Sex Roles

, Volume 67, Issue 3, pp 131–142

Feminism and Attachment Parenting: Attitudes, Stereotypes, and Misperceptions

Authors

    • Department of PsychologyUniversity of Mary Washington
  • Mindy J. Erchull
    • Department of PsychologyUniversity of Mary Washington
Original Article

DOI: 10.1007/s11199-012-0173-z

Cite this article as:
Liss, M. & Erchull, M.J. Sex Roles (2012) 67: 131. doi:10.1007/s11199-012-0173-z

Abstract

This paper investigated attitudes and stereotypes about what feminist women, primarily from the United States, believed about a number of practices associated with attachment parenting which is theorized to be both feminist and non-feminist. The goals of this study were to determine whether feminists endorsed attachment parenting and whether stereotypes of feminists’ beliefs corresponded to actual feminists’ attitudes. Women were recruited online, primarily through blogs, to complete an online survey about feminism and mothering. Four hundred and thirty one women comprised the sample for the current investigation and included heterosexual-identified feminist mothers (n = 147), feminist non-mothers (n = 75), non-feminist mothers (n = 143), and non-feminist non-mothers (n = 66). Participants were asked to rate their own attitudes towards specific practices associated with attachment parenting and to indicate their perceptions of the beliefs of the typical feminist. Results indicated that feminists were more supportive of attachment parenting practices than were non-feminists. Non-feminists, particularly mothers, held misperceptions about the typical feminist, seeing them as largely uninterested in the time-intensive and hands-on practices associated with attachment parenting. Feminist mothers also held stereotypes about feminists and saw themselves as somewhat atypical feminists who were more interested in attachment parenting than they thought was typical of feminists. Our data indicated that feminists did endorse attachment parenting and that stereotypes of feminists related to attachment parenting are untrue. Furthermore, the role of feminism in the identity of feminist mothers and whether attachment parenting is truly a feminist way to parent are discussed.

Keywords

FeminismMotheringAttachment parentingStereotypesSocial normsSocial identity theory

Introduction

Feminists are frequently positioned in the U.S. media as anti-family and anti-motherhood (Dillaway and Paré 2008; Faludi 1991). The stereotypical assumption that feminists are uninterested in caring for children has contributed to the backlash against the feminist movement (Faludi 1991; Feder 2006). Despite this, feminist women may support practices associated with attachment parenting (Bobel 2008; Friedman 2008) a particularly intensive and involved parenting style that some (e.g., Etelson, 2007) argue is a feminist way to parent while others (e.g., Jong 2010) argue is inherently non-feminist. The goal of this paper was to ascertain whether feminist and non-feminist women, mostly from the United States, differed in their endorsement of behaviors associated with attachment parenting and whether perceptions of feminist parenting were accurate. We also wanted to look at potential differences between mothers and non-mothers resulting in a two by two design (feminist self-identification by motherhood status). An online survey using a sample largely drawn from the United States was used to assess individuals’ own beliefs about a number of parenting practices as well as their stereotypes about the beliefs held by the typical feminist.

Though feminists have done a great deal of theorizing about motherhood (e.g., Friedan 1981; Ehrenreich and English 1978; Rich, 1976; Ruddick 1989; Umansky 1996), it remains unclear what it means to be a feminist mother and to parent in a feminist way. This is, in large part, due to a lack of empirical data investigating the beliefs of feminist and non-feminist identified women about parenting practices. Practices associated with attachment parenting have served as a nexus of debate about what it means to be a feminist parent (see Bobel 2008; Jong 2010; Etelson, 2007). While some feminists explicitly point to attachment parenting as feminist (e.g., Etelson, 2007), others explicitly point to it as anti-feminist (e.g., Jong 2010). Despite these active debates, whether feminist-identified women endorse beliefs associated with attachment parenting more so than do non-feminists is unknown. This research is aimed to elucidate whether attachment parenting practices are specifically endorsed by feminist women. Determining whether feminist women endorse these beliefs will help sharpen the conversation about whether or not attachment parenting is actually an empowering or an oppressive way to parent. Furthermore, an additional goal of this research is to understand whether stereotypes about feminist parenting match realities. Since the backlash has positioned feminists as anti-family (see Faludi 1991; Feder 2006) but feminists may indeed be very engaged with childcare (see Bobel 2008; Friedman 2008), this research may enhance arguments against the stereotyped image of feminists as anti-parenting. Specific attitudes about parenting and stereotypes of what feminists believe about parenting may vary internationally. Nevertheless, whether or not feminists endorse attachment parenting practices and the notion that there are misperceptions about the extent to which feminists are interested in these practices are issues that likely cross national boundaries.

Feminism and Motherhood

Historically, there has been a tension between feminism and motherhood (Simons 1984). In the early part of the second wave of the feminist movement, some feminists explicitly took the stance that motherhood was a significant source of women’s oppression (Snitow 1992; Umansky 1996). Other feminists have expressed concern about how one can be both a good feminist and a good mother (Ehrenreich and English 1978; Friedan 1981), though what it means to be either a good feminist or a good mother is unclear. Feminists have questioned whether the desire to mother is truly autonomous or the result of a patriarchal delusion and the over-valorization of a motherhood ideal (Tietjens Meyers 2001). On the other hand, feminists have noted that, while the institution of motherhood can be oppressive, the actual experience of being a mother can be empowering (Rich 1976). Furthermore, many feminist scholars hold a positive view of motherhood and have specifically pointed to motherhood as a source of women’s strength (Ruddick 1989). More recently, feminist writers have focused on the importance of recognizing that feminist mothers can experience a true, genuine desire to care for children (de Marneffe 2004). Despite the complex relationship between feminism and mothering, the stereotype that feminists are anti-motherhood has been exploited by conservatives who have accused feminists of destroying the family (Douglas and Michaels 2004; Faludi 1991; Feder 2006).

