Editorial Board: Editor: George H. Conklin, North Carolina Central University Board: Bob Davis, North Carolina Agricultural and Technical State University Richard Dixon, UNC-Wilmington Ken Land, Duke University Miles Simpson, North Carolina Central University Ron Wimberley, N.C. State University Robert Wortham, North Carolina Central University
Editorial Assistants John W.M. Russell, Technical Consultant Austin W. Ashe, North Carolina Central University
Volume 7, Number 2
Generational Difference in Feminist
Catherine E. Harnois
Wake Forest University
In June of 1998, the cover of Time magazine pictured the headshots of four prominent white women: Susan B. Anthony; Betty Friedan; Gloria Steinem; and (the only picture to appear in color) Ally McBeal. Under McBeal's picture, in bold red letters was the question, "IS FEMINISM DEAD?" Since this story appeared, debates about the decline of feminism, the significance of "Third Wave feminism" and meaning of "post-feminism" have occupied a prominent place within both popular culture and feminist scholarship. Put on the defensive by the feminist backlash in the 1980's, and by media proclaiming the death of feminism, Third Wave feminists argued that though today's feminist movement is different from that of previous generations, feminism is nonetheless alive and well among today's young women (see for example, Walker 1995, Heywood and Drake 1997, Baumgardner and Richards 2000). Today's young feminists may prefer blogging to staging public protests, and may aspire to be "career feminists" rather than "activists," but, from this perspective, are thought to embrace feminist ideals, and for the most part embrace feminist identities.
Social scientists who have investigated contemporary feminist generations have been somewhat less sanguine in their conclusions. While the overwhelming majority of research finds continued support for gender equality (e.g., Huddy et al. 2000; Bozendahl and Myers 2004; Aronson 2003), support for feminism and feminists is generally much more limited. Huddy et al. (2000:313) report that "Relatively few women and even fewer men report feminists as one of the political or sociodemographic groups to which they feel particularly close." Thirteen percent of women surveyed for one poll in 1995 even went so far as to call feminism "the work of the devil (Huddy et al. 2000, 339)."
Bivariate analyses from Peltola et al. (2004) and Huddy et al. (2000) support the claim that contemporary young women identify as feminist at rates similar to previous generations. But this finding is shown to be more complicated when multivariate models are employed. In their analysis of the 1992 National Election Study (NES) and the 1996 General Social Survey (GSS) Peltola et al. (2004:122) concluded that "Baby Bust women [those born from 1960-1978] are less apt to identify as feminist than are older women, once background characteristics and attitudes related to feminist identification are controlled." Similarly, in their multivariate models Schnittker et al. (2003:614) found that "both male and female respondents whose political coming-of-age coincides with the development of the feminist movement are more likely to think of themselves as feminists than are their older or younger counterparts." Though support for gender equality might be stable, generational experiences seem to shape how individuals choose to identify themselves (see also Whittier 1995).
Feminist Generations among African Americans
Though previous social science research and feminist scholarship more generally has yielded important insights concerning feminist generations, existing survey research is limited in the extent to which it draws meaningful conclusions about feminism among African Americans and other racial-ethnic minority groups (see Kane 2000). For example, in her analysis of data from the 1996 GSS, McCabe (2005) found that race was not a significant predictor of women's self-identification as feminist. Analyzing the same data, Schnittker et al. (2003) found no significant difference among Black and white women's feminist identities, but did find that "other nonwhite women" were more likely that white women identify as feminist. Peltola et al.'s (2004) analysis of the 1992 NES and 1996 GSS found more mixed results: race was a significant predictor of women's feminist identities in the former survey (with white women being less likely than racial-minority women to identify as feminist), but it was non-significant in their analysis of the latter. Taken as a whole, these studies imply that feminist identities among African Americans are shaped by the same processes that influence feminist identities among whites; they suggest that racial statuses play little role in shaping feminist identities.
A growing body of Black and multiracial feminist scholarship, however, suggests otherwise. Multiracial feminist theorists have argued that women's gendered experiences are fundamentally shaped by their location within the intersecting hierarchies of race, class and gender. Women do not experience gender simply as women, but as women of a particular race and class. Similarly, when responding to gender inequality, women respond not simply as "women," but as women with multiple identities, and multiple (and sometimes competing) interests. As a result, some racial and ethnic minority women who are committed to gender equality may instead choose to identify as "Black feminist," or "Womanist." Still others may prefer no gender-conscious label at all.
