Introduction

It is clear from the popular press, where books entitled Backlash: The Undeclared War against American Women and To Hell with All That: Loving and Loathing Our Inner Housewife (Faludi 1992; Flanagan 2006) compete for the public’s attention that the debate over gendered family roles remains significant, despite a societal shift toward less traditional gender attitudes (Bolzendahl and Myers 2004; Myers and Booth 2002). Attitudes about gender could potentially influence roles, relationships, and family interactions. Early in life, parents and offspring share similar gender attitudes (Kulik 2002a), yet these attitudes may diverge in adulthood once offspring leave the parental home and gain unique life experiences. In particular, attitudes toward marital and childrearing roles may be salient to adult families. Recent historical and demographic changes in family life, such as a growth of dual earner couples (Cabrera et al. 2000; Sweeney 2002; Teachman et al. 2000), may contribute to disparities in parents’ and offspring’s attitudes toward family roles. These differences may create situations where offspring are without role models and parents are unable to offer offspring advice for negotiating adult roles.

The present study expands upon prior research on gender attitudes by examining differences in attitudes toward family roles between generations (i.e. parents vs offspring) and within generations (i.e. mothers vs fathers, daughters vs sons). Further, this study explores whether some parent–offspring dyads differ more in their gender attitudes than other dyads by considering whether generational differences vary by offspring gender or ethnicity. Finally, this study examines variables that may contribute to grown offspring's gender attitudes.

Early in life, parents transmit attitudes about gender to their children through socialization and status inheritance. Prior research indicates that parents socialize their offspring directly by teaching gender role expectations or indirectly through modeling behavior (Hill 2002; Maccoby 2001; Kapinus 2000; McHale et al. 2004). Parents also transmit attitudes indirectly through status inheritance by providing access to social, cultural, and economic resources that create attitude-shaping experiences for their offspring (Glass et al. 1986; Kalmijn 1994; Kohn and Slomczynski 1986).

In adulthood, parents’ influence on their offspring’s gender attitudes may be diminished relative to offspring’s personal characteristics, and offspring’s attitudes may diverge from their parents’ attitudes. Prior studies have established that in adulthood, younger generations hold less traditional gender attitudes than older generations (Brooks and Bolzendahl 2004; Myers and Booth 2002). Indeed, offspring consistently report less traditional attitudes than their parents (Burt and Scott 2002; Moen et al. 1997). Likewise, women endorse less traditional gender attitudes than do men (Cassidy and Warren 1996; Fan and Marini 2000; Shearer et al. 2005). Yet, there remain large gaps in our understanding of gender attitudes. Few studies have examined patterns of differences in gender attitudes within the family during adulthood. When generational differences are examined, research predominantly focuses on the experiences of mothers and daughters in European American families (Moen et al.), leaving fathers, adult sons, and ethnic minority families under-represented.

It is clear that offspring’s experiences outside of the family, such as their educational and relationship experiences, contribute to their gender attitudes. Previous research revealed adult daughters’ education predicted daughters’ gender attitudes, whereas mothers’ education was unrelated to their daughters’ attitudes (Moen et al. 1997). It could be that as adults, offspring’s own educational and relationship experiences result in offspring’s gender attitudes diverging from those of their parents. The extent to which adult offspring’s gender attitudes differ from their mothers’ and fathers’ attitudes, however, may vary according to social or cultural contexts. It may be that in certain parent–offspring dyads, offspring’s gender attitudes remain similar to those of their parents, whereas gender attitudes may diverge more in other dyads. Therefore, it is important to examine not only generational differences in gender attitudes between parents and grown offspring, but also the extent to which these differences vary in certain types of parent–offspring dyads (e.g., parent–daughter vs parent–son dyads).

Specifically, the current study examines attitudes toward marital and childrearing roles because parents and grown offspring remain invested in each other’s family life throughout adulthood (Aquilino 1997; Rossi and Rossi 1990). Attitudes toward marital roles encompass views involving husbands’ and wives’ roles, whereas attitudes toward childrearing roles refer to views concerning the rearing of sons and daughters (Hoffman and Kloska 1995). We refer to these attitudes broadly as gender attitudes, although we recognize these are only one aspect of gender attitudes. The goals of the current study are to examine gender and generational differences in gender attitudes, including the extent to which these differences vary in different types of families. This study also explores whether socialization and status inheritance contribute to grown offspring’s gender attitudes.