Although a theoretical literature exploring the relationship between feminism and mothering exists (e.g., Friedan 1981; Ehrenreich and English 1978; Rich, 1976; Ruddick 1989; Umansky 1996), there is little quantitative data on this topic. The quantitative research to date has largely focused on the desire of feminists to become mothers. This research showed that feminist beliefs were negatively related to the extent to which young, undergraduate women in the United States desired to be mothers (Gerson 1980, 1984). However, many feminists are mothers, and empirical data on the attitudes towards parenting of feminist-identified mothers is lacking. There are two books that have explored this topic through interviews and essays with and by feminist mothers (Gordon 1990; O’Reilly 2008). These authors concluded that what it means to be a feminist mother is complex but includes challenging the myth of motherhood, raising children in anti-sexist ways, combining motherhood with paid employment, and expecting fathers to participate equally in childrearing (Gordon 1990). O’Reilly (2008) also specifically argued that the focus of feminist motherhood is to empower women to resist the notion that there is one prescribed way to be a “good mother.”

Attachment Parenting

Attachment parenting is a child-centric parenting technique in which children’s needs are ideally met on the child’s schedule rather than that of the parent. The techniques of attachment parenting include extended breastfeeding and breastfeeding on demand, co-sleeping, and frequently carrying one’s baby in wraps and slings (Green and Groves 2008; Sears and Sears 2003). These techniques are designed to help minimize the boundaries between parents and children. This, theoretically, increases the ability to create a strong attachment bond (Schon and Silven 2007) and has been argued to be a natural way to bond with one’s children (Etelson 2007; Schon and Silven 2007). There is little empirical data on the parenting behaviors of women who practice attachment parenting. However, one study, surveying an international sample of mothers, primarily drawn from the U.S. and Canada, found that women who self-identified as attachment parents engaged in behaviors including extended breastfeeding, maintaining a child-centric schedule rather than one driven by the mother, co-sleeping, and holding one’s child often, especially during the transition to sleep (Green and Groves 2008). Additionally, the majority of the attachment parents surveyed reported that the child spent their time almost exclusively in the care of the mother.

Feminism and Attachment Parenting

The limited data that exists on the extent to which attachment parenting is practiced and endorsed has focused on samples of women for whom feminist identification was not assessed. However, within the theoretical literature on feminism and mothering, attachment parenting is a central part of the debate and discussion about what it means to parent in a feminist way. Attachment parenting is associated with other forms of natural parenting and environmentally conscious parenting decisions and, as such, has been framed as appealing to feminist women (Bobel 2008). Other feminists have argued that attachment parenting practices specifically align with their feminist values and should be seen as a feminist way to parent (Etelson 2007). In a recent online debate in the NY Times forums, attachment parenting was advocated as an explicitly feminist way to parent in one opinion piece (Bialik 2012).

Some of the specific components of attachment parenting have also been framed as appealing to feminist women. Breastfeeding, a component of attachment parenting has been framed as a feminist act because it is a more natural, less medicalized practice (Thanyachareon 2010). One feminist scholar specifically offered co-sleeping as a way to develop an intense bond with one’s child while away from that child during that day due to work (Conrad 2009). Since shared co-parenting is often seen as a feminist way to parent and carrying one’s child and co-sleeping are behaviors in which both parents can engage, Etelson (2007) explicitly argued that attachment parenting is a way for both parents, not solely mothers, to develop strong relationships with their children.

On the other hand, other feminist theorists have positioned attachment parenting as anti-feminist. Attachment parenting, if it focuses only on the mother’s duty to engage in these behaviors, has been critiqued as being oppressive to women and a source of guilt (Eyer 1996). In the book The Conflict, Elisabeth Badinter (2011) argued that the practices associated with attachment parenting are oppressive for women. Even feminists who acknowledge the appeal of attachment parenting among feminist women note that it fails to resist patriarchal assumptions about motherhood (Bobel 2008; Friedman 2008). It has been argued that attachment theory, from which attachment parenting is derived, is a theoretical position that rose from a history of misogyny and cannot be salvaged as feminist (Franzblau 1999). Furthermore, attachment parenting techniques are extremely intensive and time consuming, especially if it is the mother alone who must engage in them, and feminists have frequently critiqued the motherhood ideal and the extent to which one must intensively engage with one’s children (Arendell 2000; Douglas and Michaels 2004; Hays 1996; O’Reilly 2008). Some of the specific components of attachment parenting, specifically breastfeeding, have been problematized by feminists as enforcing the primacy of the mother and reducing the potential for co-parenting (Friedman 2008). Other feminists have argued that the imperative to breastfeed can be oppressive to women who are unable or unwilling to do so (Crossley 2009).

Whether or not attachment parenting is an explicitly feminist or non-feminist way to parent, feminist women may indeed be engaging in this set of time-intensive parenting practices. It has been argued that a combination of rhetorical silence about mothering paired with an emphasis on personal choice in the decision to have a child has led feminists to actually embrace intensive parenting practices (O’Brien Hallstein 2008). Thus, whether feminist women support the tenets of attachment parenting and do so more than non-feminist identified women remains an empirical question, and investigating this is the central goal of the present study.

