized an alpha level of .05, moderate effect size of .06 (Cohen,
1992), and a recommended power of .80 (Cohen, 1992). Because effect size
was not reported for previous studies (e.g., Eisele & Stake, 2008; Kinsaul et
al., 2014; Roy, Weibust, & Miller, 2007; Twenge & Zucker, 1999; Zucker,
2004), a moderate effect size was chosen for the current study. G*Power
indicated a sample of 107 in order to achieve adequate power for research
question one. The power analysis conducted for research question two
indicated a sample of 265 participants for adequate power. Because the
second power analysis indicated a larger sample size (N = 265) for research
question two, the goal is to obtain at least 265 participants in order to avoid
committing a Type II error. A Type II error occurs when the researcher fails
to reject a false null hypothesis (Rubin & Babbie, 2011), meaning that the
researcher reports finding no significant differences between groups when
such differences may actually exist. A larger sample size increases the
power of the test, which in turn decreases the likelihood of failing to reject a
false null hypothesis (i.e., Type II error; Fink, 2013). Very few previous
studies included response rates in their studies (Jackson, 2005; Szymanski,
2004; Zucker, 2004). Of the studies which included response rates, Jackson
(2005) reported an average response rate of 70%; however, the researcher
did not specify the format of the survey (i.e., web, paper). Additionally, this
particular survey did not examine constructs similar to those in the current
study. Only the population used in this study was similar (i.e., young adults).
Szymanski (2004) and Zucker (2004) reported response rates of 35% and
30%, respectively; however, these studies utilized mail surveys instead of
web surveys, which will be the method of administration for the current
study. Dillman Smyth, and Christian (2009) report that the tailored design
method can yield a response rate of up to 70% for mail surveys. Conversely,
Dill et al. (2009) note that electronic surveys yield lower response rates. In a
meta-analysis of 49 studies, Cook, Heath, and Thompson (2000) found an
average response rate of 35% for electronic surveys. Therefore, 35% will be
the expected response rate for the current study.
Measures
The following instruments were administered to study participants: (a)
a researcher-developed demographic form; (b) Feminist Perspectives Scale-
Short Form (FPS3; Henley, Spalding, & Kosta, 2000); (c) General Self-
Efficacy Scale (GSE; Schwarzer & Jerusalem, 1995); and (d) Self-
Identification as a Feminist Scale (SIF; Szymanski, 2004). In order to
determine the level of reliability of each measure, a reliability analysis was
conducted for each instrument to determine Cronbach’s alpha (α). All
instruments will be completed online using Survey Monkey, and results will
be exported to SPSS for analysis.
Demographics Questionnaire
The demographics questionnaire collected basic demographic
information from study participants. The form included questions about
gender, age, years of education, year in school (e.g., freshman, sophomore,
first year Master’s), and race. The demographics questionnaire consisted of
10 questions and was administered after participants had completed all other
assessments. See Appendix C for a sample of the demographic form.
Feminist Perspectives Scale – Short Form (FPS3)
The FPS3 (Henley, Spalding, & Kosta, 2000) is a 36-item interval
scale which assesses feminist attitudes and feminist behavior. Of the 36
items on the instrument, 30 items measured feminist attitudes and 6 items
measure feminist behavior. Participants were asked to rate their level of
agreement with the first 30 items on a 7 point Likert scale, ranging from
strongly disagree (1) to strongly agree (7). For the last six items,
participants were asked to reflect on how true they feel each item is of
themselves. The 7 point Likert scale for the last six items ranged from very
untrue of me (1) to very true of me (7). The 30 attitudinal items were
comprised of 6 subscales: (a) Conservative; (b) Liberal Feminist; (c) Radical
Feminist; (d) Social Feminist; (e) Cultural Feminist; and (f) Woman of
Color/Womanist. Responses for each subscale can be summed, with higher
scores indicating greater agreement with the corresponding subscale. These
scores are then summed together to obtain a total score for feminist attitudes
(i.e., Femscore3). Responses for the behavioral items are summed separately
in order to produce a total score for feminist behavior (i.e., Fembehave3).
Femscore3 can range from 25 to 175, while Fembehave3 can range from 5
to 35, with higher scores indicating greater agreement with feminist attitudes
and higher levels of feminist behaviors, respectively. The FPS3 has shown
high internal consistency for Femscore (α = .85), although some of the
subscales have shown alpha reliability ≤ .70. The FPS3 has shown to be
positively correlated with a longer, 78 item version of the Feminist
Perspectives Scale (FPS2; Henley et al., 1998) and with FPS3 retest scores,
showing large effect sizes (r ≥ .05). Additionally, the feminist subscales
were positively correlated with each other, demonstrating moderate to large
effect sizes (r ≥ .60 to .85). However, Henley et al. (2000) suggested using a
larger sample size (a sample of 209 was used in the development of the short
form being used in this study) in order to increase reliability for the
subscales. This instrument was normed on over 300 male and female
undergraduate students representing ethnically diverse backgrounds (i.e., 25-
31% White, 28-48% Asian, 10-18% Latino/a, 4-8% African American, 5-
6% multiethnic, and 7-8% foreign born).
However, the largest two groups on which the instrument was normed were
Whites and Asians. Therefore, this instrument may not be as accurate when
administered to members of other ethnic groups.
General Self-Efficacy Scale (GSE)
The General Self-Efficacy Scale (GSE; Scwarzer & Jerusalem, 1995)
measured beliefs about general self-efficacy. The GSE is a 10 item scale
which assesses one’s belief in their ability to respond to difficult situations
and cope with associated obstacles. The 4 point Likert scale measured the
extent to which each item applied to the participant, ranging from not at all
true (1) to exactly true (4). Participant total scores can range from 10 to 40,
with higher scores indicating higher general self-efficacy. The GSE was
originally in German but it has been translated into 33 different languages, to
include English. Scholz and colleagues (2002) conducted a study analyzing
the psychometric properties of the GSE with data from 19,120 participants
(7,243 men, 9,198 women, and 2,679 did not provide their gender) from 25
countries. Internal consistency was between .75 and .91 for the GSE, with
the United States data demonstrating a Cronbach’s alpha value of .87.
Across the 25 countries, participant age ranged from 12 to 94, with an
average age of 25 years old (SD = 14.7). However, of the only 50.4% of
participants who indicated their profession, only about one-third (34.7%)
identified as students. Further, no information about race or sexual
orientation was indicated in this study.
Self-Identification as a Feminist Scale (SIF)
The SIF (Szymanski, 2004) encompassed 4 items designated to assess
(a) both public and private identification as a feminist and (b) support of the
goals and values of the feminist movement. Participants rated items on a 5
point Likert scale based on their level of agreement or disagreement with
each item, ranging from strongly disagree (0) to strongly agree (4). A total
SIF score was obtained by summing all 4 items, with higher scores
indicating stronger identification as a feminist. Szymanski (2004) found an
alpha reliability of .93 for the SIF, with inter-item correlations ranging
from .81-.89. The SIF was also positively correlated with other measures of
attitudes towards feminism (r = .75.76). The instrument was normed on 227
women between ages 18 and 72 (mean age=38.25), with 85% being White.
Of the 227 participants, 82% identified as lesbians, 15% as bisexual, and 3%
as being unsure about their sexual orientation. Although the age range of
participants differs greatly from that of this study, the majority of
participants had either a graduate/professional degree (54%) or a four year
undergraduate degree (26%).
Procedures
After obtaining IRB approval from the University of South Carolina, I
began recruiting participants. Participants were recruited using both active
(e.g., face-to-face) and passive (e.g., email) recruitment. Yancey et al.