In addition to problematizing racially "neutral" understandings of feminist identities, multiracial feminist scholars have also challenged "hegemonic" narratives of feminist waves and generations (Thompson 2002). Most scholarship on US feminism describes three waves of feminist activity. The first wave is thought to correspond with the fight for Women's Suffrage and the second with the fight for "Women's Liberation" in the 1960s and 1970s. Scholars use the "third wave" to describe the surge in young women's feminist activism beginning in the early 1990s and continuing into the following decade. In contrast to this narrative of distinct waves of feminism, Roth (2004) has argued that what is generally thought of as one single "Second Wave" movement in the 1960s and 1970s in actuality was comprised of several distinct feminist movements, each with their own origins and timelines: a Black feminist movement, a Chicana feminist movement, and a primarily white women's movement. Thompson (2002) further argues that the alleged gap between the "Second Wave" and "Third Wave" of US feminism (i.e., the 1980s) corresponds with a comparatively vibrant period for racial and ethnic minority feminists. The 1980's saw, among other things, the publication of classic works of multiracial feminist theory, including This Bridge Called my Back (1981); Homegirls: A Black Feminist Anthology (1983); Sister / Outsider (1984), and Feminist Theory: From Margin to Center (1984). As Springer (2002, 1062) argues, "the more we learn about women of color's feminist activism, the less tenable the wave analogy becomes."
The sharp contrast between multiracial feminist theories and the models of feminist generations employed by survey researchers leads one to question the relevance of the wave model for African American feminism. How well do the generational differences uncovered by previous survey research apply to African American feminism? In this article, I analyze survey data from the National Black Feminist Study (2004-2005) to assess the relevance of feminist generations for the development of African Americans' feminist identities, and gender-conscious identities more generally. I use the term "gender-conscious identities" to refer to those identities which demonstrate individuals' recognition of gender inequality, and their view of this inequality as problematic. For purposes of this paper, gender-conscious identity includes self-identification as feminist, but also includes self-identification as "Black feminist," "Womanist" and "Africana Womanist." My results demonstrate that, along with gender, generation is critically important in shaping both feminist identities and gender-conscious identities more generally.
Data and Method
In order to assess generational differences in feminist and gender-conscious identities, this study employs both bivariate and multivariate regression analyses of survey data. While this general approach has been used in a number of other studies (e.g., Harnois 2008; McCabe 2005; Peltola et al. 2004; Schnittker et al. 2003) the current study differs in its particular approach in two important ways. First, the majority of survey research on feminist identities and attitudes has relied on racially diverse samples and has included race solely as a dichotomous independent variable (white or non-white, in some cases white, Black, and Other non-white). Modeling race in this way, with no interaction effects, assumes that particular sociodemographic characteristics affect women's feminist identities in the same way, regardless of their racial status. Because the number of white women in such analyses is significantly larger than the number of non-white women, the conclusions drawn may better reflect white women's experiences than African American women's. In contrast, this study analyzes data from the 2004-2005 National Black Feminist Study, a survey of self-identified African American respondents, and thus ensures that the findings of the analyses reflect the experiences of African Americans. A second benefit of the NBFS is that it allows us to examine a range of gender-conscious identities, rather than focusing merely on "feminist" identities. The NBFS asks respondents whether they identify as "Womanist," "Africana Womanist," "Feminist," and "Black Feminist," and the analyses presented here include all of these identities as "gender-conscious identities."
The NBFS is a national survey drawn from census tracts where at least 30% of the households are African-American, a sample frame that was "designed to be identical to the targeted frame used in the 1993-1994 National Black Politics Study." The original sample consists of 500 adult respondents (278 women, 222 men), all of whom identified as African American. In my analyses, I employ the available weights, which are designed "to reflect estimated population characteristics based on the 2002 Current Population Survey." These weights are "based on age category within gender, level of educational attainment and region of the US."