Attachment Parenting and Motherhood Status

Although no one has specifically investigated whether mothers endorse attachment parenting practices more so than do non-mothers, it is reasonable to expect that they should. Research conducted in the U.S. with both first and second time parents has suggested that attitudes shift after the birth of a child such that both men and women’s beliefs about gender become more traditional (Katz-Wise et al. 2010). One traditional view is that motherhood should be of central importance to women. Given that mothers may well be placing more importance on the way people parent than do non-mothers, we would anticipate that mothers would endorse attachment parenting practices more so than would non-mothers.

Perceptions of Feminist Mothers

A secondary goal of the present investigation was to determine whether perceptions of feminists’ attitudes about attachment parenting corresponded to the actual attitudes reported by feminist mothers and non-mothers. Given that stereotypes that feminist women are anti-motherhood have been a component of the backlash against feminism (Faludi 1991; Feder 2006), research that both documents and potentially counters these stereotypes is important.

Research in the United States using undergraduate samples has suggested that a variety of stereotypical beliefs about feminists are pervasive. Some of these stereotypes are relatively benign; for example, feminists were seen as liberal and interested in promoting equality (Berryman-Fink and Verderber 1985; DeWall et al. 2005; Houvouras and Carter 2008; Twenge and Zucker 1999). Feminists were also seen as being career-oriented and frequently working outside of the home (Berryman-Fink and Verderber 1985; Twenge and Zucker 1999). More negative stereotypes of feminists also exist; for example, feminists were believed to be unattractive (Rudman and Fairchild 2007) and to dislike men (Houvouras and Carter 2008). Research has also shown that undergraduates believed that feminists were lesbian or butch (Houvouras and Carter 2008). Research has suggested that many of the negative stereotypes about feminists are actually unfounded. For example, contrary to the stereotype that feminists are man-hating and not interested in romantic relationships, research conducted with undergraduates in the United States has demonstrated that feminists were actually less hostile to men than were non-feminists (Anderson et al. 2009). Additionally, other research in the U. S. using both undergraduates and a broader internet-recruited sample found that men with feminist partners reported higher relationship stability and sexual satisfaction than did those with non-feminist partners (Rudman and Phelan 2007). Given this, we anticipated that non-feminists’ stereotypes about feminist women in regards to attachment parenting beliefs would be inaccurate and represent misperceptions.

The Current Study

In this study, we sought to investigate the beliefs of feminist and non-feminist mothers and non-mothers as well as their perceptions of the beliefs held by the typical feminist about a number of practices associated with attachment parenting. While feminism is a complex construct and there are many different forms of feminism, data from an undergraduate U.S. sample suggested that actually self-identifying as a feminist has meaningful implications for one’s attitudes, beliefs, and behaviors (Yoder et al. 2011). Given this, we were interested in how self-identified feminist women differed on their beliefs about parenting practices from those who would not self-identify as feminist.

We assessed beliefs about four different practices associated with attachment parenting: breastfeeding for extended periods of time, co-sleeping, carrying one’s child often, and setting strict schedules for one’s child (Green and Groves 2008; Schon and Silven 2007; Sears and Sears 2003). High endorsement of the first three practices and low endorsement of the importance of setting strict schedules are consistent with beliefs in attachment parenting (Sears and Sears 2003).

We had a number of specific hypotheses. We anticipated that, overall, (1) mothers would endorse attachment parenting more so than would non-mothers. Given the recent theoretical arguments connecting feminism and attachment parenting practices (e.g., Etelson 2007; O’Brien Hallstein 2008) and evidence that feminist mothers are interested in attachment parenting (e.g., Bobel 2008; Friedman 2008), we hypothesized that (2) feminists would more strongly support an attachment parenting perspective than would non-feminists. We did not anticipate any interactions.

We were also interested in whether we could identify any specific misperceptions about beliefs held by feminists. In order to answer this question, we compared self-ratings to ratings for the typical feminist for non-feminist non-mothers, non-feminist mothers, feminist non-mothers, and feminist mothers separately. We hypothesized that both non-feminist mothers and non-mothers would (3) perceive themselves as different from the typical feminist and that (4) these differences would represent misperceptions. In contrast to our expectations for non-feminist women, (5) we did not anticipate any misperceptions about feminist beliefs regarding attachment parenting among either feminist mothers or non-mothers.

Method

Procedure

Participants were recruited online using a variety of methods. Authors of many blogs concerning feminism and/or mothering as well general interest blogs targeted toward young women were contacted and asked to post a request for participants. Participants were also recruited via standardized requests through facebook posts. The request was also made to pass the link to the survey on to other women who might be interested; thus, a snowball sampling technique was used as well. Finally, the survey was publicized as a volunteer opportunity on Craigslist for a number of metropolitan areas throughout the United States. All recruitment messages contained a brief summary of the research topic identifying the research topic as “attitudes about mothering” and a link to the online survey. Messages specified that we were recruiting “feminists, non-feminists, mothers, and non-mothers.” The informed consent, the landing page for the online survey, described the study as one about “the relationship between mothering and feminism.” Recruitment messages were primarily posted on blogs from the United States. However, a few blogs outside of the U.S. did post our message, and individuals worldwide can access any blogs. While we believe that our sample was largely drawn from the United States, it is likely that it included some individuals from other countries.