(2006) emphasized the importance of using active recruitment methods
when attempting to include minority groups, suggesting that active
recruitment is more effective than passive recruitment with minority
populations. Participants who completed the survey online were provided
with a website link to access the survey on Survey Monkey. Prior to
beginning the study, participants were prompted to review the informed
consent before continuing the survey. Participation in the study was
voluntary and participants could choose to discontinue their participation at
any time without penalty. If participants chose to continue the survey, they
then completed the FPS3, GSE, and SIF. After completing the FPS3, GSE,
and SIF, participants completed the researcher-generated demographic form,
which inquired about information such as age, race, and gender. Finally,
participants were given the option to supply their name and email address
for the chance to win one of four $25 Visa gift cards. Participants’ names
and email addresses were not connected to their survey responses, as
participants who entered the gift card drawing were asked to send me an
email containing only their name and preferred email address. Finally, no
identifying information was collected on the survey, and all data was
reported in aggregate form.
Variables
For research question one, feminist perspectives (as measured by the
FPS3) and feminist identification (as measured by the SIF) served as
independent variables while self-efficacy (as measured by the GSE) served
as the dependent variable. For research question two, race and gender (as
measured by the demographics questionnaire) served as the independent
variables while feminist identification (as measured by the SIF), feminist
perspectives (as measured by the FPS3), and self-efficacy (as measured by
the GSE) served as dependent variables.
Data Analyses
I conducted a preliminary analysis of the data in order to identify
outliers, missing data, and violations of assumptions. I conducted two
statistical analyses to evaluate the aforementioned research questions. A
linear multiple regression examined the relationships among the constructs
of feminist self-identification (as measured by the SIF total score), feminist
perspectives (as measured by Femscore3 total score, and Fembehave3
score), and self-efficacy (as measured by the GSE total score) for all
participants. Linear multiple regressions test for how much unique variance
the independent variables contribute to the depend variables (Pallant, 2013).
In other words, a linear multiple regression examines the relationship
between each independent variables and the dependent variable. I also tested
for outliers, missing data, and any violations of the assumptions of
normality, multicollinearity, and singularity. A two-way factorial MANOVA
examined differences between various races (e.g., White, African American,
Hispanic) and gender (i.e., male, female) for feminist self-identification (as
measured by the SIF total score), feminist perspectives (as measured by
Femscore3 total score, and Fembehave3 total score), and self-efficacy (as
measured by GSE total score). SPSS was utilized for all statistical
procedures for this study. In order to protect participants’ rights, data was
only reported in aggregate form with no identifying information connecting
participants to their responses. All data was contained on a password
protected laptop, and I was the only person with access to the data.
Summary
The current dissertation intended to (a) examine relationships among
feminist self-identification, feminist perspectives, and self-efficacy; and (b)
explore existing differences for race and gender on feminist self-
identification, feminist perspectives, and self-efficacy in undergraduate and
graduate students. A series of assessments for the abovementioned variables
of interest were administered to 305 participants. For this study, I utilized a
multiple regression and a two-way factorial MANOVA to analyze the data
obtained from participants. For research question one, the independent
variables included feminist identification and feminist perspectives, and self-
efficacy served as the dependent variable. For research question two, the
independent variables were race and gender, while feminist perspectives,
feminist identification, and self-efficacy were dependent variables. Results
of the current study can be found in Chapter 4 of this dissertation.
CHAPTER 4
RESULTS
The current study examined the relationships among feminist
identification, feminist perspectives, and self-efficacy in young adults. By
investigating the relationship and predictive ability of feminist self-labeling
and feminist perspectives on self-efficacy, more knowledge can be gained
about possible factors that may influence levels of selfefficacy in young
adults. Further, identifying the relationship among feminist identification,
feminist perspectives, and self-efficacy for various demographic groups can
contribute much needed data to the existing literature, which is primarily
based on the relationships among these constructs for White women. Data
analyses were conducted utilizing the Statistical Package for the Social
Sciences (SPSS) Version 23.
Sampling and Data Collection Procedures
The target population for this study was young adults (ages 18-39)
who were enrolled as either undergraduate or graduate students at the
University of South Carolina. Convenience sampling was utilized to obtain
participants for this study because compared to other sampling methods (a)
it is more cost effective; (b) it has less restrictions for obtaining participants;
and (c) it may be more feasible for a particular population (Rubin & Babbie,
2011). For the current study, White students and women comprise the
majority of both undergraduate and graduate students with an even larger
gap existing between the percentage of White students and racial minority
students. Therefore, attempting to obtain equal numbers for the smaller
groups (i.e., males and minorities) for this study was less feasible
considering the population demographics. In order to obtain participants, I
contacted 29 professors/instructors who were teaching introductory courses
in various departments (e.g., Public Health, Psychology, Counselor
Education, Information Technology) to acquire permission to collect data in
their classes. Of the 29 professors I contacted, 10 did not respond to my
request and eight agreed to disperse my study information to their students
via email or Blackboard. The remaining 11 professors/instructors allowed
me to attend their classes to collect data. One of the 19 instructors also
distributed the study information to 30 students at Limestone College
(included in the invited number of 1248 mentioned below); however, I
received no responses from students at this institution. Additionally, of the
eight student groups that I contacted, one dispersed the study information
through their group’s listserv. The remaining four groups allowed me to
attend their meetings to collect data. Potential participants were invited to
complete the survey in either a paper or online format. When professors
provided participants details about the survey via Blackboard, they included
a link to Survey Monkey. When I spoke to potential participants face-to-
face, I explained the purpose of the study and provided interested
participants with a paper version of the survey. Upon the completion of the
survey (in both formats), participants had the option of providing their
names and email addresses to be entered into a gift card drawing.
Participants’ names or contact information were in no way connected to
their survey responses, as no identifying information was collected on the
survey itself.
Descriptive Data Results
Response Rate
Overall, 1,248 potential participants received invitations to participate
in the study. Initially, a total of 319 participants responded, yielding a
response rate of 25.6%. However, eight participants did not complete all
assessments, thereby reducing the response rate to 24.9%. Finally, six
participants fell outside of the desired age range (1839) for young adults,
yielding a useable response rate of 24.4% for 305 participants. The response
rate for participants who received face-to-face invitations (e.g., active
recruitment) was 96.9%, while the response rate for the online version of the
survey (e.g., passive recruitment) was 11.4%. However, a lower response
rate is common for electronic data collection procedures.
Participant Demographics
Following are descriptive statistics for the 305 participants who
participated in the study. The women comprised the majority of participants
(n = 228, 74.8%), compared to men (n = 77, 25.2%). Of the participating
females, 176 (77.2%) identified their ethnicity as White/Caucasian, while 52
(22.8%) identified as a racial/ethnic minority. Of the participating males, 58
(75.3%) identified their ethnicity as White/Caucasian, while 19 (24.7%)
identified as a racial/ethnic minority. See Tables 4.1 and 4.2 for additional
demographics related to ethnicity.
Table 4.1
Frequencies of Participants by Ethnicity
Frequency Percent
Asian 4 1.3
Black/African American 41 13.4
Hispanic/Latino(a) 7 2.3
Native Hawaiian/Pacific Islander 1 0.3
White/Caucasian 233 76.4
Other 19 6.2
Table 4.2
Frequencies of Participants by
Gender an d Racial Group
Frequency Percent
White female 176 57.7
White male 58 19.0
Non-White female 52 17.0
Non-White male 19 6.2
Participants ranged in age from 18 to 39, with a mean age of 22.05 (SD =
3.79). Undergraduate students comprised 76.4% (n = 233) of the sample,
while graduate students comprised the remaining 23.6% (n = 72) of
participants. The average years of education for participants was 15.77, and
the mean number of credit hours in which participants were enrolled was
14.04. Most participants reported that they were not currently in a
relationship (n = 158, 51.8%), and the majority of participants identified as
heterosexual (n = 261, 85.6%). See Table 4.3 for additional demographics
related to student status and sexual orientation.