The dependent variable is gender-conscious identity, which is measured by the question "Would you describe yourself as a feminist, Black feminist, womanist, Africana Womanist or none of these?" Respondents who self-identified as any of these terms are coded "1" for holding a gender-conscious identity. Those who answered "None of these" were coded "0," and respondents reported "Don't know" or who refused to answer were excluded from the analysis.
For purposes of this study, the most important independent variables are those that assess feminist generations. Following Peltola et al. (2004), I create dummy variables for generational cohorts that are thought to correspond to feminist generations. The "Baby Boom" cohort includes individuals born between 1946 and 1959 (inclusive), the "pre-Baby Boom" cohort include those born prior 1946, and the "Baby Bust" cohort includes those born from 1960 - 1986.
Additional socio-demographic variables include gender (women=1, men=0), educational attainment (measured by two variables, "less than high school" and "college degree or beyond", where "high school graduate and/or some college" is the reference category) and total family income (an 11-category variable ranging from below $10,000 to $100,000 or more). To avoid the bias introduced by listwise deletion, missing values for family income were replaced with mean values. This change affected 14.5% of observations. Marital status is measured by two variables, "single or cohabitating" and "divorced or separated", where "married or widowed" is the reference category. Religious service attendance is measured with a four-category ordinal level variable ranging from to "never" (coded 1) to "at least once a week" (coded 4). I include a dichotomous variable to assess whether respondents are currently participating in the workforce (where full-time and part-time employment is coded 1, otherwise 0) and I also include an interaction term for Baby Bust generation and workforce status.
Table 1 shows the number of men and women respondents who hold particular gender-conscious identities. As expected, the proportion of women who self-identify as feminist is significantly higher than the corresponding proportion of men. What is perhaps more interesting, is the relationship between feminist identity and gender-conscious identity more generally. Of those women who hold some kind of gender-conscious identity, the majority identify as something other than simply "feminist." The same holds true for men, with more than two-thirds of those embracing a gender-conscious identity identifying as either Black Feminist, Womanist, or Africana Womanist.
Predicting Gender Conscious Identities
Table 2 presents the results from bivariate analyses of a variety of socio-demographic characteristics on women and men's gender-conscious identities. Three characteristics in particular stand out. As expected, women are shown to be significantly more likely to hold gender-conscious identities than are men. Although age per se is not statistically significant, generational differences are very significant, with individuals from the Baby Boom generation being more than three times as likely to identify as feminist as are individuals from other generations. Individuals from the Baby Bust generation, in contrast, are less likely to hold gender-conscious identities than are individuals from previous generations. Specifically, being born after 1960 decreases the odds of holding a gender-conscious identity by a factor of 0.447. Increased family income significantly decreases the likelihood of individuals' holding a gender-conscious identity, a result consistent with Peltola et al.'s (2004) analyses of both the GSS and NES. In the bivariate models, education, marital status, frequency of religious attendance, and participation in the paid labor force show no statistically significant effect on gender-conscious identities.
Regression coefficients reflect employed weights.
* significant at 5%; ** significant at 1%;
*** significant at 0.1% (two-tailed tests)
Table 3 presents the odds ratios (Exp[B]) from multivariate analyses of gender-conscious identities on socio-demographic variables. Model 1 again depicts the bivariate relationship between gender and feminist identity, where women are shown to be more likely to embrace a gender-conscious identity than are men. Model 2 builds on the first model by including information about respondents' generation and all three variables are found to be statistically significant. Respondents from the Pre-Baby Boom generation, as well as respondents from the Baby Bust generation are less likely than women from the Baby Boom generation (the reference group) to hold a gender-conscious identity.
Note: Numbers in parentheses are standard errors.
Regression coefficients reflect employed weights.
N reflects unweighted number of cases.
