Balancing Multicultural Competence With Social Justice

Abstract

To identify a multivariate configuration of feminist beliefs best associated with optimal psychological functioning, 215 mostly White college women completed an online survey measuring their feminist beliefs (Feminist Perspectives Scale, Attitudes toward Feminism and the Women’s Movement, sense of common fate, and Feminist Identity Composite) and 13 measures of well-being with liberation, encompassing individual (e.g., well-being), interpersonal (e.g., egalitarianism), and societal (e.g., collective esteem) levels of analysis. A canonical correlation analysis revealed a significant multivariate association and yielded three distinct functions: established feminism (the strongest, most positive predictor) and its opposite (antifeminism), awakening feminism (negatively linked to individual well-being), and nonfeminist but woman-identified traditionalism (with some compromised well-being). The configuration of feminist beliefs that a woman holds, does not hold, and rejects makes a difference for her psychological functioning as well as for the roles counseling psychologists adopt to achieve multicultural competence along with social justice.

Keywords

feminism well-being mental health social justice feminist therapy counseling psychology

Counseling models of feminist and womanist identity (Downing & Roush, 1985; Moradi, 2005) and of feminist therapy (American Psychological Association, 2007; Enns, 2011; Espin, 1994; Fassinger, 2000; Worell & Robinson, 1993) have highlighted feminist understandings of both women and gender issues. Feminism and a feminist identity have been explored in relation to a social justice agenda for counseling psychology (Goodman et al., 2004), to counselors’ attitudes and practices (Moradi, Fischer, Hill, Jome, & Blum, 2000), to measurement issues (Moradi & Subich, 2002a) and empirically documented expressions in general samples (largely college students; Moradi, Subich, & Phillips, 2002), and to the intersection of potential clients’ attitudes with different therapeutic approaches (Enns, 1993). The underlying assumption that ties this research and theory together and that links it to counseling psychology is its grounding in a general social justice agenda and the specific presumption that a feminist approach is beneficial for clients, counselors, and the discipline itself. As we will see below, this assumption is generally, but not always, supported; points that remain vague concern the very definition of being feminist and how these often multifaceted definitions relate to complex configurations of outcomes relevant to the practice of counseling psychology.

Definitions of Feminism and Feminist Beliefs

Although seemingly simple on first blush (e.g., Are you a feminist?), what feminism is both in theory and in lay definitions is quite complex. Feminist scholars (e.g., Jaggar & Rothenberg, 1993) have laid out various forms of feminism, and psychologists have developed a measure of six different feminist perspectives (Henley, Meng, O’Brien, McCarthy, & Sockloskie, 1998). More convoluted are lay definitions of what it means to be a feminist. When college-aged women were asked to simply talk about “feminism” and its meaning in their lives, they started by expressing confusion, but when pressed, settled on a vague reference to social equality (Rudolfsdottir & Jolliffe, 2008). Quantitative researchers commonly measure feminist identification as a yes–no response as well as more complex self-labeling, endorsement of feminist beliefs (see below), and the combination of these components (see Moradi & Yoder, 2011, for a review).

Four prominent belief measures are Henley et al.’s (1998) Feminist Perspectives Scale (FPS), Fassinger’s (1994) Attitudes Toward Feminism and the Women’s Movement Scale (FWM), Gurin and Townsend’s (1986) sense of common fate, and Fischer and colleagues’ (2000) Feminist Identity Composite (FIC). Although certainly not an exhaustive list of measures of feminist beliefs, these four scales do represent a large scope of possibilities across the extant literature covering personal, relational, and collective domains. The FPS is the broadest attitudinal measure, recording respondents’ agreement with statements that reflect each of six sociopolitical variations of feminism: conservative, liberal, radical, socialist, cultural, and womanist. The FWM explores individuals’ affective attitudes, specifically toward feminists and the women’s movement. Gurin and Townsend’s three common fate items are intended to examine how aware the respondent is of the connection between herself and all women united together as a sociopolitical unit.

The FIC is an amalgamation of two earlier scales designed to capture the five stages of Downing and Roush’s (1985) developmental model of individual feminist identity. These theoretical dimensions include passive acceptance (endorsement of traditional gender roles and denial of sexist discrimination), revelation (a time of awakening catalyzed by a series of crises in which women begin to question themselves and their roles, commonly feel guilty and/or angry, engage in inconsistent dualistic thinking, and derogate men), embeddedness–emanation (characterized by connection with select women, affirmation of one’s emerging identity, more relativist thinking, and cautious interaction with some men), synthesis (understandings that culturally prescribed gender roles matter and can be transcended, that choices can be made based on personal values not cultural prescriptions, that femininity can be celebrated without compromising one’s feminism, and that men should be evaluated as individuals not as global oppressors), and active commitment (marked by engagement in meaningful action toward creating a less sexist world). In response to criticisms that the developmental stages of the model are without longitudinal support (Hyde, 2002), the model (Henderson-King & Stewart, 1994; Worell, 1996) and the FIC (Fischer et al., 2000) have since been reframed as a series of five dimensions that together capture a woman’s beliefs at a single point in time. Although questions have been raised about how thoroughly the FIC captures these theoretical dimensions as well as about the meaning of each dimension (see Liss & Erchull, 2010), the FIC remains a widely used measure of feminist identification. Furthermore, unlike the other two belief measures, the items of the FIC are all self-relevant (referring to “I” or “my”), thus giving the FIC a specific self-reference absent from the other belief measures.

Benefits of Feminist Beliefs

A growing body of empirical evidence supports the general conclusion that endorsing feminist beliefs is favorable for girls and women, both as individuals and in their interpersonal relationships (see review in Moradi & Yoder, 2011); however, some exceptions exist, and one wonders if nonsupportive findings remain in researchers’ “file drawers.” Supportive evidence can be found among personal outcomes where holding strong feminist beliefs has been linked with high self-esteem (Fischer & Good, 1994), self-efficacy (Eisele & Stake, 2008), sexual well-being (Schick, Zucker, & Bay-Cheng, 2008), and sexual openness (Bay-Cheng & Zucker, 2007). Women’s feminist views have been shown to moderate the negative effects of gender discrimination among college students (Fischer & Good, 1994) and among women in general (Kobrynowicz & Branscombe, 1997; Landrine & Klonoff, 1997). Among undergraduate women, the relationship between media awareness and internalization of a thin ideal was moderated by feminist beliefs (Myers & Crowther, 2007), randomly assigned women exposed to feminist perspectives showed increased physical appearance satisfaction (Peterson, Tantleff-Dunn, & Bedwell, 2006), and feminist self-identification was linked to rejection of feminine norms regarding appearance and thinness and regarding the importance of romantic relationships (Hurt et al., 2007). A recent meta-analysis further confirmed the association between being feminist and positive body image, especially among older women and purposively recruited women, such as students in women’s studies (Murnen & Smolak, 2009). In addition, negative consequences are associated with not being feminist; for example, Moradi and Subich (2002a) found that nonfeminist passive acceptance was positively related to psychological distress and lower self-esteem.

Other evidence, although limited, reaches into the domain of women’s interpersonal roles and relationships. For example, feminist African American women placed greater value on blending career and family (Weathers, Thompson, Robert, & Rodriguez, 1994), and college women reporting feminist leanings described less traditional dating scripts (Rickard, 1989). Furthermore, negative relationship consequences adhere to not adopting a feminist perspective. Passive acceptance of the status quo was related to college women’s low expectations for having an egalitarian intimate relationship as well as depressed sexual assertiveness (Yoder, Perry, & Saal, 2007).

Still, some women continue to equate feminism with heterosexual disharmony (Rudman & Fairchild, 2007). Egalitarian women scored higher than both self-labeled feminists and nonfeminists in their sexual assertiveness regarding condom use (Bay-Cheng & Zucker, 2007), showing no benefit from adopting a feminist self-label. Scoring high in the revelation stage of the FIC has been linked to psychological distress (Moradi & Subich, 2002a) and lower self-esteem (Fischer & Good, 2004; Moradi & Subich, 2002b). Furthermore, among college women, high revelation scores (and only revelation scores for the FIC dimensions) were related to psychological distress and were partially mediated by general state anger (Fischer & Good, 2004). In sum, although feminism generally appears “good” for women, this conclusion is not consistently supported across all studied measures of feminism and with all psychological outcomes. This inconsistency then suggests a more complex pattern of multivariate relationships between the domains of feminist beliefs and psychological functioning—leading us to tender the use of canonical correlation to explore correlations between these two complex sets of variables.

Counseling Outcomes: Well-Being With Liberation

At the heart of counseling psychology lie the goals of maximizing psychological functioning and minimizing psychological distress (American Psychological Association, 1999). Given our interests in the benefits of feminism and in taking a counseling approach (Lopez et al., 2006), we narrowed our focus to concentrate on generally positive indicators of (self-reported) psychological functioning. Drawing on the social justice agenda that links feminist (Worell, 2001) with counseling (Speight & Vera, 2008) psychologies through both critical (Prilleltensky & Prilleltensky, 2003) and positive (Lopez & Edwards, 2008) psychology, we broadly defined these outcomes as promoting well-being with liberation (both from oppression and to be empowered) and as operating across three spheres: the personal, relational, and collective (Prilleltensky & Prilleltensky, 2003). Given our social justice agenda, we narrowed our definition of well-being to focus specifically on eudaimonic (as opposed to hedonic or subjective emotional) well-being through which individuals seek personal growth, meaning, and purpose in their lives (Lent, 2004). Following Lent’s (2004) review, we operationally defined this construct with the Personal Growth, Autonomy, and Self-Acceptance subscales of Ryff’s (1989) multidimensional measure of psychological well-being.