Table 4.3
Frequencies of Participants by Student Status and Sexual Orientation
Frequency Percent
Student Status
Freshman
35
11.5
Sophomore 54 17.7
Junior 62 20.3
Senior 83 27.2
Graduate student,
Master’s 24 7.9
Graduate student,
Ed.S
33 10.8
Graduate student,
PhD
14 4.6
Sexual Orientation
Heterosexual
261
85.6
Lesbian 5 1.6
Gay 5 1.6
Bisexual 22 7.2
Other 12 3.9
Feminist Identification
The Self-Identification as a Feminist Scale (SIF; Szymanski, 2004)
measured feminist identification. The SIF is a four item scale that assess (a)
both public and private identification as a feminist, and (b) support of the
goals and values of the feminist movement. The items contain a 5 point
Likert scale based on level of agreement or disagreement with each item,
ranging from strongly disagree (0) to strongly agree (4). Cronbach’s α
assessing internal consistency of the SIF was .93, indicating strong internal
consistency of the scale (Pallant, 2013). Internal consistency is important
because it demonstrates the degree to which all scale items are measuring
the same construct (Pallant, 2013). Participant total scores for the SIF ranged
from 0 to 16 (M = 10.31, SD = 4.35). Because each of the four items
inquires about a different facet of feminist identity, descriptive data and
measures of central tendency are included in Tables 4.4 and 4.5.
Table 4.4
Descriptive Statistics for Individual SIF Questions
Question M Mdn SD Range
I consider myself a feminist. 2.52 3.00 1.23 0-4
I identify myself as a feminist to other
people. 2.02 2.00 1.42 0-4
Feminist values are important to me. 2.87 3.00 1.05 0-4
I support the goals of the feminist
movement.
2.90 3.00 1.02 0-4
Table 4.5
Frequencies for Level of Agreement with SIF Items
Disagree Agree Neither
Agree/Disagree
Question n % n % n %
I consider myself a
feminist. 65 21.3 167 54.8 73 23.9
I identify myself as
a feminist to other
people.
113
37.0
118
38.7
74 24.3
Feminist values are
important to me.
29 9.5 206 67.5 70 23.0
I support the goals
of the feminist
movement.
26
8.5
214
70.2
65 21.3
Feminist Perspectives
The Feminist Perspectives Scale – Short Form (FPS3; Henley,
Spalding, & Kosta, 2000) assessed feminist attitudes and behavior. The
FPS3 is a 36 item instrument, with 30 items measuring feminist attitudes and
six items measuring feminist behavior. Items are rated on a 7 point Likert
scale, ranging from strongly disagree (1) to strongly agree (7) for the
attitudinal items and from very untrue of me (1) to very true of me (7) for the
behavioral items. The 30 attitudinal items are comprised of six subscales: (a)
Conservative; (b) Liberal Feminist; (c) Radical Feminist; (d) Social
Feminist; (e) Cultural Feminist; and (f) Woman of Color/Womanist. A total
score for attitudes (i.e., Femscore3) is obtained by summing five of the
subscales (excluding Conservative), and a total score for behavior
(Fembehave3) is obtained by summing items 32 to 36. Cronbach’s alpha
assessing internal consistency of the Femscore3 and Fembehave3 was .84
and .86 respectively, indicating good internal consistency of the scales
(Pallant, 2013). Cronbach’s alpha for the subscales ranged from .58 to .86,
with only four of the subscales demonstrating acceptable internal
consistency (α ≥ .7). Low internal consistency indicates that the items on
the scale are not measuring the same underlying construct (Pallant, 2013).
Internal consistency can be negatively impacted by low number of
questions, poor interrelatedness of items, and heterogonous constructs
(Tavakol & Dennick, 2011); therefore, the subscales with lower than
acceptable internal consistency may contain items that do not measure the
same, intended constructs. See Table 4.6 for additional psychometrics for
the FPS3. Low internal consistency for some subscales is common in
previous research which uses the FPS3 (Henley, Spalding, & Kosta, 2000).
Henley, Spalding, & Kosta (2000) suggested using a larger sample size (a
sample of 209 was used in the development of the short version of this
instrument) in order to increase reliability for the subscales; however, using
a sample size of 305 did not increase subscale reliability. Due to low internal
consistency of some of the scales and small sample sizes for men and racial
minorities, subscale scores for the FPS3 were not used in analysis.
Self-Efficacy
The General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem,
1995) measured beliefs about general self-efficacy. The GSE is a 10 item
scale that assesses one’s belief in their ability to respond to difficult
situations and cope with associated obstacles. The 4 point Likert scale
measures the extent to which each item applies to the participant, ranging
from not at all true (1) to exactly true (4). Participant total scores ranged
from 21 to 40 (M = 33.53, SD = 4.27), with higher scores indicating higher
general self-efficacy. Cronbach’s alpha assessing internal consistency of the
GSE was .86, indicating good internal consistency of the scale (Pallant,
2013).
Table 4.6
Psychometric properties for the FPS3
M SD α Range
Composite subscale
Femscore3
114.71
22.45
.86
28-
153
Perspective subscales
Conservative
11.45
5.38
.70
5-30
Liberal Feminist 23.39 4.84 .66 8-35
Radical Feminist 16.21 6.33 .77 5-33
Socialist Feminist 17.22 6.39 .78 5-33
Cultural Feminist 18.70 5.00 .58 5-32
Woman of Color/Womanist 23.45 7.06 .86 5-35
Data Analysis
The following section reviews the results of preliminary analyses for
the data, as well as the results of the analyses for the two research questions
and their accompanying hypotheses. All data was analyzed using the
Statistical Package for the Social Sciences (SPSS, Version 23). An alpha
level of .05 was utilized to confirm that 95% of the variance was due to the
relationship between variables, not due to sampling error (Fink,
2013).
Statistical Assumptions
I conducted preliminary analyses to test for missing data, outliers, and
assumptions. Missing data existed for participants who completed the
survey. See Table 4.7 for missing data by assessment. Initial examination of
missing data revealed eight participants who did not complete all items on
the assessments. Therefore, sum scores could not be calculated for those
corresponding assessments. Of the eight participants who did not complete
the survey, five participants discontinued the survey before completing the
first assessment (FPS3). The remaining three participants completed the
FPS3 and did not complete the other two assessments. Because the majority
of these participants were missing total scores for the three assessments, and
less than 5% of the data was missing (Sterner, 2011), listwise deletion
excluded these cases from all data analysis.
Table 4.7
Missing Data by Assessment
Complete Missing (%) Total
SIF 311 8 (2.5%) 319
FPS3 314 5 (1.6 %) 319
GSE 311 8 (2.5%) 319
Scatterplots and normal probability plots tested for assumptions of
normality, linearity, and homoscedasticity, with their residuals identifying
any potential outliers. No assumptions were violated for the two types of
analyses used in this study: standard multiple regression and two-way,
factorial MANOVA. Prior to each analysis, an examination of univariate and
multivariate outliers is presented.
Finally, I conducted a Pearson correlation to determine possible
covariates. I utilized participant age, years of education, and number of
credit hours as intended covariates. The Pearson correlation revealed no
relationship between the intended covariates and SIF, Fembehave3, and
Femscore3 total scores. However, a relationship existed between the three
intended covariates and GSE total scores (see Table 4.8). Therefore, I
controlled for GSE total scores by age, years of education, and credit hours
by conducting two separate ANCOVAs. Gender and minority served as
independent variables and GSE total scores served as the dependent
variable.
Table 4.8
Pearson Correlations for Participant GSE Scores
Age
Years of
Education Credit Hours
GSE .23* .05* .04*
Note. * denotes significance at the .01 level
The first ANCOVA tested for differences in male and female
participants’ GSE scores while controlling for age, years of education, and
credit hours. Levene’s test revealed no violation of the assumption of
homogeneity. ANCOVA results indicated no significant differences existed
between men and women, F = .000, observed
power = .056. Participants’ GSE scores did not differ significantly by
gender. See Table 4.9 for means and standard deviations.