Participants

Selection

In total, 1723 women completed the larger online survey. For the present study, only a portion of these women met inclusion criteria. Women were eligible for inclusion in the analyses for the current study if they were between the ages of 18 and 50, as these women would be most likely to either be engaged in parenting or actively considering issues related to parenting, and were heterosexual. All of the non-feminists, both mothers and non-mothers, who met these criteria were included in the analyses for the present study. However, given that the survey was described as a study about feminism and mothering, a disproportionate number of respondents self-identified as feminists (86.6 %), as these women were likely more interested in completing a survey on the topic of feminism. Thus, we chose to select a subset of the self-identified feminist women who met the inclusion criteria for this study, using a random number generator, in an effort to create more equivalent group sizes. The number of feminist mothers and non-mothers selected for inclusion was based, roughly, on the percentage of non-feminist mothers and non-mothers in our sample. Among the 209 non-feminists who completed our online survey and met the inclusion criteria, 66 were non-mothers (31.6 %) and 143 were mothers (68.4 %). We randomly selected 222 feminist participants for inclusion: 75 feminist non-mothers (33.8 %) and 147 feminist mothers (66.2 %). This resulted in a sample of 431 women for the present study.

Demographic Characteristics and Group Differences

Demographic characteristics (i.e., age, education, socio-economic status, race/ethnicity, and marital status) divided by the four groups of participants are presented in Table 1. Country of origin was not asked.
Table 1

Demographic characteristics by group

 

Non-feminist Non-mothers

Non-feminist Mothers

Feminist Non-Mothers

Feminist Mothers

N = 66

N = 143

N = 75

N = 147

Age

 Mean

25.71ac

32.28b

26.81ac

34.75d

 SD

4.51

5.89

6.08

6.41

 Median

25.5

32

25

34

 Range

19–41

19–49

19–46

21–50

Education

M = 5.21ab (SD = 1.27)

M = 5.00a (SD = 1.47)

M = 5.57b (SD = 1.38)

M = 5.73b (SD = 1.43)

 

N, %

N, %

N, %

N, %

 1 = grade school

 2 = some high school

2, 1.4 %

1, .7 %

 3 = high school graduate

2, 3 %

15, 10.5 %

3, 4 %

2, 1.4 %

 4 = some college/associates degree

23, 34.8 %

46, 32.2 %

17, 22.7 %

33, 22.4 %

 5 = college graduate

17, 25.8 %

39, 27.3 %

18, 24 %

37, 25.2 %

 6 = some graduate school

8, 12.1 %

9, 6.3 %

14, 18.7 %

16, 10.9 %

 7 = masters level degree

15, 22.7 %

23, 16.1 %

17, 22.7 %

42, 28.6 %

 8 = doctoral level degree

1, 1.5 %

9, 6.3 %

6, 8 %

16, 10.9 %

SES

M = 2.92a (SD = .76)

M = 2.73a (SD = .76)

M = 3.01a (SD = .73)

M = 2.90a (SD = .79)

 

N, %

N, %

N, %

N, %

 1 = poverty

1, 1.5 %

8, 5.6 %

1, 1.3 %

7, 4.8 %

 2 = working class

17, 25.8 %

41, 28.7 %

15, 20 %

31, 21.1 %

 3 = middle class

31, 47 %

75, 52.4 %

41, 54.7 %

78, 53.1 %

 4 = upper middle class

17, 25.8 %

19, 13.3 %

16, 21.3 %

29, 19.7 %

 5 = wealthy

1, 1.3 %

1, .7 %

 not disclosed

1, 1.3 %

1, .7 %

Race/Ethnicity

 

N, %

N, %

N, %

N, %

 White/Caucasian

60, 90.9 %

128, 89.5 %

68, 90.7 %

127, 86.4 %

 African American

2, 3 %

5, 3.5 %

1, 1.3 %

3, 2 %

 Asian/Pacific Islander

1, .7 %

1, 1.3 %

4, 2.7 %

 Latina

1, 1.5 %

1, 1.3 %

1, .7 %

 American Indian

1, .7 %

1, .7 %

 Multiracial

2, 3 %

7, 4.9 %

2, 2.7 %

8, 5.4 %

 Other

1, 1.5 %

 not disclosed

1, .7 %

2, 2.7 %

3, 2 %

Marital Status

 

N, %

N, %

N, %

N, %

 married

12, 18.2 %

130, 90.9 %

20, 2.76 %

129, 87.8 %

 not married but in a committed relationship

19, 28.8 %

9, 6.3 %

18, 24 %

10, 6.8 %

 not married

35, 53 %

4, 2.8 %

37, 49.3 %

7, 4.8 %

 not disclosed

1, .7 %

Main effect differences are reported in the text, and there were no significant interactions. For completeness, differences between all four groups for the continuous variables are indicated with superscripts. Groups with different superscripts were significantly different at p < .05

A MANOVA where feminist self-identification and motherhood status were independent variables and age, education, and socio-economic status (SES) were dependent variables was run to determine whether there were group differences on these variables. This test was run to determine whether any of these variables should be included as covariates in analyses. These variables were selected as only continuous variables can be used as covariates, and education level and SES can be understood as continuous. The overall MANOVA was significant for both feminist self-identification, F(3, 423) = 5.71, p = .001, ηpartial2 = .04, and motherhood status, F(3, 423) = 54.06, p < .001, ηpartial2 = .28. Age differed significantly between both feminists and non-feminists, F(1,425) = 8.40, p = .004, ηpartial2 = .02, as well as between mothers and non-mothers, F(1,425) = 139.65, p < .001, ηpartial2 = .25. Feminists were older than non-feminists and mothers were older than non-mothers. Feminists were significantly more educated than non-feminists, F(1,425) = 13.79, p < .001, ηpartial2 = .03, but there was no difference by motherhood status. Non-mothers had higher self-reported SES than did mothers, F(1,425) = 4.77, p = .03, ηpartial2 = .01; there were no significant differences on SES for feminist self-identification. There were no significant interactions. Thus, age, education, and SES were used as covariates in all between group analyses. Differences between each of the four groups for these covariates can be seen in Table 1.