Turning to the liberation portion of our counseling outcomes, we purposefully sought out constructs that would span personal, relational, and collective spheres; that would work to free women from oppression as well as empower them to work toward personal and social change; and that would forge linkages across these three spheres. Although other multicultural analyses have applied these concepts to race/ethnicity (Vera & Speight, 2003) and physical disability (Prilleltensky & Prilleltensky, 2003), our study is the first known to explore this model empirically and with a focus on women.

At the individual level, we examined personal self-esteem (Rosenberg, 1965), agency (both with [Spence & Helmreich, 1978] and without [Vallacher & Wegner, 1989] ties to the gender literature), personal empowerment (Johnson, Worell, & Chandler, 2005), and personal entitlement (an exaggerated sense of deservingness; Campbell, Bonacci, Shelton, Exline, & Bushman, 2004). At the relational level, we focused on college women’s almost universal desire for egalitarian relationships (Gilbert & Radar, 2001) and the role of sexual assertiveness in promoting women’s sexual health and well-being, specifically regarding women’s attitudes about sexual refusal (Morokoff et al., 1997). At the broadest level, we included collective esteem focused on membership in and identification with social groups in general (Luhtanen & Crocker, 1992) and social justice entitlement for women collectively (Beaton & Tougas, 2001). Our selected variables forged linkages across spheres tying both agency and personal empowerment (personal) to both egalitarianism and sexual refusal (relational) as well as connecting self-esteem (personal) with collective esteem (collective) and individual entitlement (personal) with justice entitlement (collective). Although personal entitlement fell outside our superordinate interest in positive functioning, we included it as a counterweight to empowerment and to balance justice entitlement.

The Present Study

In sum, we sought to generate multivariate definitions of both feminist beliefs and well-being with liberation that are representative of the scope of these obviously complex constructs. Our goal is to bring some coherence to the literature, which to date has explored relationships between individual variables within each set, yielding a generally positive but sometimes inconsistent conclusion that holding feminist beliefs is associated with positive outcomes for women. By drawing on canonical correlation analysis, we can (a) identify how many canonical functions, which maximize the correlations between these two sets of variables, exist as well as (b) explore within each function which specific variables define each multivariate set. Our ultimate goals are to add to what we know about counseling models of feminist beliefs and to contribute to counselors’ understandings of how best to put feminism into practice in individual counseling, that is, how to best facilitate clients’ positive psychological functioning.

Method

Participants

Although 255 of 370 participants contacted actually began our 60- to 75-minute-long online survey (a 69% response rate), our final sample comprised 215 college women (an 84% completion rate). Two men were eliminated. The remaining 38 women elected to withdraw from the study by ceasing to respond: 8 stopped after completing the initial demographics, and no consistent pattern emerged to signal when the remaining 30 participants withdrew. The point in the survey at which reference to women became clearest was with the social justice entitlement items; all but 3 women withdrew before reaching this point, and the persistent 3 participants completed items well beyond this turning point. In sum, attrition appeared idiosyncratic, although we cannot determine if the length of the survey created its own selection bias.

All but one of the sampled women was 30 years old or younger; 9 (4%) were older than 21 (Mdn = 19, SD = 2.36, range = 18–39, with 2 missing). (Exclusion of students older than 21 yielded canonical correlation analysis findings comparable to the ones presented here.) Fully 85% (n = 180) identified as White, with 16 Black women and 16 others (3 missing); the majority (90%) were 1st- or 2nd-year college students. Similar proportions were either in (48%, n = 102) or not in (45%, 96) a committed intimate relationship, with an additional 13 cohabiting, 1 married, 1 previously married, and 2 missing. The majority (86%, n = 185) described themselves as exclusively heterosexual on a scale including mostly heterosexual (20), bisexual (6), mostly homosexual (2), and exclusively homosexual (2). A power analysis indicated that this sample was large enough (with α = .05, power = 80%) to detect a minimum correlation of .20.

Handling Missing Data

Although there were very few missing responses in the dataset for the majority of items (four women failed to answer all three items assessing sense of common fate), some of the items from the Well-Being scale, the Personal Progress Scale measuring empowerment, and most notably the Sexual Assertiveness Scale (with one woman skipping the full scale) had 5% to 15% missing data. We dealt with this missingness with two different methods: (a) the multiple imputation procedure recommended by Schafer and Graham (2002) and (b) ipsative mean imputation, which simply averages the valid responses for each individual for an overall scale score. Although ipsative mean imputation is far less cumbersome, we could not assume that these missing responses happened at random given the sensitivity of our items and the nature of the missingness. As Schafer and Graham pointed out, sample data with nonrandom missingness can produce biased parameter estimates. In this case, a multiple imputation procedure is recommended, whereby several datasets are created and analyzed, resulting in a range of estimates for each parameter. When we followed Schafer and Graham’s multiple imputation procedure and generated five imputed datasets, the correlations between scales created from these imputed values and the ipsative mean imputation scales ranged from .953 to .999, and all multivariate analyses with these datasets yielded the same results as the less elegant ipsative mean imputation procedure.

Given the extreme similarity of the results generated by these two different methods and our findings that the amount of missingness in our dataset did not seriously bias parameter estimates and standard errors when ipsative mean imputation was used to create composite scores, we opted to present the results from the single dataset generated with ipsative mean imputation, and thus we provide the reader with one set of parameter estimates (rather than a range of estimates for each coefficient). Thus, for all scale-level variables, an average of the valid items in the measure was computed for each individual. All means, standard deviations, ranges (possible and actual for our sample), coefficient alphas, and intercorrelations among the following measures for the full sample are reported in Table 1.

Table 1.

Means, Standard Deviations, Coefficient Alphas, and Intercorrelations Among All Variables

			Feminist Beliefs								Well-Being With Liberation
Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21
1. M-FPS	4.33	0.77	(.74)	.54^**	.22^**	−.24^**	.49^**	.44^**	.32^**	.55^**	−.06	.02	−.08	−.08	−.08	.05	.16^*	−.02	.24^**	.08	.08	−.06	.41^**
2. FWM	3.21	0.48		(.79)	.34^**	−.19^**	.32^**	.38^**	.38^**	.50^**	.04	.12	.04	.04	−.03	.08	.02	.14^*	.19^**	.24^**	.07	−.08	.23^**
3. CmFate	2.36	0.66			—	.02	.30^**	.29^**	.29^**	.34^**	−.03	.16^*	.10	.09	−.07	.08	.09	.17^*	.19^**	.07	.15^*	.09	.23^**
4. FIC-PA	3.01	0.62				(.74)	−.02	−.25^**	.01	−.08	−.23^**	−.15^*	.03	.13	.00	.16^*	.10	−.02	−.15^*	−.08	.01	.12	−.06
5. FIC-Rev	2.50	0.73					(.86)	.46^**	.06	.42^**	−.24^**	−.21^**	−.22^**	−.24^**	−.23^**	.03	.17^*	−.23^**	−.14^*	−.02	−.21^**	−.16^*	.52^**
6. FIC-EE	2.87	0.91						(.85)	.27^**	.53^**	.07	.12	−.01	−.07	−.05	.13	.09	.10	.15^*	.12	.09	−.05	.31^**
7. FIC-Syn	3.94	0.56							(.76)	.53^**	.23^**	.34^**	.20^**	.17^*	.19^**	.04	−.05	.37^**	.41^**	.21^**	.38^**	.07	.08
8. FIC-AC	3.33	0.56								(.83)	.14^*	.24^**	.14^*	.10	.10	.15^*	.10	.31^**	.25^**	.18^**	.21^**	.02	.37^**
9. WB-Auto	4.26	0.75									(.80)	.49^**	.52^**	.42^**	.53^**	.02	.02	.62^**	.33^**	.25^**	.37^**	.13	−.07
10. WB-Grw	4.67	0.62										(.71)	.55^**	.35^**	.38^**	.09	−.13	.47^**	.44^**	.20^**	.51^**	.27^**	−.09
11. WB-Acpt	4.49	0.79											(.84)	.72^**	.54^**	.14^*	.07	.62^**	.28^**	.21^**	.57^**	.30^**	−.09
12. Self-est.	3.23	0.49												(.88)	.54^**	.20^**	.19^**	.61^**	.22^**	.21^**	.46^**	.24^**	−.14
13. Agency	3.56	0.53													(.68)	.19^**	.10	.52^**	.18^**	.54^**	.40^**	.21^**	−.11
14. ActionID	8.04	3.05														(.72)	.07	.20^**	.01	.19^**	.09	.12	−.04
15. Prs Entl.	3.06	0.60															(.85)	.10	−.09	−.07	.14^*	.04	.22^**
16. Prs Emp.	5.14	0.74																(.88)	.42^**	.30^**	.51^**	.33^**	.11
17. Egalitar.	3.72	0.95																	(.72)	.20^**	.40^**	.24^**	−.05
18. SexRefuse	3.94	0.93																		(.76)	.14^*	.04	.01
19. MmbrEst.	3.72	0.95																			(.84)	.52^**	−.10
20. ID-Est.	4.52	1.11																				(.69)	−.13^*
21. JustEntl.	3.06	0.60																					(.74)