Table 4.9
Descriptive statistics for GSE by Gender
Men Women
M SD M SD
GSE 33.58 4.48 33.51 4.21
The second ANCOVA tested for differences in White and non-White
participants’ GSE scores while controlling for age, years of education, and
credit hours. Levene’s test revealed no violation of the assumption of
homogeneity. ANCOVA results indicated no significant differences existed
between White and non-White participants, F (1, 299) = .073, p = .787, 𝜂𝜌2
= .000, observed power = .058. Participants’ GSE scores did not differ
significantly by minority status. See Table 4.10 for means and standard
deviations.
Table 4.10
Descriptive Statistics for GSE by Minority Status
White Non-White
M SD M SD
GSE 33.47 4.16 33.72 4.65
Results of Data Analysis
Research Question 1
The first research question asks: What relationship exists among
feminist selfidentification, feminist perspectives, and self-efficacy?
Specifically, can feminist selfidentification, as measured by the Self-
Identification as a Feminist Scale (SIF, Szymanski, 2004), and feminist
perspectives, as measured by the Feminist Perspectives Scale – Short Form
(FPS3; Henley, Spalding, & Kosta, 2000) predict self-efficacy, as measured
by the General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem, 1995)?
For the standard multiple regression utilized to answer this question,
no violations of normality, linearity, or homoscedasticity existed. No
univariate outliers existed on the normal probability plot or scatterplot, as
residual values fell between -3.3 and 3.3 (Pallant, 2013). An investigation of
Mahalanobis distances revealed no multivariate outliers, as no cases
exceeded the chi-square critical value (16.27) associated with three
independent variables. I tested for multicollinearity by examining the
collinearity statistics of tolerance and variance inflation factor (VIF).
Multicollinearity occurs when two or more predictor variables are highly
correlated (r ≥ .9). In other words, one variable could be predicted by the
other variable or variables (Pallant, 2013), creating redundancy and
interfering with determining unique predictors. Because tolerance values
were greater than .10 and VIF values were below 10, no violations of
multicollinearity occurred (Pallant, 2013).
I conducted the following regression analysis utilizing feminist
identification (total SIF score) and feminist perspectives (Fembehave3 and
Femscore3 total scores) as predictor variables. Self-efficacy (total GSE
score) served as the dependent variable. Results for research question one
and the associated hypotheses are presented below.
Hypothesis 1A. The first hypothesis postulated that feminist
identification ratings and feminist perspectives ratings would be positively
correlated with self-efficacy ratings. The following table (Table 4.11)
presents the Pearson correlations of feminist identification, feminist
perspectives, and self-efficacy. As the table displays, a weak, positive
relationship was found for only one variable (i.e., Fembehave3). Therefore,
hypothesis 1A was partially supported.
Hypothesis 1B. The second hypothesis postulated that higher ratings
for feminist self-identification and feminist perspectives would predict
higher self-efficacy for young adults. The following table (Table 4.12)
shows the predictive ability of each independent variable.
Table 4.11
Pearson Correlations for Feminist Identification, Perspectives, and Self-
Efficacy
Self-
Efficacy
Feminist
Identification
Feminist
Attitudes
Feminist
Behavior
Self-Efficacy 1.00
Feminist
Identification
-0.05 1.00
Feminist
Attitudes
-.07 .65 1.00
Feminist
Behavior
.20* .19 .19 1.00
Note. * denotes significance at the .05 level
Table 4.12
Predicting Relationship between Feminism and Self-Efficacy
B SE B t p
Constant 30.34 1.62 18.78 .00
Feminist
Identification -.02 .07 -.25 .80
Feminist Attitudes -.02 .01 -1.39 .17
Feminist Behavior .23 .06 3.82* .00
Note. * denotes significance at the .05 level
Only one predictor variable, feminist behavior (Fembehave3) was a
significant predictor of self-efficacy at the .05 level (p < .001). Therefore,
the hypothesis was partially supported. The r2 value indicates that
approximately 8% of the variance in self-efficacy scores can be accounted
for by the given model (see Table 4.13). The r2 value (r2 = .05) also indicates
a small effect size, meaning that any effect smaller than .10 indicates the
relationship has little practical significance (Cohen, 1992).
Table 4.13
Model Summary
Model R R2 Adjusted
R2
Std. Error of
the Estimate ΔR2
1 .23 .05 .04 4.18 .05
In order to determine how well the independent variables predict the
dependent variable, I assessed values presented in the ANOVA table (Table
4.14). Overall, the model was significant, F (3, 301) = 5.44, p < .001.
However, feminist behavior was a unique, significant predictor of general
self-efficacy. Therefore, higher ratings for feminist behavior are more likely
to predict higher ratings for general self-efficacy. Further, when scores for
feminist behavior are predicted to increase by one, scores on self-efficacy
would increase by .23.
Table 4.14
ANOVA Table
Model
Sum of
Squares
284.28
df Mean
Square F Sig.
Regression 3 94.76 5.44 .001
Residual 5245.74 301 17.43
Total 5530.01 304
Research Question Two
The second research question asks: What differences exist among
various races (e.g., White, African American, Hispanic) and gender (i.e.,
male, female), as measured by the demographics questionnaire, between
feminist self-identification, as measured by the Self-Identification as a
Feminist Scale (SFI; Szymanski, 2004), feminist perspectives, as measured
by the Feminist Perspectives Scale – Short Form (FPS3; Henley, Spalding,
& Kosta, 2000), and self-efficacy, as measured by the General Self-Efficacy
Scale (GSE; Schwazer & Jerusalem, 1995)?
For the two-way factorial MANOVA utilized to answer this question,
no violations of normality or linearity existed. Scatterplots revealed no
univariate outliers; therefore the assumption of linearity was not violated
(Pallant, 2013). For research question two, there are four dependent
variables: (a) feminist attitudes scores; (b) feminist behavior scores; (c)
general self-efficacy scores; and (d) feminist identification scores. I tested
for multicollinearity and singularity by examining the correlations between
the dependent variables. Because none of the correlations for dependent
variables were greater than .70, the assumptions of multicollinearity and
singularity were not violated (Pallant, 2013). The assumption of
homogeneity was not violated, as the significance value for Box’s M test
was .56. Because the significance was greater than .05, this means there are
no significant differences between the covariance matrices across groups
(Pallant, 2013). Levene’s test of equality of variances revealed a violation of
the assumption of equal variances for gender on GSE scores, F (3, 301) =
3.075, p = .028. I utilized Pillai’s Trace instead of Wilk’s Lambda and
adjusted the alpha level to .01 to account for violations of equal variances
(Pallant, 2013).
I conducted the following factorial MANOVA utilizing gender and
minority status as independent variables. The four dependent variables were:
(a) feminist attitudes scores; (b) feminist behavior scores; (c) general self-
efficacy scores; and (d) feminist identification scores. Results for research
question two and the associated hypotheses are presented below.
Hypothesis 2A. The first hypothesis postulates that female
participants will have higher ratings than male participants for feminist
identification, feminist perspectives (made up of feminist behaviors and
feminist attitudes), and self-efficacy. Results indicated a main effect for
gender and the combined dependent variables, F (4, 298) = 4.74, p = .001;
Pillai’s Trace = .06; 𝜂𝜌2 = .06. Table 4.15 illustrates the between subjects
statistics for gender.
Table 4.15
Between Subjects Statistics by Gender
df F p η2
Feminist Attitudes 1 8.09 .005* .03
Feminist Behavior 1 11.03 .001* .04
Self-Efficacy 1 .116 .734 .00
Feminist Identification 1 9.38 .002* .03
Note. * denotes significance at the .05 level
An inspection of mean scores revealed that women (n = 228) reported
significantly higher scores for feminist behaviors, feminist attitudes, and
feminist identification than men (n = 77). Therefore, the first hypothesis is
partially supported. Table 4.16 displays descriptive statistics for gender and
the four dependent variables.