Measures

Demographics

Demographic questions included age, race, sexual orientation, education level, self-reported socio-economic status, whether the participants reported having any children, and whether they wanted children if they did not have children. Participants were also asked if they were married, single, or in a committed relationship.

Feminist Self-Identification

Participants were asked whether they considered themselves to be a feminist in a yes-no format. As a check that self-identified feminists held more feminist beliefs, differences on the short form of the Liberal Feminist Attitudes and Ideologies Scale (Morgan 1996) were tested. Self-identified feminists scored significantly higher (M = 5.46, SD = .46) than did non-feminists (M = 4.37, SD = .74), F(1, 420) = 337.02, p < .001, ηpartial2 = .45. The LFAIS is scored on a scale ranging from 1 (strongly disagree) to 6 (strongly agree) and was found to have acceptable reliability in both original (Cronbach’s alpha = .81) and the current investigation (Cronbach’s alpha = .89).

Parenting Practices

Participants were asked to provide information about their views on four different parenting practices. Participants were asked to indicate the extent to which they agreed with the following six statements on a scale ranging from 1 (strongly disagree) to 6 (strongly agree): Parents should carry their children as often as possible (e.g., using wraps and slings); It is important to co-sleep with your child (e.g., sleep in the same bed); It is important to set a strict a strict schedule for your child. Participants were also asked to indicate the length of time for which they felt it was important to breastfeed a child. Response options included 1 (not at all), 2 (less than 3 months), 3 (36 months), 4 (612 months), 5 (1218 months), 6 (more than 18 months). Participants were then asked to answer these questions as they believed the typical feminist would as has been done in previous research (Liss et al. 2000).

Operational Definition of Misperceptions

For the purpose of this study, we defined a misperception for non-feminist women as a statistically significant perceived difference between the self and the typical feminist that was in the opposite direction of the actual difference. For feminist women, misperceptions were defined as any significant difference between the self and the typical feminist.

Results

Self Ratings for Parenting Practices Variables

A 2x2 MANCOVA in which feminist self-identification and motherhood status were the two independent variables, each parenting practice was a dependent variable, and age, education, and SES were covariates was run in order to test hypotheses 1 through 6. The main effect for feminist self-identification was significant, F(4, 401) = 6.25, p < .001, ηpartial2 = .06, indicating that feminists, overall, had different perceptions about the parenting practices than did non-feminists. The main effect of motherhood status was also significant, F(4, 401) = 10.66, p < .001, ηpartial2 = .10, indicating that mothers had different perceptions about the parenting practices than did non-mothers. The interaction was not significant, F(4, 401) = .18, p = .95, ηpartial2 = .002. The covariate of age was significant, F(4, 401) = 2.72, p = .03, ηpartial2 = .03. The covariates of education, F(4, 401) = .49, p = .75, ηpartial2 = .005, and SES, F(4, 401) = .86, p = .49, ηpartial2 = .009, were not significant. Means, standard deviations, and univariate F-test results for the main effects, controlling for age, education, and SES can be seen in Table 2.
Table 2

Self ratings on all variables

Variable

Non-feminist Non-mothers

Non-feminist Mothers

Feminist Non-mothers

Feminist Mothers

Main Effect of Feminist Identification

Main Effect of Motherhood Status

N = 63

N = 137

N = 70

N = 141

M (SD)

M (SD)

M (SD)

M (SD)

Breastfeeding

3.44 (.84)

4.16 (1.39)

3.83 (1.31)

4.38 (1.46)

F(1, 404) = 6.12, p = .01, η2 = .02

F(1, 404) = 30.27, p < .001, η2 = .07

Carry Children

3.29 (1.39)

3.72 (1.60)

3.60 (1.37)

4.13 (1.44)

F(1, 404) = 6.30, p = .01, η2 = .02

F(1, 404) = 12.54, p < .001, η2 = .03

Co-sleep

2.11 (1.25)

2.70 (1.76)

2.60 (1.36)

3.27 (1.64)

F(1, 404) = 11.83, p = .001, η2 = .03

F(1, 404) = 14.37, p < .001, η2 = .03

Strict Schedule

3.78 (1.05)

3.13 (1.50)

3.13 (1.38)

2.55 (1.27)

F(1, 404) = 18.39, p < .001, η2 = .04

F(1, 404) = 19.93, p < .001, η2 = .05

Sample sizes differ from those in Table 1 as listwise deletion was used. Higher sores represent greater endorsement of the measured idea. Endorsement of all parenting practices were measured on a scale ranging from 1 to 6; for breastfeeding, 1 = not at all, 2 = less than 3 months, 3 = 3–6 months, 4 = 6–12 months, 5 = 12–18 months, 6 = more than 18 months. All η2 reported are partial η2

Univariate analyses, after controlling for age, education, and SES, indicated, in support of hypothesis 1, that mothers were more supportive of extended breastfeeding, carrying one’s child, and co-sleeping than were non-mothers, and non-mothers felt it was more important to set strict schedules than did non-mothers. Univariate analyses, after controlling for age, education, and SES, indicated that feminists were more likely to support extended breastfeeding, carrying one’s child, and co-sleeping than were non-feminists, and non-feminists were more likely to endorse strict schedules for one’s child, supporting hypothesis 2.