N = 215, except for Sense of Common Fate (n = 211) and Sexual Assertiveness-Refusal (n = 214). Coefficient alphas are presented in parentheses along the diagonal. M-FPS = Modified Feminist Perspectives Scale (possible range = 1–7; actual range = 1.83–6.58); FWM = Feminism and the Women’s Movement (1–5; 1–5); CmFate = Sense of Common Fate (1–4; 1–4); FIC = Feminist Identity Composite (PA = Passive Acceptance, REV = Revelation, EE = Embeddedness, Syn = Synthesis, AC = Active Commitment; 1–5; 1–4.71, 1–4.38, 1–5, 2.60–5, 1.56–5, respectively); WB = Well Being (Auto = Autonomy, Grw = Personal Growth, Acpt = Self-Acceptance; 1–6; 2.11–5.89, 3–5.89, 1.78–6, respectively); Self-est. = Self-Esteem (1–4; 2–4); Agency (1–5; 1.88–4.88); ActionID = Action Identification (0–25; 0–25); Prs. Entl. = Personal Entitlement (1–7; 1–5.89); Prs. Emp. = Personal Empowerment (1–7; 3.12–7); Egalitar. = Egalitarianism (1–5; 2.18–4.18); SexRefuse = Sexual Assertiveness–Refusal (1–5; 1.67–5); MmbrEst. = Member Esteem (1–7; 1.75–6.25); ID-Est. = Identity Esteem (1–7; 1.50–7); and JustEntl. = Justice Entitlement (1–5; 1.22–5).

p < .05. **p < .01.

Measures of Feminist Beliefs

Modified Feminist Perspectives Scale

Henley and colleagues (1998) developed the FPS as a 78-item measure of attitudes toward women that included 10 attitudinal and three behavioral items for each of six perspectives derived from feminist theories describing conservative, liberal, radical, socialist, cultural, and womanist feminism. Subsequent research with 25 attitudinal items (Henley, Spalding, & Kosta, 2000), chosen for their high factor loadings, yielded a strong correlation (.85) with the full scale as well as solid internal consistency reliability (α = .85) and convergent validity with both the Attitudes Toward Women’s Scale (Spence & Helmreich, 1978) and the Bem Sex Role Inventory (Bem, 1974). Similar to this shortened version of this scale, we further reduced the attitudinal scale to 12 items (M-FPS), selecting the two top loading items for each of the six perspectives and averaging responses across all items such that higher scores represented stronger endorsement of feminist attitudes. A sample item from the Liberal subscale is, “Whether one chooses a traditional or alternative family form should be a matter of personal choice.” Items were rated on 7-point scales ranging from 1 (strongly disagree) to 7 (strongly agree), with an undecided midpoint. The alpha and correlations among the feminist measures presented in Table 1 provide support for the viability of our modification.

Feminism and the Women’s Movement

Fassinger (1994) developed the FWM as a 10-item measure of affective attitudes toward the feminist movement. With a sample of college women, she reported Cronbach’s alpha of .87. Fassinger established the scale’s convergent validity with the Attitudes Toward Women’s Scale (Spence & Helmreich, 1978), the Attitudes Toward Feminism Scale (Smith, Ferree, & Miller, 1975), the Sex Role Egalitarianism Scale/Form BB (King & King, 1990), and two single items of identification with feminism and favorability toward the women’s movement as well as its divergent validity from the Personal Attributes Questionnaire (Spence & Helmreich, 1978), Rokeach’s Dogmatism Scale (Troldahl & Powell, 1965), and the Marlowe-Crowne Social Desirability Scale (Crowne & Marlowe, 1961). A sample item is, “The leaders of the women’s movement may be extreme, but they have the right idea,” and items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree), with a midpoint of undecided. An average of the items yielded a single composite measure such that higher scores indicated more favorable attitudes toward feminism and the women’s movement.

Sense of common fate

To capture collective feminist consciousness, we included the three items reflecting a sense of common fate developed by Gurin and Townsend (1986): (a) “How often in your everyday life do you have to think about being a woman and what you have in common with women?,” (b) “To what extent will what happens to women generally in this country affect what happens in your life?,” and (c) “How much do you have in common with most women?” Response choices were 1 (hardy at all), 2 (a little), 3 (some), and 4 (a lot). Their average produced a single score such that higher scores represented stronger beliefs that a woman’s fate is linked to that of women in general. The three items were significantly intercorrelated in the present study (rs = .17–.42, ps < .05).

Feminist Identity Composite

The FIC (Fischer et al., 2000) is a 33-item, more psychometrically sound (Moradi & Subich, 2002a) combination of items from the Feminist Identity Scale (Rickard, 1987) and the Feminist Identity Development Scale (Bargad & Hyde, 1991), both of which were created in response to Downing and Roush’s (1985) model of feminist identity development (Moradi et al., 2002). A one-time administration of the FIC captures a self-relevant snapshot of an individual at a single point in time within each of the five coexisting dimensions (Fischer & Good, 1994; Moradi et al., 2002). The highest loading items identified by Fischer and her colleagues (2000) to represent each subscale were “I don’t see much point in questioning the general expectation that men should be masculine and women should be feminine” (FIC-Passive Acceptance), “Gradually, I am beginning to see just how sexist society really is” (FIC-Revelation), “I am very interested in women writers” (FIC-Embeddedness-Emanation), “I feel like I have blended my female attributes with my unique personal qualities” (FIC-Synthesis), and “I am very committed to a cause that I believe contributes to a more fair and just world for all people” (FIC-Active Commitment). All items were rated on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree) with 3 (neutral or undecided) as the midpoint. Each item was coded so that higher scores reflected more agreement with the ideology of the stage or dimension targeted and then was averaged within its subscale. Thus, high scores on the 7-item Passive Acceptance subscale reflect nonfeminist views; on Revelation (8 items), changing views; and on the Embeddedness-Emanation (4 items), Synthesis (5 items), and Active Commitment (9 items) subscales, endorsement of feminism and a feminist ideology. Coefficient alphas from previous research ranged from .74 to .75, .75 to .80, .84 to .86, .68 to .73, and .77 to .81 for the five subscales, respectively (Fischer et al., 2000; Moradi & Subich, 2002a). In the present study, the alpha for FIC-Passive Acceptance was lower (see Table 1); the remainder, except for Embeddedness-Emanation, higher.

Well-Being

To concentrate on the dimensions of well-being most relevant to our interests (as noted previously), we included three of the six independent scales of Ryff’s (1989) Scales of Psychological Well-Being: Autonomy, Personal Growth, and Self-Acceptance. There are various iterations of these scales ranging from 3 to 14 items (Springer & Hauser, 2006). Following Ryff’s (personal communication) recommendation, we settled on the 9-item scales used with Dutch college students (Autonomy: α = .78, Personal Growth: α = .65, and Self-Acceptance: α = .83; van Dierendonck, 2004). A sample item evaluating Autonomy is, “I am not afraid to voice my opinions, even when they are in opposition to the opinions of most people”; Personal Growth, “I am not interested in activities that will expand my horizons” (reverse coded); and Self-Acceptance, “When I look at the story of my life, I am pleased with how things have turned out.” Responses were recorded on a 6-point scale ranging from 1 (strongly disagree) to 6 (strongly agree), with higher composite scores indicating heightened well-being.

Liberation in the Personal Sphere

Self-esteem and agency

The widely used and validated (Blascovich & Tomaka, 1991) Rosenberg (1965) Self-Esteem Scale included 10 items scored on a scale from 1 (strongly agree) to 4 (strongly disagree), where higher scores indicated stronger self-esteem. The frequently used and validated (Lenney, 1991) Personal Attributes Questionnaire’s (PAQ; Spence & Helmreich, 1978) Agency subscale included eight semantic differential items rated on a 5-point scale. People who score high in agency or instrumentality regard themselves as autonomous and active doers (Eagly, 1987; Spence & Helmreich, 1980). Higher averaged scores represented higher self-esteem and agency, respectively.

As a second measure of personal agency independent of the gender literature with which the PAQ was associated, Vallacher and Wegner’s (1989) Behavior Identification Form presented 25 behaviors (e.g., “making a list”), forcing respondents to choose between two alternatives that “best describes the behavior for you” (e.g., 1 = getting organized or 0 = writing things down). The higher-level option delineated why or with what effect the action is performed (coded 1); the other choice simply specified how it was done (coded 0). Summing across choices, higher scores (ranging from 0 to 13 with our sample) portrayed individuals’ self-concept as an agent who thinks about her own acts in terms of their motives and meaning, as opposed to details or means of action. Convergent validity was established across a series of action and motivation measures as well as discriminant validity across 13 personality measures (Vallacher & Wegner, 1989).