Table 4.16
Differences between Gender and Dependent Variables
n M SD
Feminist attitudes*
Female 228 105.72 22.04
Male 77 95.99 26.56
Feminist behavior*
Female
228
23.21
3.79
Male 77 21.30 4.30
Self-efficacy
Female
228
33.51
4.20
Male 77 33.58 4.48
Feminist identification*
Female
228
10.92
4.09
Male 77 8.49 4.58
Note. * denotes significance between groups at the .05 level
Hypothesis 2B. The second hypothesis postulates that White
participants will have higher ratings than non-White participants for feminist
identification, feminist perspectives, and self-efficacy. Results indicated no
main effect for minority status and the combined dependent variables, F (4,
298) = 1.19, p = .32; Pillai’s Trace = .02; = .02. Therefore, the second
hypothesis is not supported by the data. In other words, White participants
did not have higher ratings than non-White participants for feminist
identification, feminist perspectives (made up of feminist behaviors and
feminist attitudes), and self-efficacy. Tables 4.17 illustrates the between
subjects statistics for minority status. Additionally, there was no significant
interaction between gender and minority status, F (4, 298) = 1.00, p = .41.
Table 4.17
Between Subjects Statistics for Minority Status
df F p η2
Minority Status
Feminist Attitudes
1
3.84
.05**
.01
Feminist Behavior 1 .00 .99 .00
Self-Efficacy 1 .36 .55 .00
Feminist Identification 1 2.30 .13 .01
Gender*Minority Status
Feminist Attitudes
1
.02
.88
.00
Feminist Behavior 1 .10 .75 .00
Self-Efficacy 1 .20 .65 .00
Feminist Identification 1 1.84 .18 .01
Note. ** p value was .051 before being rounded
Because the effect size values for minority and for gender*minority status
were less than
.10, the strength of the relationships was not significant.
Summary
The purpose of conducting the current study was to determine the
relationship among feminist identification, feminist perspectives, and self-
efficacy for young adults as it relates to counselor education. Additionally, I
sought to identify what differences existed between gender and minority
status for the aforementioned constructs. A total of two research questions
and four hypotheses were utilized to understand the relationship between
feminist identification, feminist behavior, feminist attitudes, and general
selfefficacy. Survey data was collected from 305 participants who are
enrolled as graduate or undergraduate students at the University of South
Carolina. I utilized standard multiple regression and two-way factorial
MANOVA to analyze the data. Results only partially supported two of the
four hypotheses. Regression analysis only partially support of research
question one and hypothesis 1B, identifying feminist behavior as the only
significant predictor of self-efficacy. Factorial MANOVA revealed a
significant main effect between gender and three of the four dependent
variables; however, no significant main effect was identified between
minority status and the four dependent variables.
CHAPTER 5
DISCUSSION
The current study examined the relationship among feminist
identification, feminist perspectives, participant demographic factors, and
self-efficacy for young adults who are enrolled as either undergraduate or
graduate students at the University of South Carolina. Three hundred five
participants completed all three assessments and the demographics
questionnaire. The study aimed to (a) examine relationships among feminist
self-identification, feminist perspectives, and self-efficacy; and (b) explore
existing differences for race and gender on feminist self-identification,
feminist perspectives, and self-efficacy in undergraduate and graduate
students. These aims resulted in two research questions. Research question
one examined the predictive ability of feminist identification and feminist
perspectives (which consists of feminist behaviors and feminist attitudes) on
self-efficacy. Research question two examined differences among race and
gender for feminist identification, feminist perspectives, and selfefficacy.
Following is a brief discussion of the study results, limitations to the study,
and implications for practice and future research.
Overview of Findings
Feminist Identification and Perspectives as Predictors of Self-Efficacy
The first research question asked: What relationship exists among
feminist selfidentification, feminist perspectives, and self-efficacy?
Specifically, can feminist selfidentification, as measured by the Self-
Identification as a Feminist Scale (SIF, Szymanski, 2004), and feminist
perspectives, as measured by the Feminist Perspectives Scale Short Form
(FPS3; Henley, Spalding, & Kosta, 2000) predict self-efficacy, as measured
by the General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem, 1995)?
Two hypotheses were identified for research question one. Both hypotheses
utilized selfefficacy total scores as the dependent variable, with feminist
self-identification, feminist attitudes, and feminist behaviors as predictors.
The first hypothesis for research question one postulated that feminist
identification and feminist perspectives (consisting of feminist behaviors and
feminist attitudes) would be positively correlated with selfefficacy. The
second hypothesis postulated that higher feminist identification and feminist
perspectives scores would predict higher self-efficacy scores. I conducted a
standard multiple regression to examine the predictive relationship among
the four aforementioned constructs.
Results indicated that neither feminist identification nor feminist
attitudes were significant predictors of self-efficacy. However, feminist
behavior was identified as a significant predictor of general self-efficacy,
with Pearson correlations indicating a small, positive correlation between
feminist behavior and self-efficacy. This relationship indicates that higher
scores for feminist behavior are more likely to predict higher levels of self-
efficacy. Although the strength of the relationship between feminist behavior
and self-efficacy does not indicate practical significance, previous research
has typically included advocacy in its examination of the relationship
between feminism and selfefficacy. While prior research has demonstrated a
positive relationship between feminist perspectives and self-efficacy (Eisele
& Stake, 2008), this may be partially explained by the positive relationship
between nontraditional gender role attitudes and self-esteem (Szymanski,
2004), which are components of feminist perspectives and self-efficacy. If
an individual feels more empowered (empowerment is also a component of
feminism) about an issue, they may feel more encouraged and more capable
(i.e., self-efficacy) to engage in advocacy for the corresponding group
(Eisele & Stake, 2008; Zimmerman, 1995). By not including advocacy as a
construct in this study, an important piece may have been left out of the
equation to determine relationships between feminism and selfefficacy.
However, while advocacy is a tenet of feminist behavior, feminist behavior
encompasses other components. Therefore, the current research contributes
new knowledge about the relationship between feminism and self-efficacy.
Further, the current results may highlight a more complex understanding of
contributors to general self-efficacy.
The current study indicated no relationship between feminist
identification and self-efficacy. This finding does not support previous
research, which suggested that feminist self-labeling bridges the relationship
between feminist perspectives and selfefficacy (Eisele & Stake, 2008).
Despite agreeing with feminist perspectives, men and women are often
reluctant to identify as feminists, either publicly or privately (Kelly, 2015).
This reluctance may be due to fear of being perceived negatively or
inaccurately by others (Anderson, 2009; McCabe, 2005; Roy, Weibust, &
Miller, 2007; Williams & Wittig, 1997; Zucker, 2004), or due to a lack of
education about feminism (Kelly, 2015). Results from the current study
support such research, with fewer participants identifying as a feminist
(54.8%), despite acknowledging feminist values as important (67.5%) and
supporting feminist goals (70.2%). Further, over half (54.8%) of participants
considered themselves feminists, but just over one third (38.7%) identified
as such to others. These findings are comparable to previous research on
discrepancies between feminist identification and supporting feminist
perspectives. In a study conducted by Anderson (2009), the majority of men
in the study (59.7%) did not identify as feminists, but 32.6% of men said that
they agreed with some feminist objectives but did not call themselves
feminists (which was the most popular choice of female participants in the
study [45.4%]). Including an assessment of why participants did or did not
identify (publically or privately) would have provided more depth to the
current study. Additionally, knowing whether or not someone has been
exposed to or has a knowledge of feminism may also be important in
explaining choices about feminist self-labeling, as Zucker (2004) suggested
that being exposed to feminism in various contexts influences feminist
identification.