* significant at 5%; ** significant at 1%;
*** significant at 0.1% (two-tailed tests)
The third model builds on the previous model by including information about respondents' education, marital status, participation in the workforce, religious service attendance, as well as their family income. In this model we see that even when other sociodemographic variables are controlled for, gender and generation have significant effects on gender-conscious identities. Compared with men, women are shown to be 2.8 times as likely as to identify as feminist or womanist. Consistent with previous models, individuals in the Baby Boom generation are much more likely to adopt gender-conscious identities than are earlier or later generations. Interestingly, participation in the paid labor force also affects gender-conscious identities, but its effects differ for individuals of different generations. For those of the pre-Baby Boom and Baby Boom generation, participation in the labor force decreases the likelihood of identifying as feminist (Exp[b] = 0.347). For individuals of the Baby Bust cohort, aged 18-44 at the time of this survey, working significantly increases the likelihood of their adopting a gender-conscious identity. Baby Bust respondents who were currently working, either full time or part time, at the time of this survey were nearly six times more likely to identify as feminist, as compared with Baby Bust respondents who were not working.
In order to compare these results with those of other studies, I include two additional models, which are run separately for men and women respondents. Model 4 largely echoes the same relationships found in the previous models: Women of the Pre-Baby Boom and Baby-Bust generations are significantly less likely to identify as feminist compared with women of the Baby Boom cohort. Working again is shown to have a particularly strong effect for Baby Bust women. Compared to those Baby Bust women who were not currently working, those who were working either part time or full time were eighteen times as likely to identify as feminist. Model 5 shows the odds ratios from the logistic regression analysis for African American men. In this model, only one independent variable is shown to be significant: family income, which is shown to be inversely related to the likelihood of men's adopting a gender-conscious identity.
Taken as a whole the results suggest that generation plays an important role in shaping gender-conscious identities, though this effect may be limited to women. While it is tempting to compare the strength and significance of particular coefficients across models 4 and 5, doing so is somewhat problematic due to the difference in the number of men and women respondents in these models (179 and 285 respectively).
Discussion and Conclusion
This paper began by describing the contrast between multiracial feminist theories of feminism and the models of feminist generations employed by sociological survey researchers. In particular, multiracial feminist theory suggests that race likely effects women's choice of gender-conscious identities and that the Wave model of US feminism may misrepresent feminism within the Black community. Three main conclusions stand out. First, when given a choice among several different gender-conscious identities, African American women and men are more likely to describe themselves as Black feminists, Womanists or Africana Womanists than they are to describe themselves as "feminists." This finding is consistent with multiracial feminist theorists' suggestion that 'feminist" may be a racially biased identity. Some African American women and men certainly do identify as feminist, but more than two thirds of those who claim a gender-conscious identity will choose one other than simply "feminist."
A second contribution of this study concerns the importance of generational difference in African American feminism. Here the results are largely consistent with previous sociological survey research. Both the bivariate and the multivariate regression analyses suggest that generation is critically important for shaping gender-conscious identities. As in Schnittker's et al.'s (2003) and Peltola et al.'s (2004) studies, when socio-demographic characteristics are controlled for, individuals in the Baby Boom generation are more likely than either earlier or later generations to adopt a gender-conscious identity. This particular finding poses a challenge for feminist theorists who have argued that feminism is alive and well among the younger generation. Feminist beliefs may be alive and well, but gender-conscious identities, at least among African Americans, are not as common among the Baby Bust generation as they are among Baby Boomers. This finding poses a further challenge for those who have suggested that the Wave model of US feminism misrepresents the experiences of African American feminism. At least in terms of the differences between the "second" and "third" waves, generation is an important predictor of gender-conscious identities.
A third contribution of this study concerns the relationship among generation, working in the paid labor force, and adopting a gender-conscious identity. Though previous studies include work force participation in their models, most have found it to be non-significant in predicting feminist identities (e.g., Schnittker et al. 2003; McCabe 2005). Peltola et al. (2004) run separate regression models for women of the Baby Boom, Pre-Baby Boom, and Baby Bust generations and find that employment does not significantly predict the likelihood of respondents' identifying as feminist. The results from this study suggest that Baby Bust individuals, and women in particular (Table 3, Models 3 and 4), are much more likely to adopt a gender-conscious identity if they are currently working either part-time or full-time. For African Americans of the Baby Bust generation, working in the paid labor force seems an especially important event in the development of gender-conscious identities.