Personal entitlement

A personal sense of both entitlement and deservingness was assessed with the 9-item Personal Entitlement Scale (Campbell et al., 2004). Campbell et al. (2004) define psychological entitlement “as a stable and pervasive sense that one deserves more and is entitled to more than others” (p. 31), and this sense is conveyed well in the first item: “I honestly feel I’m just more deserving than others.” Items were rated on a scale from 1 (strong disagreement) to 7 (strong agreement), with a midpoint of 4 (neither agreement or disagreement), and an averaged composite score reflected higher levels of perceived entitlement. Internal consistency reliability (α = .85) was strong, test–retest reliability across 2 months (r = .70) supported the scale’s stability, and no gender differences were reported by Campbell et al. (2004). Psychological entitlement was independent from social desirability and predicted the actual behavior of taking candy from children as well as pay entitlement and overuse of common resources in the commons dilemma. Although it is a negative outcome outside our interest in optimal psychological functioning, we included this variable because stereotyping of feminists includes being egotistical and self-serving (Twenge & Zucker, 1999), a charge that may work to mute women’s empowerment and support for justice entitlement.

Personal empowerment

The 28-item Personal Progress Scale–Revised (Johnson et al., 2005) was built on Worell and Remer’s (2003) empowerment model, which seeks to establish personal empowerment as a goal for women-centered interventions. The scale developers used Wyche and Rice’s (1997) definition of personal empowerment as “helping women to become more independent and assertive about attaining her goals and achieving change and psychological growth” (p. 60). Items such as “I have equal relationships with important others in my life” were evaluated on a 7-point scale ranging from 1 (almost never) through 4 (sometimes true) to 7 (almost always). With a mostly college student and community sample (Johnson et al., 2005), alpha (.88) was strong, and the scale established its convergent validity by predicting overall well-bring, autonomy, and self-acceptance using Ryff’s (1989) measures and discriminant validity through its independence from negative functioning and distress as measured by the Outcome Questionnaire (Lambert, Okiishi, Finch, & Johnson, 1998). In the present study, a single, averaged score reflected a heightened sense of personal empowerment.

Liberation in the Relational Sphere

Egalitarian attitudes

The Marriage Role Expectations Inventory tracked the expectancies of six cohorts of women from 1961 through 1996 (Botkin, Weeks, & Morris, 2000). The full measure includes six subscales, one of which explores the anticipated balance of power in a “committed intimate relationship” (a modification from the original focus on marriage used by Yoder et al., 2007; α = .71). A sample item from the 11 in this subscale on Authority is, “In my marriage or committed intimate relationship, I expect that if there is a difference of opinion, my partner will decide where to live.” Respondents rated each item on a 5-point scale ranging from 1 (strongly agree) to 5 (strongly disagree), with a midpoint of 3 (undecided). Following Yoder et al. (2007), we reverse coded the Authority items and averaged scores so that higher scores indicated stronger endorsement of egalitarianism.

Sexual refusal

Morokoff et al. (1997, Study 2) documented that the 6-item Sexual Refusal subscale (α = .71) of the Sexual Assertiveness Scale (SAS) was negatively related both to undergraduate women’s expectations that their partner would react negatively to sexual refusal and to experiences of sexual coercion, making this an appropriate measure to assess women’s assertiveness in sexual relationships. A sample item is, “I refuse to have sex if I don’t want to, even if my partner insists.” Responses categorized frequency ratings ranging from 1 (never, 0% of the time) to 5 (always, 100% of the time), with intervals of 25% and were coded and averaged so that higher scores denoted higher sexual refusal. Our survey included the complete 18-item SAS; however, the present analyses focus solely on the Sexual Refusal subscale.

Liberation in the Collective Sphere

Collective esteem

Luhtanen and Crocker (1992) developed the Collective Self-Esteem Scale in line with social identity theory’s conception of collective identity, including individuals’ assessments (a) of how good they are as members of their social groups (membership esteem) and (b) of the importance of their social groups to their self-concept (identity esteem). Sample items for the four-item Membership and 4-item Identity subscales are, “I am worthy of the social groups I belong to” and “The social groups I belong to are an important reflection of who I am,” respectively. Items were rated on scales ranging from 1 (strongly disagree) to 7 (strongly agree), with a midpoint of 4 (neutral). The scale authors reported alphas of .73 to .80 for college student samples as well as discrimination validity from Rosenberg’s (1965) self-esteem scale and convergent validity with Hui’s (1988) Individualism-Collectivism scale. Each subscale was scored such that higher scores represented greater collective esteem.

Justice entitlement

Justice entitlement concentrates on issues of extending fair treatment to others, focusing specifically on groups targeted for affirmative action (Beaton & Tougas, 2001). Concentrating solely on women, a sample from the nine items is, “Many people do not realize the problems women face,” rated on a scale from 1 (strong disagree) to 5 (strongly agree), where 3 is undecided. Cronbach’s alpha with college students was .75, and high scores were negatively correlated with beliefs that affirmative action threatens nondesignated groups and positively correlated with endorsement of affirmative action (Beaton & Tougas, 2001). In the present study, a higher averaged score indicated a broader scope of justice entitlement for women.

Procedures

College student recruiters solicited names and email addresses from women enrolled in psychology classes who were interested in an online survey of “your attitudes.” An email, which noted that students could “earn extra credit . . . for participating in psychology research,” cautioned students that they would need to complete the online survey in one sitting of “up to 75 minutes.” A URL link to the web-based SurveyMonkey questionnaire was provided so that each student (and only recruited students) could participate at a time and place convenient for her. Data collection took place between November 17, 2007 and March 7, 2008.

The survey itself began with an opening description and informed consent form that was titled, “A Study of Your Belief System.” The introduction started, “Some social and personality psychologists study individuals’ belief systems, looking at how people’s views of the world fit together. We are interested in how you see and evaluate yourself, your relationships, and your place in various social groups.” The debriefing and educational component at the conclusion of the survey referred back to the social psychological literature on beliefs.

Data collection began with basic demographic information (age, sex, race/ethnicity, year in college, relationship status, and sexual orientation), and then presented the following measures in order: self-esteem, PAQ-agency, the three well-being subscales, action identification, personal entitlement, collective member and identity esteem, egalitarianism, empowerment, justice entitlement, sense of common fate, the FIC, the FWM, the M-FPS, and sexual refusal. The order of measures was chosen to mask our interest in women’s and feminist issues, with the first mention of women appearing in social justice entitlement. The SAS was placed last given the sensitive nature of its sexual content (and it could be skipped in its entirety without penalty). Three additional measures taken from Zucker (2004) were included but were analyzed elsewhere (Yoder, Tobias, & Snell, 2011): her three cardinal beliefs (placed after justice entitlement), a forced choice item of feminist labeling (am, am not), and six items regarding political actions (e.g., signing a petition) on behalf of women’s rights, lesbian or gay rights, and pro-choice (both immediately before the SAS). All measures included a neutral choice as the midpoint of the rating scale and a “skip” option (coded as missing) to allow for nonresponses. Peppered throughout the survey were markers indicating how much of the survey had been completed and providing encouragement to continue.

Results

Preliminary Analyses

Table 1 presents the means, standard deviations, and intercorrelations among all study variables, and it divides these measures into two distinct sets representing Feminist Beliefs and Well-Being With Liberation. Feminist beliefs were represented by eight variables: the M-FPS, the FWM, common fate, and the five facets of the FIC (Passive Acceptance, Revelation, Embeddedness-Emanation, Synthesis, and Active Commitment). Well-Being With Liberation was operationalized with 13 variables: three forms of well-being (autonomy, personal growth, and self-acceptance), self-esteem, agency, action identification, personal entitlement, personal empowerment, egalitarianism, sexual refusal, collective esteem (both member and identity), and justice entitlement.

The designated sets of intercorrelations among the measures of Feminist Beliefs and among the indices of Well-Being With Liberation made a case for the convergent validity of each set. The measures of feminist beliefs were mainly positively correlated, with the exception of FIC-Passive Acceptance, which was expectedly negatively related or unrelated to the other measures. Of our measures of Well-Being With Liberation, action identification and both forms of entitlement (personal and justice) stood out as least related to the other measures in this set. However, the most far-reaching measure of individual liberation in this set is self-esteem, a well-established and widely used measure and construct, and it was positively correlated with every other measure in this block of Well-Being With Liberation measures.

Canonical Correlation Analysis

For the current study, we were most interested in examining how the set of variables that describe feminist beliefs relate to a woman’s well-being with liberation rather than an examination of the simple bivariate relationships between these two sets of variables. We were intrigued with the idea that there may be different patterns of feminist beliefs that were linked to different patterns of well-being with liberation. A canonical correlation analysis (CCA), which is a form of the general linear model, is specifically engineered for this type of research question.

CCA has the requirement of multivariate normality, which is typically evaluated by examining the univariate normality for the variables and their homoscedasticity (Tabachnick & Fidell, 2007). The Sexual Refusal scale was the only variable in the analysis that was significantly skewed and therefore generated the only violations regarding linearity and homoscedasticity. Thus, we conducted a CCA with the reflected logarithmic transformation for the Sexual Refusal measure. This alternate CCA yielded the same number of interpretable functions as well as the same pattern of relationships as the final CCA with the more readily interpretable original raw variable that we report here.