Gender and Generational Differences in Gender Attitudes

The first study goal establishes whether gender and generational differences found in previous studies of gender attitudes also describe differences between grown offspring, their mothers, and their fathers. Research suggests both one’s position within the larger society, such as being a woman or a man, and one’s generation matter for individuals’ gender attitudes (Brooks and Bolzendahl 2004; Cassidy and Warren 1996; Moen et al. 1997; Shearer et al. 2005). Prior research has examined generational differences in gender attitudes between mothers and grown daughters (Moen et al.), yet few studies include fathers or grown sons. In general, women and younger generations endorse less traditional attitudes, whereas men and older generations report more traditional gender attitudes (Brooks and Bolzendahl; Moen et al.; Shearer et al.). Differential experiences both within and outside the family contribute to discrepancies in gender attitudes between women and men and between generations.

For example, both fathers and sons benefit from holding more traditional gender attitudes because these attitudes help maintain men’s advantaged position within the family, whereas both mothers and daughters may reject this status quo and instead endorse less traditional gender attitudes (Ferree 1990; Zinn 2000). Indeed, prior studies suggest boys and men express more role-differentiated attitudes and attribute less value to gender equality than girls and women (Burt and Scott 2002; Fan and Marini 2000; Galambos et al. 1990; Jackson and Tein 1998). Given these discrepancies in experiences between women and men, we expected mothers to report less traditional gender attitudes than fathers; daughters to report less traditional gender attitudes than sons (Hypothesis 1).

In addition to anticipated gender differences in mothers’ and fathers’ attitudes and in daughters’ and sons’ attitudes, we also anticipated generational differences in gender attitudes. Offspring likely encounter different attitude-shaping experiences than their parents, and may hold less traditional views of marital and childrearing roles as a result. Over the past several decades, shifting opportunities and expectations for both genders to thrive in the domains of work and family have contributed to a blurring of women’s and men’s family roles (Cabrera et al. 2000; Sweeney 2002). Other experiences, including demographic shifts in the timing of marriage and childrearing, increasing opportunities for offspring of both genders to pursue higher education (Arnett 2000), and the growth of dual earner families (Sparks et al. 2001) may have modified offspring’s beliefs about gender, subsequently shifting their attitudes away from their parents’ attitudes. Therefore, we hypothesized that offspring will report less traditional gender attitudes than parents (Hypothesis 2).

Variability in Generational Differences in Gender Attitudes

Despite a divergence in attitudes between the generations, parents may continue to transmit their gender attitudes to their offspring. Offspring characteristics, such as gender or ethnicity, may facilitate or inhibit parents’ attempts to share their gender attitudes with their adult offspring. Parents tend to identify more with their same-sex offspring (Raley and Bianchi 2006; Starrels 1994). As a result, parents, particularly fathers, often spend more time with their same-sex offspring (Harris and Morgan 1991; Raley and Bianchi) which presents more opportunities for transmitting messages about gender to their same-sex offspring. In adolescence and early adulthood, same-sex parent-offspring relationships intensify, limiting parents’ opportunities to share their attitudes with their opposite-sex offspring (McHale et al. 1999). In adolescence and adulthood, mothers and daughters share similar gender attitudes (Ex and Janssens 1998; Moen et al. 1997). Fathers’ attitudes are more strongly associated with adolescent sons’ than with daughters’ attitudes (Kulik 2002b), although little research has examined generational differences between fathers and their adult sons.

On the other hand, parents’ success at transmitting their gender attitudes to their offspring may also depend on the messages about family roles society conveys. Arguably, historical influences on gender roles brought about the most profound changes in women’s roles within the family (Carr 2004; Moen et al. 1997). Prior research indicates women’s gender attitudes have changed more rapidly than men’s attitudes (Myers and Booth 2002). The lag in men’s attitudes could serve to keep sons’ attitudes more in line with their parents’ attitudes, whereas daughters’ attitudes may diverge further from the attitudes that their mothers and fathers hold. Therefore, this study explores whether generational differences in gender attitudes vary by offspring gender without testing specific hypotheses.