Race and Gender Outcomes
The second research question asked: What differences exist among
various races (e.g., White, African American, Hispanic) and gender (i.e.,
male, female), as measured by the demographics questionnaire, between
feminist self-identification, as measured by the Self-Identification as a
Feminist Scale (SFI; Szymanski, 2004), feminist perspectives, as measured
by the Feminist Perspectives Scale – Short Form (FPS3; Henley, Spalding,
& Kosta, 2000), and self-efficacy, as measured by the General Self-Efficacy
Scale (GSE; Schwazer & Jerusalem, 1995)? Both hypotheses utilized
feminist identification, feminist behaviors, feminist attitudes, and self-
efficacy as dependent variables. However, hypothesis one included gender
as the independent variable, while hypothesis two utilized race as the
independent variable. The first hypothesis postulated that women would
have higher scores than men on all assessments. The second assessment
postulated that White participants would have higher scores on all
assessments compared to nonWhite participants. I conducted a two-way
factorial MANOVA to examine outcomes between gender and race for
feminist identification, feminist perspectives (attitudes and behaviors), and
self-efficacy.
Results indicated statistically significant differences in feminist
attitudes, feminist behaviors, and feminist identification between men and
women. On average, women reported higher scores for all three constructs.
These findings underscore previous research on minority status, which
suggests that minorities (e.g., women) may be more likely to support groups
outside of their own because of their own status as a minority (Kelly &
Breinlinger, 1995; Wiley et al., 2012). Because postmodern feminist
perspectives include attitudes of inclusion and advocacy behaviors, these
findings are consistent with those of earlier research. Higher scores for
women on the FPS3 suggest stronger gender consciousness than that of
males in this study, challenging previous research in which women have
demonstrated weaker understanding about how gender affects one’s
experiences in the world (Aronson, 2003). Research conducted by Zucker
(2004) attempts to highlight the importance of feminist consciousness, with
feminists in the study rating higher on feminist consciousness and
experiences of sexism compared to non-feminists and egalitarians (i.e.,
participants who agreed with feminist ideas but did not label themselves as
feminists). Non-feminist and egalitarian participants may have rated lower
on experiences of sexism due to their lower feminist consciousness. By
being unaware of the role that gender plays in one’s experiences in the
world, non-feminist and egalitarian participants may have been unable to
recognize instances of sexism when they are actually occurring. If
individuals are unable to identify occurrences of sexism, they may also be
unable to identify when such experiences are impacting their health,
attributing their symptoms to other factors.
While women reported higher total scores for feminist identification,
examining data for the four items that comprise the SIF reveals new
knowledge about gender differences for various facets of feminist identity.
For example, over half of men (55.8%) identified feminist values as being
important to them (item 3). The same percentage of men (55.8%) also
reported supporting the goals of the feminist movement (item 4).
Additionally, the majority of women (61.4%) and over one third of men
(35.1%) reported identifying as a feminist (item 1). This is in contrast to
results of Anderson’s (2009) study, in which less than 1% of men identified
as feminist either publicly (to others) or privately (to themselves) compared
to nearly 7% of women. This suggests that more men and more women may
be claiming the feminist label both publicly and privately, although more
than half (53.2%) of men and over one third (31.6%) of women in this study
reported that they did not identify as a feminist to other people (item 2). The
adoption of the private feminist label was significantly higher for women in
this study compared to women in previous research (7%; Anderson, 2009).
For women, this item revealed the highest percentage for disagreement of
any of the four items.
A discrepancy between public and private feminist identification was
evident across gender, with more male and female participants identifying
privately (M = 54.8%) as a feminist than identifying as a feminist publicly
(M = 38.7%). This finding echoes previous research that both men and
women may be reluctant to identify as a feminist to others, even if they
consider themselves to be a feminist (Kelly, 2015). This finding supports
previous research that suggests the fear of being perceived negatively by
others can prevent the adoption of the label (Anderson, 2009; McCabe,
2005; Roy, Weibust, & Miller, 2007; Williams & Wittig, 1997; Zucker,
2004). Although both men and women face this fear, it may be particularly
influential for men due to notions of masculinity being arguably more rigid
than notions of femininity (Ratele, 2013).
Results indicated no significant differences in feminist identification,
feminist attitudes, or feminist behaviors between White and non-White
participants. This finding challenges speculation from previous research that
minorities may not support feminism due to its origination in White, middle-
class culture (Aronson, 2003; Hunter & Sellers, 1998; Williams & Wittig,
1997; Zucker, 2004). This may be due, in part, to the current wave of
feminism being more inclusive and diverse in its views and expression of
feminist identity (Aronson, 2003; Heywood & Drake, 1997). Conversely, no
significant differences for feminist identification, feminist perspectives, and
self-efficacy also does not support previous research which suggests that
minority status makes an individual more likely to support minority groups
outside of their own (Hunter & Sellers, 1998; Kelly & Breinlinger, 1995;
Wiley et al., 2012). However, due to the low numbers of minority
participants, the results from this study may not be an accurate
representation of non-Whites’ views on feminism and self-efficacy. Previous
research has also suggested that where an individual is in their racial identity
impacts their views on feminism (Martin & Hall, 1992; Myaskovsky &
Wittig, 1997). Martin and Hall (1992) purported that the further along an
African American woman is in her racial identity, the more likely she is to
view feminism as important. Similarly, Myaskovsky and Wittig (1997)
found that African American women who had stronger racial identities were
more likely to have been exposed to feminism, to recognize racial
discrimination, and to support collective action. Therefore, including a
measure of racial identity for racial minorities may aid in explaining
differences between White participants and participants of color in their
feminist identification, support of feminist perspectives, and levels of self-
efficacy.
Limitations of the Study
The current study contained limitations to both internal and external
validity. Internal validity refers to the confidence that an outcome was the
result of the studied variable, while external validity refers to the extent of
generalizability of results to the population (Rubin & Babbie, 2011).
Selection bias was a threat to internal validity, as participants who chose to
complete the online version of the study may have chosen to do so because
they have strong views, either positive or negative, about feminism. In other
words, students who were willing to participate may be more or less
interested in the topics of feminism and self-efficacy than the general
population. This may have particular meaning for participants who were
recruited through passive strategies (and completed the online version), who
may have participated because they had higher motivation to do so. While
motivation to participate in the study may have been due to polarized views
about feminism, additional instructor incentives may have also played a role
in motivating students to participate. Some instructors offered additional
incentives (beyond the gift card drawing offered to all participants) to their
students who chose to participate in the study.
The use of convenience sampling to obtain participants was a threat to
the external validity of the study. This sampling method can bias results
because it is impossible to determine if the sample is representative of the
overall population and may have led to an inadequate representation of
groups in my sample (Rubin & Babbie, 2011).
However, since little is known about how men and minorities feel
about the variables that will be examined in this study, convenience
sampling may provide insight into whether or not a problem exists in a
biased sample. For example, there was a significant difference between men
and women for feminist identification and feminist perspectives, which
supports previous research comparing men’s and women’s support of
feminism. Further, because convenience samples are typically already biased
(Rubin & Babbie, 2011), uncovering the perspectives on feminism of men
and minorities in a biased sample can provide valuable information into how
to proceed in future studies with these two groups (Rubin & Babbie, 2011).
In other words, since no significant differences existed between White and
non-White participants on the outcome variables, it may be unlikely that
significant differences would exist in an unbiased sample (Rubin & Babbie,
2011). However, because there were unequal groups for White and non-
White participants, this may not be the case.
While the sample obtained for this study appeared demographically
similar to the population at the University of South Carolina, results may be
difficult to generalize to similar populations across the country. Specifically,
because the University is located in the Southern region of the United States,
participants’ formative experiences surrounding feminism and self-efficacy
may not be comparable to those of individuals living in other geographic
regions. This may be particularly true for the participants who identified as
racial minorities, as the South has historically held more negatively biased
views of such groups.
Diversity within the sample may have also impacted the results. While
my original goal was to obtain equal numbers for racial and gender
categories for this study, participation from minorities and men was
significantly lower than that of Whites and women. This is consistent with
previous research on feminism and self-efficacy, which has also had lower
representation of men and minorities. Because of the overrepresentation of
Whites and women in this study, it is difficult to draw conclusions about
male and minority participants’ views on the measured constructs.