Results from this final CCA demonstrate that these two sets of variables were significantly related to each other as evidenced by a Wilks’s lambda of .202, F(104, 1312) = 3.26, p < .001. Even more interesting are the unique functions uncovered by the CCA to illuminate how these two sets of variables are maximally correlated with each other. The dimension reduction analysis indicated that three roots were significant at a p < .05 level, establishing that these sets of variables were related to each other in three distinct ways.

An effect size calculation in CCA is not a straightforward calculation like a R² in multiple regression, and a number of different types of indices can be created (Stevens, 1986). Given our interest in understanding the multivariate relationship between Feminist Beliefs and Well-Being With Liberation, the redundancy index suggested by Stewart and Love (1968) was calculated. Based on this redundancy index, the first Feminist Belief function explained 9.01% of the variance in Well-Being With Liberation variables (see Table 2). The second function explained an additional 4.41% of this variance; the third, 1.80%. To complicate matters further, the effect size calculations for CCA are not symmetrical. In our study, all three distinct Feminist Belief functions accounted for 15.22% of the total variance, whereas the three functions that were derived for the Well-Being With Liberation variables explained 23.73% of the variance in the Feminist Beliefs variables (see Table 2).

Table 2.

Results of the Canonical Correlation Analysis

	Canonical Functions
	Established Feminism		Awakening Feminism		Woman-Identified Traditionalism
Multivariate Sets	Corr.	Coeff.	Corr.	Coeff.	Corr.	Coeff.
Feminist Beliefs
Feminist Perspectives	.39	0.18	.75	0.67	−.17	0.14
Women’s Movement	.42	−0.06	.37	−0.23	−.08	0.04
Common Fate	.40	0.33	.39	0.25	.32	0.59
FIC-Passive Acceptance	−.26	−0.28	.04	0.27	.88	1.45
FIC-Revelation	−.14	−0.76	.89	0.78	−.20	−0.58
FIC-Embeddedness	.43	0.20	.49	0.05	−.22	−0.04
FIC-Synthesis	.81	0.88	.06	−0.43	.02	−0.43
FIC-Active Commitment	.67	0.71	.53	0.21	.06	0.47
Percentage of variance	20.27		10.93		9.32
Redundancy	9.01		4.41		1.80
Well-Being with Liberation
Well-Being: Autonomy	.48	−0.12	−.39	−0.43	−.33	−1.01
Well-Being: Personal growth	.67	0.56	−.23	0.01	−.03	−0.13
Well-Being: Self-acceptance	.41	−0.31	−.27	−0.02	.31	0.02
Self-esteem	.34	−0.23	−.27	−0.20	.55	1.18
Agency	−.17	−0.17	.09	0.21	.20	0.58
Action Identification	.13	0.01	.14	0.06	.45	0.22
Personal Entitlement	−.05	0.03	.35	0.23	.25	0.02
Personal Empowerment	.72	0.73	−.28	−0.10	.31	0.53
Egalitarianism	.72	0.78	−.02	0.44	−.04	−0.15
Sexual Refusal	.35	0.10	−.02	0.07	−.08	−0.17
Collective Member Esteem	.66	0.43	−.17	0.04	.27	0.01
Collective Identity Esteem	.17	−0.28	−.15	−0.07	.44	0.09
Justice Entitlement	.08	0.25	.89	1.35	−.18	−0.14
Percentage of variance	23.31		27.19		12.48
Redundancy	10.36		10.96		2.41

N = 210. Corr. = structural correlation; coeff. = unstandardized function coefficient; FIC = Feminist Identity Composite.

To interpret the meaning of each of these three canonical functions, we explored the structural correlations and unstandardized function coefficients (see Table 2). We interpreted the structural correlations because they best reflect the underlying structure of the extracted function (Huberty, 1984), and, unlike the function coefficients, they are not vulnerable to problems with multicollinearity. Each canonical function is readily interpreted by paying attention to the strongest correlations, both positive and negative (similar to factor loadings in factor analysis). We named each function in accordance with its highest loading Feminist Belief variables.

Function 1: Established feminism

As can be seen in Table 2, the first extracted function was characterized by high levels of synthesis and active commitment, with moderate loadings of embeddedness, positive attitudes toward the women’s movement, a sense of common fate, and endorsement of feminist perspectives, along with low scores on passive acceptance. In sum, this function described a rather classic definition of feminism as being most strongly, and notably positively, associated with our set of measures of Well-Being With Liberation. Specifically, established feminism was characterized by very high levels of personal empowerment (and not personal entitlement) and egalitarianism as well as high levels of personal growth well-being and collective member esteem. It also was associated with moderate degrees of autonomy and self-acceptance well-being, sexual refusal, and overall self-esteem as well as some collective identity esteem. Only agency showed a small but noteworthy negative correlation within this set of Feminist Belief variables. In sum, this function uncovered a strong and positive correlation between feminist beliefs indicative of established feminism and well-being with liberation.

As with any correlation, reversing our perspective can also prove intriguing. Examining the set of variables that make up Feminist Beliefs for this function in reverse makes nonfeminist passive acceptance the most influential variable and links its endorsement with active rejection of many of these feminist beliefs, including the three FIC dimensions, endorsement of the women’s movement, common fate, and feminist perspectives. Our interpretation of this multivariate set, then, from this vantage does not represent passive nonfeminism alone, but rather a richer pattern of active antifeminism. The correlations of this antifeminism with the set of Well-Being With Liberation variables reverses the outcomes that we found above to be favorable for established feminism—thus linking antifeminism to compromised personal empowerment and egalitarianism, well-being, collective member esteem, sexual refusal, and self-esteem. Furthermore, no compensatory positive outcomes emerged from the remaining indicators of Well-Being With Liberation.

Function 2: Awakening feminism

The second multivariate function to emerge, above and beyond the first function, was based on a set of Feminist Beliefs most strongly characterized by FIC-Revelation or the component of feminism that captures an awakening to feminist beliefs (see Table 2). In addition, this set of Feminist Belief variables drew on those feminist principles (feminist perspectives, active commitment, embeddedness, common fate, and support for the women’s movement) that were included in the first function for established feminism, underscoring that this second function appears positively feminist (as opposed to nonfeminist or antifeminist). This set of awakening feminist beliefs also correlated with a different configuration of Well-Being With Liberation variables than established feminism. Topping this list by far was justice entitlement, followed by moderate endorsement of personal entitlement and characterized by low levels of well-being, personal empowerment, and self-esteem—a very different relationship between beliefs and well-being with liberation than found for established feminism and certainly not a relationship suggesting generally positive psychological functioning.

Function 3: Woman-identified traditionalism

At first blush, the set of variables defining our third function looked like the antifeminist combination of Feminist Beliefs that defined our reversed first function in that both sets were defined by passive acceptance (see Table 2). However, a closer look at this third variable set showed not only stronger endorsement of passive acceptance but also neutrality to most feminist attitudes (as opposed to the outright rejection of feminist principles that characterized the antifeminism in the reversed first function) along with a sense of common fate with women as a group yet resistance to seeking embeddedness with feminist figures. This third combination of Feminist Belief variables included acceptance of the nonfeminist status quo, indicating traditionalism, but simultaneously included identification with women as a group by sharing a common fate (but clearly an understanding that was devoid of feminist understandings). Thus, the defining features of this set of Feminist Beliefs were that they were (a) traditional in their support of the status quo (passive acceptance) and (b) although woman-identified, not feminist.

This pattern of woman-identified traditionalism was linked to a rather distinctive pattern of Well-Being With Liberation variables exemplified by moderately high self-esteem as well as by moderate levels of action identification (i.e., gender-neutral agency), collective identity esteem, personal empowerment (along with personal entitlement), self-acceptance well-being, and collective member esteem. Negatively related to this set of beliefs was autonomy well-being. In sum, this combination of nonfeminist traditional yet woman-centered beliefs is associated with liberation for one’s self (self-esteem, self-acceptance, personal entitlement, action identification, and personal empowerment) and with collective liberation (collective identity and member esteem), but with compromised individual autonomy and collective justice entitlement.

Discussion

Given the two sets of variables we constructed to capture women’s feminist beliefs and well-being with liberation (as indicative of self-reported, generally positive psychological functioning), two broad conclusions emerge from our findings: feminist beliefs and well-being with liberation are significantly related (confirming prior research connecting feminist beliefs with generally positive outcomes for women) and this relationship is complex (explaining why there are some inconsistencies in this literature). This complexity extends both (a) to how various measures of feminist beliefs interrelate and (b) to how this set of beliefs links with well-being with liberation. Regarding this first point, our CCA did not identify just the first function (which identified a continuum from established feminism to antifeminism) that we might have expected. Rather, the first function also underscored differences between passive nonfeminism and active antifeminism, and two additional and orthogonal functions (awakening feminism and woman-identified traditionalism) captured the complicated ways in which women’s feminist beliefs can coexist.