Differences Between Self and Typical Feminist Ratings for Non-feminist Non-mothers

Hypotheses 3 and 4 were tested for non-feminist non-mothers using within subject ANOVAs comparing self ratings with ratings for the typical feminist. The purpose of this analysis was to determine whether these women perceived differences between themselves and the typical feminist (hypothesis 3) and whether these differences would represent misperceptions as operationalized above (hypothesis 4). Table 3 gives F-test results for each of these analyses. The results of the ANOVAs indicated that non-feminist non-mothers perceived differences between themselves and the typical feminist in terms of extended breastfeeding and co-sleeping, partially supporting hypothesis 3 for this group. For breastfeeding, the differences were in the opposite direction of the actual differences between feminists and non-feminists; in other words, while feminists placed greater importance on extended breastfeeding than did non-feminists, non-feminist non-mothers saw the typical feminist as valuing extended breastfeeding less than they themselves did, supporting hypothesis 4 for this variable. The perceptions about the importance placed on co-sleeping represented an accurate perception of the difference between feminist and non-feminist women, in contrast to hypothesis 4.
Table 3

Self versus typical feminist within-subject ANOVA results for non-feminist and feminist mothers and non-mothers

 

Breastfeeding

Carry Children

Co-sleep

Strict Schedule

Non-feminist Non-mothers

Mself (SD)

3.45 (.84)

3.33 (1.38)

2.09 (1.24)

3.76 (1.07)

Mtyp (SD)

3.05 (1.45)

3.52 (1.51)

3.25 (1.62)

3.71 (1.46)

F(1, 163) = 4.77, p = .03, η2 = .07

F(1, 163) = .56, p = .46, η2 = .009

F(1, 163) = 25.14, p < .001, η2 = .29

F(1, 162) = .06, p = .80, η2 = .001

Non-feminist Mothers

Mself (SD)

4.17 (1.40)

3.69 (1.58)

2.70 (1.78)

3.10 (1.48)

Mtyp (SD)

3.09 (1.75)

3.02 (1.66)

2.89 (1.61)

3.77 (1.63)

F(1, 134) = 34.57, p < .001, η2 = .21

F(1, 134) = 1.55, p = .001, η2 = .08

F(1, 131) = .76, p = .39, η2 = .006

F(1, 133) = 11.80, p = .001, η2 = .08

Feminist Non-mothers

Mself (SD)

3.76 (1.38)

3.56 (1.34)

2.47 (1.29)

3.17 (1.37)

Mtyp (SD)

3.68 (1.23)

3.58 (1.12)

3.19 (.97)

2.99 (.97)

F(1, 70) = .23, p = .64, η2 = .003

F(1, 71) = .03, p = .88, η2 < .001

F(1, 69) = 18.63, p < .001, η2 = .21

F(1, 69) = 1.38, p = .24, η2 = .02

Feminist Mothers

Mself (SD)

4.33 (1.44)

4.09 (1.43)

3.23 (1.64)

2.06 (1.27)

Mtyp (SD)

3.80 (1.34)

3.55 (1.31)

3.06 (1.22)

2.81 (1.17)

F(1, 140) = 16.77, p < .001, η2 = .11

F(1, 138) = 14.32, p < .001, η2 = .09

F(1, 137) = 1.23, p = .27, η2 = .009

F(1, 138) = 3.09, p = .08, η2 = .02

Some means are slightly different than those reported in Table 2 as listwise deletion was not used for these analyses. Mtyp indicates the mean score given by each group for how the “typical feminist” would respond. All η2 reported are partial η2

Differences Between Self and Typical Feminist Ratings for Non-feminist Mothers

Hypotheses 3 and 4 were tested for non-feminist mothers using within subject ANOVAs comparing self ratings with ratings for the typical feminist. Table 3 gives F-test results for each of these analyses. The results of the ANOVAs indicated that non-feminist mothers perceived differences between themselves and the typical feminist on all variables except for co-sleeping, largely supporting hypothesis 3. Non-feminist mothers perceived themselves as placing more importance on extended breastfeeding and carrying children and less importance on setting strict schedules than they perceived was true of the typical feminist. However, feminist women, in fact, placed more importance on extended breastfeeding and carrying children and less importance on setting strict schedules. Thus each of these significant differences represented a misperception, supporting hypothesis 4 for non-feminist mothers on these three variables.

Differences Between Self and Typical Feminist Ratings for Feminist Non-mothers

Within subject ANOVAs comparing ratings for the self with ratings for the typical feminist were conducted in order to test hypothesis 5 among feminist non-mothers. F-test results for each of these analyses are provided in Table 3. Results indicated that feminist non-mothers saw themselves as similar to the typical feminist on all variables except for co-sleeping. For co-sleeping, they perceived that the typical feminist felt it was more important than they did. Thus, hypothesis 5 was largely, although not entirely, supported for feminist non-mothers.