Additionally, results indicating no difference between White and non-White
participants for research question two are subject to Type II error due to the
low, unequal numbers of minority participants. A Type II error occurs when
the researcher fails to reject a false null hypothesis (Rubin & Babbie, 2011),
meaning that the researcher reports finding no significant differences
between groups when such differences may actually exist. Higher, equal
numbers between groups also contribute to higher power (Rusticus &
Lovato, 2014) and larger effects, leading to less likelihood of a type II error
(Balkin & Sheperis, 2011). In other words, having higher, equal numbers of
non-White participants may have changed the results for research question
two, indicating a significant difference between White and non-White
participants on the measured constructs. Therefore, future research should
utilize probability sampling methods (e.g., stratified random sampling) in
order to ensure equal representation of groups. Further, the decision to
combine non-White participants into one group did not allow for
examination of individual racial category data. However, some categories
(e.g., Native American, Pacific Islander) did not have enough participants to
perform data analysis, contributing to the decision to combine the groups.
Some items on the FPS3 did not represent more modern, diverse
feminist perspectives, potentially impacting results. For example, gendered
items on the FPS3 positioned the male as the perpetrator and the woman as
the victim of the oppression; however, modern feminist perspectives support
the idea that like women, men are also negatively impacted by gender
inequality and oppression (Mohanty, 1988). Further, some items on the
FPS3 made assumptions about participants’ lives or intentions that were
rooted in traditional gender norms. For example, item 31 stated, “My
wedding was, or will be, celebrated with a full traditional ceremony,”
assuming that participants desired to get married if they were not already.
Similar assumptions were found in item 33 and item 35, regarding the
assumption of religion and of desire for children, respectively. Lastly, item
36 stated, “I often encourage women to take advantage of the many
educational and legal opportunities available to them,” which assumes that
all women have equal access to multiple educational and political resources.
This assumption is related to the intersectionality of identities, which is a
central focus of modern feminist perspectives.
Finally, subscores for the FPS3 were not used in data analysis. Only
four of the six subscales demonstrated acceptable internal consistency (α
≥ .7). Additionally, the small sample size for men and racial minorities
prevented the use of subscale scores. Differences may have existed between
men and women, and between White and nonWhite participants for feminist
perspectives. However, inclusion of additional variables in my analyses
would have decreased the observed power.
Implications
Counselor Education
The aim of this study was to identify the significance and role of
feminism in determining self-efficacy in young adults. While results
identified feminist behavior as a predictor of self-efficacy, neither feminist
identification nor feminist attitudes correlated with levels of general self-
efficacy within the sample. However, feminist identification and feminist
perspectives were significantly higher for women compared to men. This
finding is noteworthy, considering the typical composition of Master’s level
counseling programs. Because these programs generally contain more
women than men, understanding that feminism may be a significant part of
their identities is important when thinking about program curriculum.
Including more knowledge about feminist theory in coursework may allow
faculty to cater to an already present identity within students, as the majority
of participants in this study identified as supporting feminist values and
goals. For example, dedicating an entire course to feminism, or creating a
hybrid multiculturalism/feminism course may help increase students’
knowledge feminist theory. Having feminist speakers, counselors, and
supervisors may also provide more context to and a better understanding of
feminist theory for students. The rationale for this suggestion is that students
may have no prior exposure to feminists (or, more likely, they may not have
been exposed to individuals who publicly identify as feminists). This may
also provide vicarious experiences (modeling; Bandura, 1977) with
feminism for students; by being exposed to successful feminist clinicians,
students may feel that they can also exhibit feminist behaviors successfully.
Because modeling is a contributing factor for self-efficacy, providing
opportunities for exposure to feminism may increase their confidence in
their ability to engage in feminist behaviors. Additionally, interacting with
positive feminist models may challenge pre-existing negative or inaccurate
stereotypes they have about the feminist label. Previous research has
suggested that even brief exposure to positive portrayals of feminists can
positively influence perceptions of feminism, as well as increase the desire
to participate in collective action for women (Roy, Weibust, & Miller, 2007;
Wiley et al., 2012). Further, research has found that selfidentifying with a
group increases the likelihood of activism on behalf of that group
(Leaper & Arias, 2011; Wiley et al., 2012; Zucker, 2004).
Incorporating more education about feminism in counselor education
curriculum may also help educate individuals who cite a lack of knowledge
about feminism as the reason they do not identify as such (Kelly, 2015).
Further, requiring a class-wide advocacy project based on feminist issues,
along with a corresponding research paper, may also aid in (a) gaining
accurate knowledge about feminist issues (both through research and
practical experience); (b) encouraging feminist identification through the
impact of the group experience (i.e., feminist labeling becomes the social
norm of the class); and (c) inciting student interest in and understanding of
advocacy in the field. If students realize that they can successfully engage in
advocacy (i.e., performance accomplishments; Bandura, 1977), they may be
more likely to persist in their advocacy behaviors. Further, continued
mastery experiences with advocacy and other feminist behaviors may also
increase their perceived self-efficacy.
For counselors, supervisors, and counselor educators, understanding
the relationship among such factors will not only help them better
understand the worldview of clients who do not ascribe to traditional gender
roles, but it may also encourage selfreflection on the impact that gender
stereotyping and sexism can have on interactions with and treatment of
clients and counselors-in-training. This is relevant to the results of this study
because the majority of participants reported supporting goals of the feminist
movement and identified feminist values as being important to them.
Because feminism typically does not do not align with traditional gender
roles and norms, being undereducated about feminism puts counselors at risk
of providing biased treatment to clients based on their own values and
unintentionally supporting traditional gender concepts (Crethar, Rivera, &
Nash, 2008; DeVoe, 1990; Good, Gilbert, & Scher, 1990). This imposition
of values is addressed in the ACA Code of Ethics (2014), which states that
counselors must be aware of their own values, resist imposing those values,
and “seek training in areas in which they are at risk of imposing their own
values onto clients” (A.4.b). DeVoe (1990) found that participating in
advocacy efforts and feminist consciousness raising helped make counselors
more aware of feminist issues, resulting in increased insight into power
differentials between men and women and awareness of how sexist values
can negatively impact relationships. Therefore, counselors, supervisors, and
counselor educators have an ethical responsibility to become more educated
on feminism in order to better inform their work with clients, supervisees,
and students. Finally, the common threads between feminist theories and
multicultural counseling theories (Crethar, Rivera, & Nash, 2008; Goodman
et al., 2004), which are woven throughout counselor education programs,
make the incorporation of feminism a logical next step for curriculum.
Counseling Practice
Results from the current study indicated that the majority of
participants were in agreement of supporting feminist goals and of viewing
feminist value as being important to them. These findings have strong
implications for counselors in their work with young adults. For example,
because power imbalance is a significant focus of feminism, young adult
clients who ascribe to feminist perspectives may value a more egalitarian
relationship with their counselors. If left unacknowledged, this power
differential can negatively impact the client-counselor relationship by
ignoring the client’s expertise on their own lives. Additionally, counselors
who may be unaware of or undereducated about feminist perspectives risk
making inaccurate interpretations of client problems, imposing their own
traditional values, implementing techniques and strategies that reflect the
counselor’s perspectives instead of the client’s, and establishing goals that
do not meet client needs and perspectives (Ivey et al., 2011). In order to
better meet the needs of feminist young adult clients, counselors do not have
to be experts in feminist therapy; however, being informed about the tenets
of feminism and feminist therapy can prevent some of the aforementioned
issues.
Feminism’s view on oppression as a contributor to physical and
mental health is also important for counselors to understand. Through the
feminist lens, client issue arise from experiences of oppression and power
imbalance (Brady-Amoon, 2011); therefore, feminist clients may reject the
idea of diagnosis altogether. In other words, they may identify their
symptoms as being normal responses to oppression, not as psychopathology.