Regarding the second point, there appear to be two different configurations of women’s feminist beliefs that are associated with a positive pattern of well-being with liberation: established feminism and woman-centered traditionalism. The more college women endorsed the beliefs of established feminism, the higher their reported levels of personal empowerment, egalitarianism, personal growth well-being, and collective member esteem. These beliefs were also favorably linked to autonomy well-being, self-acceptance well-being, sexual refusal, global self-esteem, and collective identity esteem. Looked at another way, although not positively nor substantially correlated with all of our well-being with liberation measures (excluding action identification and justice entitlement), only agency showed a small negative loading, suggesting that, at least among the indices of psychological functioning we studied, there is not much of a psychological downside to endorsing established feminism (at least within the limitations of our mostly White college sample).

High woman-identified traditionalism combined passive acceptance of the status quo (i.e., traditionalism) with recognition of the common fate that ties women together as women (i.e., a woman identification). From a feminist vantage, this combination of nonfeminism with recognition of women’s common fate seems oxymoronic; however, it is congruent with a profile of women associated with conservative women’s organizations (Schreiber, 2002). Such woman-identified traditionalism in our data is related to positive psychological functioning (self-esteem, gender-neutral action identification agency, collective esteem, personal empowerment, and self-acceptance well-being) but carries with it moderate threats to autonomy well-being as well as nonrelationships with important variables in both the personal (personal growth well-being) and relational (egalitarianism and sexual refusal) spheres. In sum, this pattern of beliefs certainly is generally positive for women but with a bit more of a downside than established feminism.

The combination of feminist beliefs that generally characterize high established feminism take on a very different relationship with well-being with liberation when the dominant correlation in this set is FIC-Revelation (a time of crisis and developing feminist awareness)—forming a separate function we defined as awakening feminism. Spearheading this belief set’s correlation with our set of liberation variables was entitlement, both personal and, more strongly, justice entitlement. However, moderate to low threats to well-being with liberation (autonomy well-being, personal empowerment, self-esteem, both self-acceptance and personal growth well-being, and collective member and identity esteem) characterize high awakening feminism as quite threatening to psychological functioning. If as Downing and Roush (1985) theorized, revelation is a stage through which many women pass toward feminist development, then our data suggest that this period may be a time of psychological vulnerability for college women. This conclusion is consistent with existing research that has associated the revelation stage with anger (Fischer & Good, 2004), psychological distress (Moradi & Subich, 2002a), and threatened self-esteem (Moradi & Subich, 2002b). However, in our data there are signs of positive possibilities in this pattern: By further rejecting the status quo and by replacing revelation with synthesis, the configuration of beliefs may shift to the pattern of established feminism and thus may forge a positive link to psychological well-being with liberation (see Brown, 2008). If indeed revelation is a step toward establishing one’s feminism (along with its associated psychological benefits), then our data highlight the potential benefits of strengthening synthesized feminist beliefs and further rejecting passive acceptance—areas that can be readily targeted by practicing counseling psychologists.

Of special interest across these three configurations of feminist beliefs are their relationships with personal entitlement and empowerment. Empowerment is considered by some feminist theorists as a central outcome of holding feminist beliefs (Worell & Remer, 2003), and indeed it is highly positively correlated with endorsing established feminism, less strongly associated with traditionalism, and negatively linked with awakening feminism in our data. Although leveled against feminists as part of disparaging stereotyping (Twenge & Zucker, 1999), personal entitlement is unrelated to established feminism in our study, yet it is positively associated with high levels of both awakening feminism and traditionalism. We speculate that feeling narcissistically deserving may underwrite some of the justice entitlement that most strongly characterizes awakening beliefs and that replacing it with personal empowerment may help shift women from high awakening to high established feminism. Personal entitlement, devoid of justice entitlement, may also further signify what is troublesome about traditionalism—which may reflect a willingness to benefit from collective connections to women in self-serving, rather than in sociopolitical-serving, ways.

By far, the pattern of beliefs related to the most negative configuration of self-reported psychological functioning came in the form of the converse of established feminism. This set of antifeminist beliefs is characterized not only by nonfeminist passive acceptance of the status quo but also by active rejection of the key features of established feminism—most notably, synthesis and active commitment. Not only does this antifeminism show negative associations with all the beneficial indicators of well-being with liberation related to established feminism, but it also offers just a small positive relationship with agency to compensate for these losses.

Implications for Researchers and Practitioners

As for the practice and training implications of our findings for counseling psychologists, it helps to step back to consider the various roles that counseling psychologists can play around issues of gender and sexism. Atkinson, Thompson, and Grant (1993) describe a multidimensional framework for categorizing these roles within cultural acculturation (low or high, although the assumption with gender is that we all are highly acculturated via socialization practices), problem etiology (internal or external), and goals of helping (prevention or remediation). Crossing the last two of these dimensions yields four roles, with two focused on internal problems (and prevention–counselor; remediation–psychotherapist) and two on external problems (prevention–consultant; remediation–change agent). The set of two roles focused on internal problems is the more conventional one for counseling psychologists (Atkinson et al., 1993); however, advocates for a social justice agenda in counseling psychology argue for the addition of the set of two externally focused roles (e.g., Goodman et al., 2004; Vera & Speight, 2003). Indeed, all four roles work to accomplish the goals of critical psychology—promoting optimal psychological functioning and social justice (Prilleltensky & Prilleltensky, 2003)—the perspective that guided the selection of the outcome variables in the present study.

Guidance on how to operate as internally focused counselors and psychotherapists can be found in the counseling literature on multicultural competencies, which outlines three critical competency domains: (a) counselor awareness of personal assumptions, values, and biases; (b) understanding these aspects in clients; and (c) development of cultural intervention strategies and techniques (American Psychological Association, 2003; Arredondo & Perez, 2003). Our findings address each of these competencies. By being aware of different configurations of feminist beliefs, counselors and psychotherapists can more fully understand their own and their clients’ gendered assumptions, values, and biases. Our identification of related outcomes helps counselors and psychotherapists further pinpoint areas of likely compromised functioning as well as frame interventions for more optimal functioning. For example, our data suggest that the configuration of feminist beliefs defining established feminism is associated with compromised agency, possibly reflecting frustration with the large-scale goals of achieving gender equity. For women scoring high in awakening feminism, well-being, personal empowerment, and self-esteem may be compromised, whereas woman-identified traditionalism may call for assistance with autonomy well-being and collective justice entitlement (beyond personal entitlement). Our findings further suggest that developing justice entitlement is a key part of awakening feminism so that interventions that focus in this area may convert traditionalism or antifeminism into a more individually distressful period of revelation that ultimately may be necessary to achieve established feminism along with its largely positive outcomes of well-being with liberation. Indeed, in the contrast between traditionalism and established feminism may rest a core conundrum for counseling psychologists who must juggle the optimal functioning of the individual with the social justice imperative to seek liberation.

One solution to this dilemma is to concentrate mainly on the internally focused roles of counselor and psychotherapist; indeed, Pieterse, Evans, Risner-Butner, Collins, and Mason (2009) concluded from their content analysis of the syllabi of multicultural counseling courses that there is stronger adherence to the competencies paradigm of knowledge, awareness, and skills than to the social justice dictum for liberation. However, even the present findings argue against tipping these scales too far in the direction of exclusive internal focusing in that our least collectively connected pattern was for active antifeminists whose psychological functioning appeared the most impaired. Rather, our findings suggest that connecting collectively with women, especially with a feminist understanding of that connection (but even without it), enhances women’s psychological functioning over antifeminism. This conclusion serves then to underscore the underlying assumption that multicultural competencies and a social justice agenda need to be blended in counseling psychology by bringing the externally focused roles of consultant and change agent into both training and practice (Atkinson et al., 1993). Possible models for achieving this balance may be found in both the disability (e.g., Olkin, 2009) and feminist (e.g., Worell, 2001) literatures, wherein counseling psychologists are encouraged to develop expertise across all four roles without demanding that their clients adopt the practitioner’s assumptions, values, and biases.

This last point about potentially imposed values from practitioner as change agent to client may be informed by a second set of analyses with the current sample and variables as well as with the variables excluded from the present analyses, specifically feminist self-labeling and reported behavioral activism (e.g., signing petitions). Labeling alone, independent of all the feminist beliefs explored here, predicted feminist activities (Yoder et al., 2011). This finding highlights the critical importance of saying “I am a feminist” for bringing about social change—a label that may be important for counseling psychologists to embrace for themselves but debatable as a client outcome within the therapeutic relationship.

Limitations

The obvious limitation of our data and analysis plan concerns the choice of variables on both sides of each multivariate correlation. Although we have tried to represent a range of feminist beliefs and measures of these beliefs as well as an array of positive variables capturing well-being with liberation (across personal, relational, and collective spheres) as suggested by a social justice framework, CCA is necessarily limited by the variables included (and not included). Most notable among the excluded outcome variables are those that capture psychological distress and measures beyond self-reports. We also used a modified version of the FPS, and concerns about the construct validity of Ryff’s (1989) measures persist across various item choices (Springer & Hauser, 2006).

Furthermore, our sample was restricted to college students, representing a narrow age range, with limited racial/ethnic diversity, willing to begin and complete a long survey, and recruited from a participant pool in psychology. Our sample does tie our study into a larger existing literature that focuses on college women, and thus it offers insights into this not-inconsequential population. However, the limits of the generalizability of our findings and possible selection biases are important for both researchers and practitioners to keep in mind. For example, Rederstorff, Buchanan, and Settles (2007) found that gender attitudes moderated the relationship between sexual harassment and psychological well-being in opposing ways for African American and White American women. It is very possible that this and other moderators could intervene in our association of feminist beliefs with well-being and liberation.