Differences Between Self and Typical Feminist Ratings for Feminist Mothers

Within subject ANOVAs comparing ratings for the self with ratings for the typical feminist were conducted for feminist mothers to test hypothesis 5 for this group. Results of these analyses are presented in Table 3. In contrast to our hypothesis, feminist mothers saw themselves as different from the typical feminist on two of the four measured variables. They perceived themselves as placing more importance on extended breastfeeding and carrying one’s child than they perceived was true of the typical feminist. In support of hypothesis 5, they did not see themselves as significantly different from the typical feminist on co-sleeping or setting a strict schedule for one’s child.

Discussion

In investigating women’s attitudes about attachment parenting practices, our hypothesis about the differences between mothers and non-mothers was confirmed (hypothesis 1). Mothers more strongly supported extended breastfeeding, co-sleeping, and carrying one’s child, and less strongly endorsed setting strict schedules for their children. Of greater importance to the current investigation were the differences in beliefs about attachment parenting practices between self-identified feminist and non-feminist women. Consistent with our hypothesis, feminists endorsed attachment parenting practices, including extended breastfeeding, co-sleeping, carrying one’s child, and not setting strict schedules for their children, more so than did non-feminists (hypothesis 2). This was consistent with recent theorizing that attachment parenting is a feminist way to parent (Etelson 2007) or, at least, a type of parenting that is attractive to feminist women (Bobel 2008; Friedman 2008).

Despite our data, as well as theorizing that attachment parenting is a feminist practice (e.g., Etelson 2007), it remains unclear whether it is empowering or disempowering for feminist women. Erica Jong, a famous second wave feminist, wrote an essay in the Wall Street Journal strongly arguing that attachment parenting is indeed oppressive for women: “attachment parenting…has encouraged female victimization…It's a prison for mothers, and it represents as much of a backlash against women’s freedom as the right-to-life movement.” (Jong 2010, para. 5). Future research should look at the extent to which these different beliefs translate into actual parenting practices. This question would be more appropriate when investigating a sample of solely mothers, rather than a combined group of mothers and non-mothers as was true in the present investigation. If, in fact, feminist mothers engage in higher levels of attachment parenting practices than do non-feminist mothers, as our data would suggest, the question arises whether attachment parenting is an explicitly feminist way to parent or whether it is simply a way of parenting that happens to be used by feminist-identified women.

It may be the case that feminist mothers are understanding attachment parenting in a less essentialist way than is generally assumed given its roots in attachment theory (Franzblau 1999). Feminist mothers may indeed see attachment parenting practices as activities that can be performed by both parents or other members of an extended family or community, as has been suggested by Friedman (2008). However, it should be noted that breastfeeding, a central component of attachment parenting, can only be performed by lactating mothers. Future research may want to consider feminist and non-feminist perceptions about the role that fathers and other caregivers should take in parenting. If attachment parenting practices, excluding breastfeeding, are shared equally within the family, then it would be harder to argue that these practices are disempowering to feminist women, or to women in general. We did not assess whether the women in our sample believed that attachment parenting behaviors should be performed only by mothers or by mothers and fathers, and future research may wish to ask these more specific questions. However, the one study that assessed attachment parenting behaviors among an international group of self-identified attachment parents (for whom feminist identity was not assessed) revealed that the vast majority of attachment parenting was being done by the mother (Green and Groves 2008). In fact, 30 % of the sample reported never leaving their infant in the care of another person, including the father, and an additional 28 % reporting leaving the infant the care of a close relative (including the father) for no more than two hours per week. Whether this high level of maternal involvement would be true of feminist-identified women practicing attachment parenting remains an open question.

While understanding what feminists and non-feminists actually believed about attachment parenting practices was the central goal of the present investigation, a second goal was to understand the stereotypes about how feminists view attachment parenting and the extent to which these stereotypes reflected actual beliefs of feminist women. Research with undergraduates in the United States has suggested that feminists are perceived as career-oriented and working outside of the home and, thus, would be less invested in childcare activities (Berryman-Fink and Verderber 1985; Twenge and Zucker 1999). The notion that feminists are uninterested in the family is a cornerstone of the backlash against feminism (Faludi 1991; Feder 2006).

In partial support of hypothesis 3, non-feminist women did see themselves as different from the typical feminist in regards to some of the assessed beliefs about attachment parenting practices. Specifically, non-feminist non-mothers saw themselves as being more interested in breastfeeding and less interested in co-sleeping than they believed was true of the typical feminist. Non-feminist mothers saw themselves as being more interested in breastfeeding and carrying their children and less interested in setting strict schedules for their children than they perceived was true of the typical feminist. Thus, non-feminist mothers appeared to hold stronger stereotypes about feminist women’s attachment parenting beliefs than did non-feminist non-mothers. These stereotypes are consistent with the image of a feminist woman as being less invested in her children and family, perhaps because she is more invested in aspects of her life outside of the home. However, in support of hypothesis 4, four of these five differences represented misperceptions of the attitudes of actual feminists. Specifically, misperceptions were found for breastfeeding, carrying one’s children and setting strict schedules, attitudes for which feminists more strongly endorsed an attachment parenting perspective than did non-feminists. The only accurate difference was found for non-feminist non-mothers who correctly perceived feminists as being more interested in co-sleeping than they themselves were.