To the counselor who is undereducated about feminism, this may present as
client resistance to treatment, further impacting how the counselor interacts
with the client in regards to goal setting and treatment planning. Counselors
are not expected to divorce their theoretical orientations; however, it is their
professional responsibility to utilize techniques and strategies that best fit a
client’s needs. Being knowledgeable about feminist therapy, which
addresses such issues through collaborative, nonhierarchical counselor-client
relationships, can aid counselors in aligning with their feminist client’s
needs and worldviews. Awareness of feminism and feminist perspectives
can also aid in illuminating similarities between feminist and multicultural
theories in counseling, which are woven throughout counselor education
programs. Combining multicultural and feminist counseling theories may
help counselors better address issues of social justice, privilege and
oppression in their work with clients. In fact, Fassinger and Gallor (2006)
cite being informed about both perspectives as a necessary prerequisite for
social justice and advocacy work with clients.
Research
In order to gain more understanding of the relationship between
feminism and self-efficacy, more research is needed on specific factors that
contribute to feminist identity. While there is previous qualitative research
on feminist identity, scant research exists on how such influential factors
may interact with general self-efficacy beliefs. Results also revealed higher
mean scores for supporting feminist goals (M = 70.2%) than for public (M =
38.7%) and private (M = 54.8%) feminist identification. Therefore, more
research is needed to identify what has encouraged or prevented individuals
from adopting the label despite adopting feminist values and perspectives.
For example, qualitative inquiry into contributing factors for feminist
identification may provide deeper explanations about differences between
groups, as well as between public and private identification. Additionally,
examining the relationship among individual SIF items, feminist attitudes,
and feminist behaviors may provide further knowledge about differences for
public and private identity, as well as for support and value of feminist
perspectives. Further, incorporating a measure of racial identity might also
help explain feminist identification reasoning in racial minorities (Martin &
Hall, 1992; Myaskovsky & Wittig, 1997).
While advocacy was not a focus of this study, understanding how self-
efficacy impacts advocacy, for self and for others, is also an area of further
research. Previous research has suggested a relationship among feminism,
self-efficacy, and advocacy, and such a relationship may provide insight into
what increases young adults’ decision to engage in advocacy efforts. Zucker
(2004) suggested that participants who identified as feminists were more
likely to engage in feminist activism, regardless of favorable conditions or
barriers to feminist identification. This may illuminate the link between
feminism and advocacy, suggesting that feminist identification is a better
predictor of social justice participation, even if individuals are faced adverse
conditions or possible negative consequences. However, because one’s
belief in their self-efficacy determines the initiation of coping behaviors, the
amount of effort presented, and the amount of time someone will continue to
exert effort when they encounter obstacles (Bandura, 1997), this seems
particularly important for feminists engaging in advocacy. Individuals who
publically identify as feminists, as well as individuals who engage in
collection action, are bound to face adverse experiences related to both the
feminist label and advocacy. This may illuminate important knowledge
regarding how the relationship among feminist identification, self-efficacy,
and advocacy affects the desire of clients, students, and supervisees to
engage in advocacy for any group or issue. Therefore, utilizing a measure of
advocacy in future research may aid in identifying whether a group identity
(e.g., feminist) or self-efficacy beliefs predicts advocacy intentions,
providing guidance for counselor educators on how to motivate students to
fulfill their ethical and professional obligation as advocates. In other words,
findings may provide suggestions for what may be more important when
discussing advocacy intentions in class: (a) an individual’s selfefficacy
levels (and how to increase them); or (b) an individual’s group
identifications (and how to strengthen them).
More research should be conducted on how men and minorities differ
in their views on feminist self-labeling and feminist perspectives. This was a
goal of the current study; however, due to the use of convenience sampling
and overrepresentation of women and White participants, generalizability of
results is low. In recruiting male and minority participants, making initial
contact with these groups at events where they are well-represented would
have allowed for more active recruitment opportunities. Although face-to-
face recruitment was the end goal for these groups, initial introductory
emails about the study were unsuccessful in peaking participation interest
from male and minority groups and organizations. Therefore, without an
invitation from male and minority groups and organization, I was never able
to obtain participation through active strategies. This resulted in inadequate
representation of both groups in the current study, and therefore results
could not be generalized to the overall population of male and minority
young adults. The numbers for minority participants in the current study was
likely most impacted by this, as active recruitment strategies are more
effective with culturally diverse populations (Yancey, Ortega, &
Kumanyika, 2006).
Utilizing probability sampling would have also increased
generalizability of results. In stratified random sampling, the population is
divided into strata (i.e., subgroups) and a desired number or proportion of
participants are selected from each stratum (e.g., White men, White women,
Minority men, Minority women) for the sample (Fink, 2013). Stratified
random sampling allows the researchers to choose a sample that represents
groups in desired proportions (Fink, 2013). Therefore, employing stratified
random sampling in future research will allow researchers to obtain equal
numbers for groups in the sample while also allowing random selection of
participants for each group. It is important to understand men’s perspectives
on feminism because they have historically been seen as the oppressor of
women. Minorities’ perspectives on feminism are needed in order to learn
more about how minority status influences advocacy, as well as to identify
whether minorities see postmodern feminism as being inclusive of and
relevant to them.
Of the assessments used in this study, not all were normed on diverse
samples. Additionally, the FPS3 may not have included feminist
perspectives that represent late third/early fourth wave feminism. Creating
an instrument which includes more current feminist perspectives may
provide a better measure for whether or not participants support feminism as
it stands today. For example, future instruments might include questions
about (a) the use of social media (e.g., Facebook, Twitter) to address forms
of oppression in an effort to include and support minority groups (Phillips &
Cree 2014); (b) the intersection of gender and other forms of oppression
(Wrye, 2009); and (c) the encouragement of male involvement in the
feminist movement (Phillips & Cree, 2014). Additionally, items should be
more focused on egalitarianism across identities, not specific to gender. For
example, gendered items on the FPS3 positioned the male as the perpetrator
and the woman as the victim of the oppression; however, modern feminist
perspectives support the idea that like women, men are also negatively
impacted by gender inequality and oppression (Mohanty, 1988). Future
instruments should include items that explore the impact of gender inequity
and masculinity on men. Further, some items on the FPS3 made assumptions
about participants’ lives or intentions that were rooted in traditional gender
norms. For example, item 31 stated, “My wedding was, or will be,
celebrated with a full traditional ceremony,” assuming that participants
desired to get married if they were not already. Similar assumptions were
found in item 33 and item 35, regarding the assumption of religion and of
desire for children, respectively. Lastly, item 36 stated, “I often encourage
women to take advantage of the many educational and legal opportunities
available to them,” which assumes that all women have equal access to
multiple educational and political resources. This assumption is related to
the intersectionality of identities, which is a central focus of modern feminist
perspectives. Therefore, items on future instruments should also be
examined for inclusivity across multiple identities. Contrasting the more
singular views of earlier feminist waves, these suggestions reflect some of
the more varied views of third/early fourth wave feminism (Aronson, 2003;
Heywood & Drake, 1997).
Conclusion
Results from the current study indicated feminist behavior as a
predictor of selfefficacy. Additionally, results indicated no differences
between White and non-White participants for feminist identification,
feminist perspectives, or self-efficacy. However, women had significantly
higher scores for the feminist identification and feminist perspectives, which
is consistent with previous research comparing men and women on these
constructs.
Results suggest counselor educators should consider incorporating
feminism into their curriculum when teaching young adults, as the majority
of participants in this study reported supporting goals of the feminist
movement and identified feminist values as important the them. For these
same reasons, supervisors and counselors who work with young adults
should also consider a place for feminism in their practice in order to better
understand the worldview of supervisees and clients who adopt such
perspectives. Researchers should continue to research factors that impact
self-efficacy, feminist identification, and feminist perspectives. Obtaining
qualitative (and perhaps observational) data on factors that contribute to
young adults’ decisions to adopt the feminist label and perspectives, both
publicly and privately, may help identify additional connections between
feminism and self-efficacy. Additionally, incorporating a measure of racial
identity may aid in identifying contributing factors to feminist identification.