The extent to which parents transmit their gender attitudes to their offspring may also vary by ethnicity. Clearly, societal experiences shape gender attitudes as well as parents’ opportunities to impart their attitudes to their offspring. Research suggests that shared experiences with racial discrimination and parental socialization often result in African American women and men being more sensitive to and less tolerant of gender inequality compared to European Americans (Kane 1992; Orbuch and Eyster 1997). Still, prior work examining ethnic differences in gender attitudes remains equivocal. Some studies indicate African Americans hold less traditional gender attitudes (Kane 1992; McLoyd et al. 2000), whereas others reveal African Americans endorse more traditional attitudes (Hill 2001; Hoffman and Kloska 1995). Discrepancies across studies may represent the specific gender attitudes being examined. African Americans report less traditional attitudes toward maternal employment (Kane; McLoyd et al.), while continuing to accept more traditional attitudes toward masculinity (Hill) and family roles (Hoffman and Kloska). Regardless of whether African American families report more or less traditional gender attitudes, there may be unique circumstances in African American families that facilitate parents’ ability to transmit their gender attitudes to their offspring.

A strong sense of family, frequent interactions with relatives, and an emphasis on parental authority and children’s obedience to family norms characterize African American family patterns (Barber 1994; Jayakody et al. 1993; Parke and Buriel 2006; Smetana and Chuang 2001). African American parents teach their children to be more responsible for household activities than do European American parents, such that from childhood through adulthood African American offspring receive more direct messages from their parents regarding the division of family roles (Hill 2001). Further, shared experiences, such as norms of female employment and exposure to discrimination (Blee and Tickamyer 1995; Hill 2002; Hill and Sprague 1999) may present a context for ongoing gender attitude socialization in African American families more so than in European American families. Early childhood socialization combined with adult experiences may keep African American offspring’s gender attitudes more similar to those of their parents. For these reasons, we expected European American parents and offspring to differ more in their gender attitudes than African American parents and offspring (Hypothesis 3).

It is important to note that the present study focuses on predominantly middle-class African American and European American adults and their parents. Although this select sample does not generalize to families from different socioeconomic backgrounds, prior research supports the importance of considering the familial experiences of middle-class ethnic minority families in an effort to begin to disentangle socioeconomic status and ethnicity (Smetana et al. 2000; Smetana and Chuang 2001). Ethnic differences in this middle-class sample likely represent cultural differences between African American and European American families rather than discrepancies in socioeconomic factors.

Potential Contributors to Offspring’s Gender Attitudes

Finally, this study considers how socialization or status inheritance variables may contribute to grown offspring’s gender attitudes. Based on prior research, we consider parents’ gender attitudes as a marker of socialization and parents’ educational attainments as a proxy for status inheritance (Glass et al. 1986; Moen et al. 1997). We expect socialization and status inheritance to differentially contribute to grown offspring’s gender attitudes. Parents’ social position contributes to offspring’s educational attainments through the transmission of resources (Kalmijn 1994; Kohn and Slomczynski 1986). In general, those with more education hold less traditional gender role attitudes (Brooks and Bolzendahl 2004; Cassidy and Warren 1996; Harris and Firestone 1998). Consequently, offspring’s own life experiences, such as advanced education, may overshadow parents’ educational attainments and contribute to less traditional attitudes toward family roles. Indeed, mothers’ educational attainments appear less important for adult daughters’ gender attitudes than daughters’ own educational experiences (Moen et al. 1997). It is possible, however, that parents may continue to directly socialize their offspring regarding the nature of family roles. Therefore, we hypothesize that offspring’s gender attitudes will be associated with their own educational attainment, rather than with their parents’ educational attainments, and that parents’ gender attitudes will be associated with offspring’s gender attitudes even after accounting for offspring’ characteristics (Hypothesis 4).

In summary, the present study tests the following hypotheses:

  1. Hypothesis 1

    Mothers are expected to report less traditional gender attitudes (i.e. attitudes toward marital and childrearing roles) than fathers; daughters are expected to report less traditional gender attitudes than sons.

  2. Hypothesis 2

    Offspring are expected to report less traditional gender attitudes than their mothers and their fathers, although this difference may vary in families with daughters vs families with sons.

  3. Hypothesis 3

    European American parents and offspring are expected to differ more in their gender attitudes than African American parents and offspring.