Conclusion

Our data and analysis plan argue that what it means to be a feminist cannot be considered independent of its relationships with other variables, and counseling psychology readily contributes what these variables should be (both enhanced well-being and mitigated distress), at least for the individuals involved. The canonical functions that resulted from this understanding helps then clarify what it means for women themselves to hold feminist beliefs. Despite emerging concerns about the FIC-Synthesis dimension as both a valid operationalization of the construct of synthesis and the construct itself (Liss & Erchull, 2010), our data highlight the importance of this variable as the dominant (but certainly not singular) factor in the profile of women who endorse established feminism. Our canonical functions also go on to demarcate four consequential functions—established feminism, established antifeminism, awakening feminism, and woman-identified traditionalism—that move not only beyond the simple labels of being a feminist or not (e.g., Liss, Crawford, & Popp, 2004) but also beyond more nuanced categorizations such as feminist, egalitarian, and nonfeminist (Zucker, 2004). Importantly, our findings highlight the value of expansive feminist beliefs in promoting women’s well-being with liberation, not simply identification with women devoid of feminist understandings. As such, they challenge future researchers in counseling psychology and other disciplines not to ask what being feminist is in the abstract but rather to put these definitions into the broader context of women’s lives and relationships.

Footnotes

Acknowledgements

The authors thank Ann Fischer for her helpful comments on a draft of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Bios

Janice D. Yoder is a professor of psychology at the University of Akron.

Andrea F. Snell is an associate professor at the University of Akron.

Ann Tobias collected these data for her undergraduate honors project in the Department of Psychology at the University of Akron.

References

American Psychological Association. (1999). Archival description of counseling psychology. The Counseling Psychologist, 27, 589-592.

American Psychological Association. (2003). Guidelines on multicultural education, training, research, practice, and organizational change for psychologists. American Psychologist, 58, 377-402.

American Psychological Association. (2007). Guidelines for psychological practice with girls and women. American Psychologist, 62, 949-979.

Arredondo

Perez

(2003). Expanding multicultural competence through social justice leadership. The Counseling Psychologist, 31, 282-289.

Atkinson

D. R.

Thompson

C. E.

Grant

S. K.

(1993). A three-dimensional model for counseling racial/ethnic minorities. The Counseling Psychologist, 21, 257-277.

Bargad

Hyde

J. S.

(1991). Women’s studies: A study of feminist identity development in women. Psychology of Women Quarterly, 15, 181-201.

Bay-Cheng

L. Y.

Zucker

(2007). Feminism between the sheets: Sexual attitudes among feminists, nonfeminists, and egalitarians. Psychology of Women Quarterly, 31, 157-163.

Beaton

A. M.

Tougas

(2001). Reactions to affirmative action: Group membership and social justice. Social Justice Research, 14, 61-78.

Bem

S. L.

(1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.

10.

Blascovich

Tomaka

(1991). Measures of self-esteem. In Robinson

J. P.

Shaver

P. R.

Wrightsman

L. S.

(Eds.), Measures of personality and social psychological attitudes (pp. 115-160). San Diego, CA: Academic Press.

11.

Botkin

D. R.

Weeks

M. O.

Morris

J. E.

(2000). Changing marriage role expectations: 1961–1996. Sex Roles, 42, 933-942.

12.

Brown

L. S.

(2008). Feminist therapy. In Lebow

J. L.

(Ed.), Twenty-first century psychotherapies: Contemporary approaches to theory and practice (pp. 277-306). Hoboken, NJ: John Wiley.

13.

Campbell

W. K.

Bonacci

A. M.

Shelton

Exline

J. J.

Bushman

B. J.

(2004). Psychological entitlement: Interpersonal consequences and validation of a self-report measure. Journal of Personality Assessment, 83, 29-45.

14.

Crowne

D. P.

Marlowe

(1961). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354.

15.

Downing

Roush

(1985). From passive acceptance to active commitment: A model of feminist identity development for women. The Counseling Psychologist, 13, 695-709.

16.

Eagly

A. H.

(1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Lawrence Erlbaum.

17.

Eisele

Stake

(2008). The differential relationship of feminist attitudes and feminist identity to self-efficacy. Psychology of Women Quarterly, 32, 233-244.

18.

Enns

C. Z.

(1993). Twenty years of feminist counseling and therapy: From naming biases to implementing multifaceted practice. The Counseling Psychologist, 21, 3-87.

19.

Enns

C. Z.

(2011). Feminist therapy. In Altmaier

E. M.

Hansen

J. C.

(Eds.), The Oxford handbook of counseling psychology. New York, NY: Oxford University Press.

20.

Espin

O. M.

(1994). Feminist approaches. In Comas-Diaz

Greene

(Eds.), Women of color: Integrating ethnic and gender identities in psychotherapy (pp. 265-286). New York, NY: Guilford.

21.

Fassinger

R. E.

(1994). Development and testing of the Attitudes toward Feminism and the Women’s Movement (FWM) scale. Psychology of Women Quarterly, 18, 389-402.

22.

Fassinger

R. E.

(2000). Gender and sexuality in human development: Implications for prevention and advocacy in counseling psychology. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (3rd ed., pp. 346-378). New York, NY: John Wiley.

23.

Fischer

A. R.

Good

G. E.

(1994). Gender, self, and others: Perceptions of the campus environment. Journal of Counseling Psychology, 41, 343-355.

24.

Fischer

A. R.

Good

(2004). Women’s feminist consciousness, anger, and psychological distress. Journal of Counseling Psychology, 51, 437-446.

25.

Fischer

A. R.

Tokar

D. M.

Mergl

M. M.

Good

G. E.

Hill

M. S.

Blum

S. A.

(2000). Assessing women’s feminist identity development: Studies of convergent, discriminant, and structural validity. Psychology of Women Quarterly, 24, 15-29.

26.

Gilbert

L. A.

Rader

(2001). Current perspectives on women’s adult roles: Work, family, and life. In Unger

R. K.

(Ed.), Handbook of the psychology of women and gender (pp. 156-169). New York, NY: John Wiley.

27.

Goodman

L. A.

Liang

Helms

J. E.

Latta

R. E.

Sparks

Weintraub

S. R.

(2004). Training counseling psychologists as social justice agents: Feminist and multicultural principles in action. The Counseling Psychologist, 32, 793-837.

28.

Gurin

Townsend

(1986). Properties of gender identity and their implications for gender consciousness. British Journal of Social Psychology,25, 139-148.

29.

Henderson-King

D. H.

Stewart

A. J.

(1994). Women or feminists? Assessing women’s group consciousness. Sex Roles, 31, 505-516.

30.

Henley

N. M.

Meng

O’Brien

McCarthy

W. J.

Sockloskie

R. J.

(1998). Developing a scale to measure the diversity of feminist attitudes. Psychology of Women Quarterly, 22, 317-348.

31.

Henley

N. M.

Spalding

L. R.

Kosta

(2000). Development of the short form of the Feminist Perspectives Scale. Psychology of Women Quarterly, 24, 254-256.

32.

Huberty

(1984). Issues in the use and interpretation of discriminant analysis. Psychological Bulletin, 95, 156-171.

33.

Hui

C. H.

(1988). Measurement of individualism-collectivism. Journal of Research in Personality, 22, 17-36.

34.

Hurt

M. M.

Nelson

J. A.

Turner

D. L.

Haines

M. E.

Ramsey

L. R.

. . .Liss

(2007). Feminism: What is it good for? Feminine norms and objectification as the link between feminist identity and clinically relevant outcomes. Sex Roles, 57, 355-363.

35.

Hyde

J. S.

(2002). Feminist identity development: The current state of theory, research, and practice. The Counseling Psychologist, 30, 105-110.

36.

Jaggar

A. M.

Rothenberg

P. S.

(1993). Feminist frameworks: Alternative theoretical accounts of the relations between women and men (3rd ed.). New York, NY: McGraw-Hill.

37.

Johnson

D. M.

Worell

Chandler

R. K.

(2005). Assessing psychological health and empowerment in women: The Personal Progress Scale Revised. Women & Health, 41, 109-129.

38.

King

L. A.

King

D. W.

(1990). Abbreviated measure of sex role egalitarian attitudes. Sex Roles, 15, 207-214.

39.

Kobrynowicz

Branscombe

N. R.

(1997). Who considers themselves victims of discrimination? Individual difference predictors of perceived gender discrimination in women and men. Psychology of Women Quarterly, 21, 347-363.

40.

Lambert

M. J.

Okiishi

J. C.

Finch

A. E.

Johnson

L. D.

(1998). Outcome assessment: From conceptualization to implementation. Professional Psychology: Research and Practice, 29, 63-70.

41.

Landrine

Klonoff

E. A.

(1997). Discrimination against women: Prevalence, consequences, remedies. Thousand Oaks, CA: Sage.

42.

Lenney

(1991). Sex roles: The measurement of masculinity, femininity, and androgyny. In Robinson

J. P.

Shaver

P. R.

Wrightsman

L. S.

(Eds.), Measures of personality and social psychological attitudes (pp. 573-660). San Diego, CA: Academic Press.