Further research is needed to understand the presence of generational divisions among African American feminists and womanists. African American feminism may not have experienced the same degree of abeyance in the 1980s as did white women's feminism, but what, then, brought about the generational differences that we see here? It is possible that generational decline in gender-conscious identities is the result of the abeyance of the Civil Rights Movement; it also might result from the general conservatism of the Reagan era in which many of the Baby Bust individuals were raised. A third possibility is that individuals raised with Title IX, Affirmative Action, and who were born after Roe v. Wade face less gender inequality in their younger years relative to earlier generations. If the interaction between working and Baby Bust generation tells us anything, it is that when young African Americans go to work, and when young African American women in particular work, they are more likely to adopt gender-conscious identities. Perhaps work-related experiences help to reveal the extent of gender inequality to young women, and thus encourage them to adopt gender-conscious identities? Further research is needed, but the results here suggest this is a strong possibility.
Aronson, Pamela. 2003. "Feminists or Postfeminists Young women attitudes toward Feminism and Gender Relations" in Gender & Society 17: 903-922.
Baumgardner, Jennifer and Amy Richards. 2001. Manifesta. New York: Farrar, Straus, and Giroux.
Bolzendahl, Catherine I. and Daniel J. Myers. 2004. "Feminist Attitudes and Support for Gender Equality: Opinion Change in Women and Men, 1974-1998." Social Forces 83:759-790.
Baca Zinn, Maxine and Bonnie Thornton Dill. 1996. "Theorizing Difference from Multiracial Feminism." Feminist Studies 22:321-33
National Black Feminist Study, 2004-2005. CSRA/UCONN Poll # 2004-FEM, Study # USCSRA2004-FEM. Sponsored by Evelyn M. Simien. Acquired from the Roper Center for Public Opinion Research.
Dicker, Rory and Alison Piepmeier. 2003. Catching a Wave. Boston: Northeastern University Press.
Finland, Barbara 1995. Listen Up: Voices from the Next Feminist Generation. Seattle: Seal.
Heywood, Leslie and Jennifer Drake (eds.) 1997. Third Wave Agenda: Being Feminist, Doing Feminism. Minneapolis: University of Minnesota Press.
Harnois, Catherine E. 2008. "Representing Feminism: Past, Present Future." NWSA Journal 20:120-145.
Huddy, Leonie, Francis K. Neely and Marilyn R LaFay. 2000. "The Polls-Trends: Support for the Women Movement" in Public Opinion Quarterly 64: 309-50.
Kane, Emily. 2000. "Racial and Ethnic Variations in Gender-Related Attitudes." Annual Review of Sociology 26: 419-39.
Mann, S. A. and D. J. Huffman. 2004. "The Decentering of Second Wave Feminism and the Rise of the Third Wave" in Science and Society 69: 56-91.
McCabe, Janice. 2005. "What's in a Label? The Relationship between 'Feminist' Self-Identification and Feminist Attitudes among U.S. Women and Men." Gender & Society 19: 480-505.
Peltola, Pia, Melissa A. Milkie, Stanley Presser. 2004. The "Feminist" Mystique: Feminist Identity in Three Generations of Women. Gender & Society 18: 122-144.
Purvis, Jennifer. 2004. "Grrrls and Women Together in the Third Wave: Embracing the Challenges of Intergenerational Feminism(s)" in NWSA Journal 16: 93-123.
Roth, Benita. 2004. Separate Roads to Feminism: Black, Chicana, and White Feminist Movements in America Second Wave. Cambridge, UK: Cambridge University Press.
Schnittker, Jason, Jeremy Freese and Brian Powell. 2003. "Who are Feminists and What Do They Believe?" in American Sociological Review 68: 607-622.
Siegel, Deborah L. 1997. "Reading between the Waves: Feminist Historiography in a 'Postfeminist' Moment" in Third Wave Agenda. edited by Heywood and Drake:55-82.
Springer, Kimberly. 2002. "Third Wave Black Feminism?" in Signs 27:1057-1082.
Thompson, Becky. 2002. "Second Wave Feminism in the United States." Feminist Studies 28:336-360.
Walker, Rebecca. 1995. To be Real. New York: Anchor Books.
Whittier, Nancy. 1995. Feminist Generations.
Philadelphia: Temple University Press.
©2009 by the North Carolina Sociological Association