  4. Hypothesis 4

    Offspring’s gender attitudes will be associated with their own educational attainment, rather than with their parents’ educational attainments, whereas parent’s gender attitudes will be associated with offspring’s gender attitudes.

Method

Participants

The sample was obtained as part of a larger study of emotional qualities of ties between adults and their parents (N = 213 families; Fingerman et al. 2005). The current study includes the 158 offspring–mother–father triads who completed all phases of the study (N = 474). As discussed elsewhere, this sub-sample did not differ from the larger sample on demographic or relationship characteristics (Fingerman et al. 2006). Offspring ranged in age from 22 to 49 (daughters M = 34.2, SD = 7.4, sons M = 34.9, SD = 7.2). Parents ranged in age from 40 to 84 years (fathers M = 62.9, SD = 9.3, mothers M = 61.3, SD = 8.8). A stratified sampling technique focusing on offspring gender, age, and ethnicity assured that the sample included comparable numbers of daughters (n = 82) and sons (n = 75) distributed by offspring age (younger vs older offspring) and ethnicity.

Participants were recruited from five counties in the greater Philadelphia Metropolitan Statistical area. The majority of participants (85%) were recruited through either the offspring or the parents using purchased lists of phone numbers. The remaining participants were recruited through convenience sampling (e.g. church and community center bulletins, 7%) and snowball sampling (8%). A series of ANOVAs revealed there were no significant differences between participants obtained through each type of recruitment on demographic variables, such as educational attainment ps > .05. We recruited equal proportions of the sample in each stratification cell (offspring gender, age, ethnicity) using each of the three recruitment strategies.

In order to be eligible to participate, a family had to include an offspring age 22 to 49 years who lived within 50 miles of both parents and identify as African American or European American. Parents included whomever the offspring identified as their mother and father. Biological parents made up the majority of parents in the study (97% of mothers and 91% of fathers), although offspring also identified step-parents (1% of mothers and 7% of fathers) or adopted parents (2% of mothers and fathers).

Table 1 presents demographic characteristics, including age, education, marital status, work status, and income. There were no significant gender differences in educational attainment between daughters and sons or between fathers and mothers [Fs(1,151) < 1.00, ps > .05]. The majority of participants were currently married (64% of offspring, 89% of mothers, and 90% of fathers) with 87% of the parents in the study married to one another. There were gender differences in work status χ 2 = 31.94, p < .001, with women more likely to be homemakers/caretakers and less likely to be retired than men. Chi-square tests (n = 474) revealed ethnic differences in marital status χ 2 = 63.86, p < .001 and parent’s work status χ 2 = 19.70, p < .01, with African American participants less likely to be married and less likely to be employed for pay or retired than European American adults.

Table 1 Sample characteristics.
Full size table

Procedure

First, parents and offspring separately participated in phone interviews and each received $10. Next, members of the 158 triads participated in videotaped interviews in either the parent’s or the offspring’s home. Offspring participated in two videotaped interviews, one with their father and one with their mother. Whenever possible, we randomized whether the offspring first participated with their father or mother. The videotaped data were not used in the current study, however, at the conclusion of the interview, family members completed a series of questionnaires, including a measure of gender attitudes. For participating in both videotaped interviews, offspring received $40 and each parent received $20. In addition, either the parent or the offspring received an additional $10 for traveling to the other’s home.

Measures

Gender Attitudes

Participants completed an adapted version of the Attitudes Toward Family Roles Scale (ATFRS; Hoffman and Kloska 1995; Shearer et al. 2005), a 13-item measure assessing gender-based attitudes toward traditional family roles. It consists of two sub-scales: attitudes toward marital roles (e.g. “men should make the really important decisions in the family”) and childrearing roles (e.g. “education is more important for a son than for a daughter”). Participants answered on a 4-point scale ranging from 1 (strongly disagree) to 4 (strongly agree). It has been used previously in samples of both African American and European American adults (Hoffman and Kloska). In this sample, it demonstrated satisfactory reliability for the parents (α = .68 to .85) and offspring (α = .74 to .89). Higher values indicate more traditional gender attitudes (the adapted measure is available by request from the second author).