43.

Lent

R. W.

(2004). Toward a unifying theoretical and practical perspective in well-being and psychosocial adjustment. Journal of Counseling Psychology, 51, 482-509.

44.

Liss

Crawford

Popp

(2004). Predictors and correlates of collective action. Sex Roles, 50, 771-779.

45.

Liss

Erchull

M. J.

(2010). Everyone feels empowered: Understanding feminist self-labeling. Psychology of Women Quarterly, 34, 85-96.

46.

Lopez

S. J.

Edwards

L. M.

(2008). The interface of counseling psychology and positive psychology: Assessing and promoting strengths. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (4th ed., pp. 86-99). Hoboken, NJ: John Wiley.

47.

Lopez

S. J.

Magyar-Moe

J. L.

Petersen

S. E.

Ryder

J. A.

Krieshok

T. S.

O’Byrne

K. K.

. . . Fry

N. A.

(2006). Counseling psychology’s focus on positive aspects of human functioning. The Counseling Psychologist, 34, 205-227.

48.

Luhtanen

Crocker

(1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18, 302-318.

49.

Moradi

(2005). Advancing womanist identity development: Where we are and where we need to go. The Counseling Psychologist, 33, 225-253.

50.

Moradi

Yoder

J. D.

(2011). Current status and future directions in research on women’s experiences. In Altmaier

E. M.

Hansen

J. C.

(Eds.), The Oxford handbook of counseling psychology (pp. 346-374). New York, NY: Oxford University Press.

51.

Moradi

Fischer

A. R.

Hill

M. S.

Jome

L. M.

Blum

S. A.

(2000). Does “feminist” plus “therapist” equal “feminist therapist”? An empirical investigation of the link between self-labeling and behaviors. Psychology of Women Quarterly, 24, 285-296.

52.

Moradi

Subich

L. M.

(2002a). Feminist identity development measures: Comparing the psychometrics of three instruments. The Counseling Psychologist, 30, 66-86.

53.

Moradi

Subich

L. M.

(2002b). Perceived sexist events and feminist identity development attitudes: Links to women’s psychological distress. The Counseling Psychologist, 30, 44-65.

54.

Moradi

Subich

L. M.

Phillips

J. C.

(2002). Revisiting feminist identity development theory, research, and practice. The Counseling Psychologist, 30, 6-43.

55.

Morokoff

P. J.

Quina

Harlow

L. L.

Whitmire

Grimley

D. M.

Gibson

P. R.

Burkholder

G. J.

(1997). Sexual Assertiveness Scale (SAS) for women: Development and validation. Journal of Personality and Social Psychology, 73, 790-804.

56.

Murnen

S. K.

Smolak

(2009). Are feminist women protected from body image problems? A meta-analytic review of relevant research. Sex Roles, 60, 186-197.

57.

Myers

T. A.

Crowther

J. H.

(2007). Sociocultural pressures, thin-ideal internalization, self-objectification, and body dissatisfaction: Could feminist beliefs be a moderating factor? Body Image, 4, 296-308.

58.

Olkin

(2009). Disability-affirmative therapy. In Marini

Stebnicki

M. A.

(Eds.), The professional counselor’s desk reference (pp. 355-369). New York, NY: Springer.

59.

Peterson

R. D.

Tantleff-Dunn

Bedwell

J. S.

(2006). The effects of exposure to feminist ideology on women’s body image. Body Image, 3, 237-246.

60.

Pieterse

A. L.

Evans

S. A.

Risner-Butner

Collins

N. M.

Mason

L. B.

(2009). Multicultural competence and social justice training in counseling psychology and counselor education: A review and analysis of a sample of multicultural course syllabi. The Counseling Psychologist, 37, 93-115.

61.

Prilleltensky

(2003). Synergies for wellness and liberation in counseling psychology. The Counseling Psychologist, 31, 273-281.

62.

Rederstorff

J. C.

Buchanan

N. T.

Settles

I. H.

(2007). The moderating roles of race and gender-role attitudes in the relationship between sexual harassment and psychological well-being. Psychology of Women Quarterly, 31, 50-61.

63.

Rickard

K. M.

(1987, March). Feminist identity development: Scale development and initial validation studies. Paper presented at the annual meeting of the Association for Women in Psychology, Denver, CO.

64.

Rickard

K. M.

(1989). The relationship of self-monitored dating behaviors to level of feminist identity on the Feminist Identity Scale. Sex Roles, 20, 213-226.

65.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

66.

Rudman

L. A.

Fairchild

(2007). The F word: Is feminism incompatible with beauty and romance? Psychology of Women Quarterly, 31, 125-136.

67.

Rudolfsdottir

A. G.

Jolliffe

(2008). “I don’t think people really talk about it that much”: Young women discuss feminism. Feminism & Psychology, 18, 268-274.

68.

Ryff

C. D.

(1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57, 1069-1081.

69.

Schafer

Graham

J. W.

(2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147-177.

70.

Schick

V. R.

Zucker

A. N.

Bay-Cheng

L. Y.

(2008). Safer, better sex through feminism: The role of feminist ideology in women’s sexual well-being. Psychology of Women Quarterly, 32, 225-232.

71.

Schreiber

(2002). Injecting a woman’s voice: Conservative women’s organizations, gender consciousness, and the expression of women’s policy preferences. Sex Roles, 47, 331-342.

72.

Smith

Ferree

M. M.

Miller

F. D.

(1975). A short scale of attitudes toward feminism. Representative Research in Social Psychology, 6, 51-56.

73.

Speight

S. L.

Vera

E. M.

(2008). Social justice and counseling psychology: A challenge to the profession. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (4th ed., pp. 54-67). Hoboken, NJ: John Wiley.

74.

Spence

J. T.

Helmreich

R. L.

(1978). Masculinity and femininity: Their psychological dimensions, correlates, and antecedents. Austin: University of Texas Press.

75.

Spence

J. T.

Helmreich

R. L.

(1980). Masculine instrumentality and feminine expressiveness: Their relationships with sex-role attitudes and behaviors. Psychology of Women Quarterly, 5, 147-163.

76.

Springer

K. W.

Hauser

R. M.

(2006). An assessment of the construct validity of Ryff’s Scales of Psychological Well-Being: Method, mode, and measurement effects. Social Science Research, 35, 1080-1102.

77.

Stevens

(1986). Applied multivariate statistics for the social sciences. Hillsdale, NJ: Lawrence Erlbaum.

78.

Stewart

Love

(1968). A general canonical correlation index. Psychological Bulletin, 70, 160-163.

79.

Tabachnick

B. G.

Fidell

L. S.

(2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson.

80.

Troldahl

V. C.

Powell

F. A.

(1965). A short-form of the dogmatism scale for use in field studies. Social Forces, 44, 211-214.

81.

Twenge

J. M.

Zucker

A. N.

(1999). What is a feminist? Evaluations and stereotypes in closed- and open-ended responses. Psychology of Women Quarterly, 23, 591-605.

82.

Vallacher

R. R.

Wegner

D. M.

(1989). Levels of personal agency: Individual variation in action identification. Journal of Personality and Social Psychology, 57, 660-671.

83.

van Dierendonck

(2004). The construct validity of Ryff’s Scales of Psychological Well-being and its extension with spiritual well-being. Personality and Individual Differences, 36, 629-643.

84.

Vera

E. M.

Speight

S. L.

(2003). Multicultural competence, social justice, and counseling psychology: Expanding our roles. The Counseling Psychologist, 31, 253-272.

85.

Weathers

P. L.

Thompson

C. E.

Robert

Rodriguez

(1994). Black college women’s career values: A preliminary investigation. Journal of Multicultural Counseling and Development, 22, 96-105.

86.

Worell

(1996). Feminist identity in a gendered world. In Chrisler

J. C.

Golden

Rozee

P. D.

(Eds.), Lectures on the psychology of women (pp. 358-370). New York, NY: McGraw-Hill.

87.

Worell

(2001). Feminist interventions: Accountability beyond symptom reduction. Psychology of Women Quarterly, 25, 335-343.

88.

Worell

Remer

(2003). Feminist perspectives in therapy: Empowering diverse women (2nd ed.). Hoboken, NJ: John Wiley.

89.

Worell

Robinson

(1993). Feminist counseling/therapy for the 21st century. The Counseling Psychologist, 21, 92-96.

90.

Wyche

K. F.

Rice

J. K.

(1997). Feminist therapy: From dialogue to tenets. In Worell

Johnson

N. G.

(Eds.), Shaping the future of feminist psychology: Education, research, and practice (pp. 57-72). Washington, DC: American Psychological Association.

91.

Yoder

J. D.

Tobias

Snell

A. F.

(2011). When declaring “I am a feminist” matters: Labeling is linked to activism. Sex Roles, 64, 9-18.

92.

Yoder

J. D.

Perry

R. L.

Saal

E. I.

(2007). What good is a feminist identity? Women’s feminist identification and role expectations for intimate and sexual relationships. Sex Roles, 57, 365-372.

93.

Zucker

A. N.

(2004). Disavowing social identities: What it means when women say, “I’m not a feminist, but . . . .” Psychology of Women Quarterly, 28, 423-435.