One image of mothers that is consistent with stereotypes of feminist mothers is that of the superwoman: a woman who can manage both her family and the boardroom but may not be physically present for every moment of her child’s life (Johnston and Swanson 2003). Another image is the earth mother: a woman who is very involved with and invested in the lives of her children while also being somewhat anti-establishment and interested in more “natural” practices (Johnston and Swanson 2003). The feminist women in our sample more closely aligned with the image of the “earth mother” than was stereotypically believed to be true of feminists. Thus, conservative writers who argue that feminists are anti-family and not interested in their children are misrepresenting feminist mothers, at least the feminist mothers in our sample. While there is likely a great diversity in what it means to be a feminist mother, the feminist mothers in our sample are endorsing a type of parenting that is extremely involved.

Our hypothesis that feminists would see themselves as similar to the typical feminist in regards to attachment parenting beliefs (hypothesis 5) was partially supported. Feminist non-mothers saw themselves as similar to typical feminists on all variables except co-sleeping where they felt typical feminists would be more interested than they themselves were. Feminist mothers, however, saw themselves as different from the typical feminist in terms of attitudes about extended breastfeeding and frequently carrying one’s child. For these variables, the direction was such that feminist mothers saw the typical feminist as being less likely to endorse attachment parenting techniques than was true of themselves.

Feminist non-mothers generally saw themselves as similar to the typical feminist regarding beliefs about attachment parenting. The exception of co-sleeping may reflect unfamiliarity with this practice among non-mothers (a finding seen with the non-feminist non-mothers as well). Furthermore, to the extent that the typical feminist could be perceived as a feminist mother (a level of specificity our prompt did not include), the perception about feminists being more interested in co-sleeping was accurate. On the other hand, feminist mothers were more likely to see themselves as atypical feminists in terms of attachment parenting beliefs and, to the extent that our prompt about the typical feminist was perceived as referencing a feminist mother, these differences represented misperceptions.

The fact that feminist mothers saw themselves as somewhat atypical in terms of attachment parenting beliefs is intriguing and raises questions about the role of feminism in these women’s sense of identity. Generally, social identity theory would predict that individuals would see themselves as similar to members of their in-group and evaluate this group positively (Tajfel 1981). Feminist mothers, however, saw themselves as somewhat different on this dimension from a social group with which they identified. One explanation may be found in optimal distinctiveness theory: the idea that individuals have simultaneous needs for group identification and individual differentiation (Brewer 1991). Future research should combine quantitative with qualitative methods in order to better understand the role that feminism has in the identity of feminist mothers.

The generalizability of this study is limited by the homogenous nature of our sample which was largely White, well-educated, and consisted of only heterosexual women. Intersecting social identities including those of race, class, and sexual orientation impact the ways in which women understand motherhood (Collins 1994). The experiences of women who parent outside of the traditional hegemonic expectations of parenting, including poor mothers, single mothers, mothers of color, and lesbian or bisexual mothers, should be investigated. It has been suggested that the experiences of these women have been framed through a “deviancy discourse” (Arendell 2000) based on the fact that particular sociodemographic characteristics place them outside of the cultural expectations for ideal motherhood. Future researchers should work to recruit a more demographically diverse sample as attitudes and behaviors surrounding attachment parenting may differ based on the social location in which women parent.

Another limit of this study was the use of internet recruitment. Much of our recruitment was done through contacting women who had active blogs and asking them to post links to our study. Blogs about mothering were particularly targeted in our recruitment, although we also contacted blogs run by non-mothers. “Mommy blogging” is a relatively new cultural phenomenon, and mothers who blog or read blogs may be different than mothers who do not. It has been suggested that blogging is a radical act and that women who blog may be more likely to reject dominant patriarchal views of motherhood (Lopez 2009). Furthermore, future research using internet recruitment should include assessment of the country of origin for respondents. We did not assess country of origin, and, while we recruited primarily from U.S. blogs and our survey was in English, it is possible for people from any country to access any blog.

Finally, it is important to acknowledge that, while our study asked women to give their perceptions of the typical feminist, there is actually no such thing as a typical feminist. The differing perspectives of each individual participant were averaged in order to access a collective stereotype of feminists. We recognize that all feminists are different and that stereotypes about feminists, even when somewhat accurate, fail to acknowledge the diversity of perspectives and life positions within those who self-identify as feminist.

This paper provides information about the stereotypes and realities of what feminist women believe about attachment parenting. We found evidence of pervasive stereotypes that feminists would endorse a less involved parenting style. This belief may be a result of media portrayals of feminism as inconsistent with and detrimental to family life; beliefs that have been part of the conservative backlash against feminism (Faludi 1991; Feder 2006). However, inconsistent with this stereotype, feminist mothers endorsed the importance of the time-intensive, hands-on, parenting techniques associated with attachment parenting. Thus, the results of this study are consistent with those from the small body of literature conducted with U.S. samples (e.g., Rudman and Fairchild 2007; Rudman and Phelan 2007) which suggests that pervasive stereotypes that feminists are in opposition to romantic and family relationships are incorrect. It is important to consider, however, that the parenting techniques examined in this work are contentious, and the value of attachment parenting practices has been a topic of debate within the feminist community (e.g., Etelson 2007; Jong 2010). The extent to which attachment parenting practices may be empowering or disempowering for feminist women remains an open question. Our data indicated that feminist women, are, indeed, interested in these practices, and future research is needed to better understand the implication of holding these beliefs on the lives of feminist women.

Copyright information

© Springer Science+Business Media, LLC 2012