Demographics

Generation and gender are dichotomous variables. Respondents identified primarily as African American or European American. Two mothers who marked their ethnicity as Hispanic and African American were classified as African American for purpose of analysis in this study. Respondents indicated how many years of education they had completed. Thus, education is a continuous variable in this study.

Results

This study includes both within family and between family effects. In this study, offspring gender and ethnicity are both between family effects because sons and daughters are from different families and families differed by ethnicity. In contrast, parent gender and generation are within family effects because mothers, fathers, and offspring are from the same family. When necessary, we used mixed method analysis of variance to take into account the fact our study includes both within and between family effects.

Hypothesis 1: Within Generation Gender Differences

Hypothesis 1 tests gender differences in mothers’ vs fathers’ and in sons’ vs daughters’ gender attitudes. For parents, we did 2 (Parent gender) × 2 (Offspring gender) × 2 (Ethnicity) mixed method ANOVAs, where parent gender is a within family effect. By contrast, for offspring we did 2 (Offspring gender) × 2 (Ethnicity) between subject ANOVAs. For both parents and offspring, we conducted two sets of ANOVAs, one examining attitudes toward marital roles and one examining attitudes toward childrearing roles.

Parent Gender Differences

Although we predicted that mothers would be less traditional than fathers, mothers and fathers did not significantly differ in their attitudes toward marital roles (Table 2). For childrearing roles, however, there was a main effect of parent gender and as expected, mothers reported less traditional childrearing role attitudes than did fathers.

Table 2 ANOVA results for offspring and parent gender differences in gender attitudes.
Full size table

Offspring Gender Differences

There were main effects of offspring gender for both marital and childrearing role attitudes. As predicted, daughters reported less traditional attitudes than sons (Table 2).

Hypotheses 2 and 3: Generational Differences in Gender Attitudes

The second and third hypotheses examine generational differences in gender attitudes and the extent to which these differences vary by offspring gender (Hypothesis 2) and ethnicity (Hypothesis 3). To test these hypotheses, we conducted a series of 2 (Generation, within factor) × 2 (Offspring gender, between factor) × 2 (Ethnicity, between factor) mixed method ANOVAs. This analysis strategy represents a parsimonious approach to examining variability in gender attitudes, as this strategy makes it possible to simultaneously test the main effect of generation, the interaction between generation and offspring gender, and the interaction between generation and ethnicity (i.e. interactions that include between × within effects) within the same model. The composition of families (i.e. father, mother, and offspring) does not lend itself to a traditional single mixed method ANOVA that simultaneously includes both parents’ gender attitudes. Therefore, we conducted mixed method ANOVAs separately for mothers and fathers.

A significant main effect of generation indicates parents and offspring differ in their gender attitudes. Significant generation × offspring gender or generation × ethnicity interactions indicate that generational differences vary by offspring gender and/or ethnicity. In the event of a significant interaction, we conducted follow-up mixed method ANOVAs to determine the location of the mean difference.

Fathers and Offspring

A main effect of generation on marital role attitudes indicated that offspring were less traditional than their fathers were (See Table 3). Results for childrearing role attitudes revealed a main effect for generation and an interaction between generation and ethnicity. Follow-up analyses separate by ethnicity indicated that both European American offspring, F(1,103) = 53.08, p < .001 and African American offspring, F(1,47) = 7.20, p < .01 were less traditional than their fathers, although European American fathers and offspring differed more in their attitudes (2.6 points) than did African American fathers and offspring (1.4 points).

Table 3 Mixed method ANOVA for generational differences in gender attitudes.
Full size table

Mothers and Offspring

There was a main effect of generation and a generation × ethnicity interaction for mothers’ marital role attitudes (Table 3). Follow-up analyses separate by ethnicity indicated that European American offspring had less traditional marital role attitudes than their mothers, F(1,105) = 5.04, p < .05, whereas African American mothers and offspring did not differ, F(1,48) < 1.00, p > .05. For childrearing roles, there was a main effect of generation, a generation × gender interaction, and a generation × ethnicity interaction. Follow-up analyses separate by offspring gender revealed that daughters had less traditional childrearing roles than mothers, F(1,80) = 13.70, p < .05, whereas mothers and sons did not differ, F(1,73) < 1.00, p > .05. As expected follow-up analyses separate by ethnicity indicated that European American offspring were less traditional about childrearing than their mothers, F(1,105) = 40.69, p < .001, whereas African American mothers and offspring did not differ, F(1,48) < 1.00, p > .05.

In summary, results provided partial support for Hypothesis 1; mothers and daughters reported less traditional childrearing role attitudes than fathers and sons. Daughters also reported less traditional marital role attitudes than sons did, but mothers and fathers did not differ in their marital role attitudes. Results also partially supported Hypothesis 2; parents were more traditional than their offspring, although mothers and daughters differed more in their childrearing role attitudes than did mothers and sons. Consistent with Hypothesis 3, European American parents and offspring differed more in their attitudes toward childrearing roles than African American parents and offspring. European American mothers and offspring also differed more in their marital role attitudes than African American mothers and offspring.

Hypothesis 4: Potential Contributors to Offspring’s Gender Attitudes

Finally, we tested whether parental socialization or status inheritance variables contribute to grown offspring's gender attitudes. Fathers’ and mothers’ years of education were highly correlated (r = .57). For this reason, the average of parents’ education was included in the analyses to represent status inheritance. Parents’ gender attitudes were included in the analyses to represent socialization.

Next, we used hierarchical regression to predict offspring’s gender attitudes because we wanted to first examine the effects of offspring’s educational attainment on their gender attitudes, and then consider whether socialization or status inheritance variables were associated with offspring’s gender attitudes after controlling for offspring’s characteristics. We conducted two regressions, one where the dependent variable was offspring’s marital role attitudes and one where the dependent variable was offspring’s childrearing role attitudes. Table 4 presents the three steps of the regression examining offspring’s marital role attitudes, and Table 5 presents the three steps of the regression for offspring’s childrearing role attitudes. Offspring gender, ethnicity, and offspring’s years of education were included in the first step in both analyses. Offspring gender and ethnicity were dichotomous variables and were included in the models because the previous set of analyses revealed significant gender and ethnic differences in gender attitudes. For both analyses, each parent’s attitudes were included in the second step and the average of parents’ years of education was included in the third step of the regression. We entered each parent’s gender attitudes into the models before parents’ average education based on our expectation that parents’ gender attitudes would be more strongly associated with offspring’s attitudes than parents’ educational attainment (Hypothesis 4).

Table 4 Summary of regression analysis for variables predicting offspring's marital role attitudes.
Full size table
Table 5 Summary of regression analysis for variables predicting offspring’s childrearing role attitudes.
Full size table

Offspring’s Attitudes Toward Marital Roles

Contrary to our expectations, offspring education was not a significant predictor in the first step of the regression to predict offspring’s attitudes toward marital roles (Table 4). In the second step, fathers’ attitudes emerged as a significant predictor. The more traditional fathers’ attitudes were toward marital roles, the more traditional offspring’s attitudes were. As anticipated, the addition of parents’ average education in step 3 did not add significantly to the explained variance. The final model explained 16% of the variance in offspring’s attitudes toward marital roles.

Offspring’s Attitudes Toward Childrearing Roles

In the first step of the regression to predict offspring’s attitudes toward childrearing roles, offspring’s education was significant (see Table 5). As expected, more educated offspring reported less traditional attitudes toward childrearing roles. The addition of parents’ attitudes (step 2) and parents’ average education (step 3) did not add significantly to the explained variance. The final model explained 35% of the variance in offspring’s attitudes toward childrearing roles.

Overall, results partially supported Hypothesis 4. Fathers’ attitudes toward marital roles were significantly associated with offspring’s marital role attitudes after controlling for offspring’s characteristics. In contrast, parents’ attitudes toward childrearing roles did not significantly contribute to offspring’s attitudes toward childrearing roles once offspring’s educational attainment was considered.

Discussion

The present study expanded upon prior work on generational differences in gender attitudes in several important ways: (a) by including fathers and sons as well as mothers and daughters, (b) by including both African American and European American families, and (c) by examining attitudes toward marital and childrearing roles relevant to adult families. Findings from this study reveal that generational differences vary by offspring gender (son vs daughter) and ethnicity. In addition, results suggest fathers’ attitudes toward marital roles are associated with offspring’s marital role attitudes, whereas offspring’s own life experiences appear to be more important for offspring’s childrearing role attitudes.

Gender and Generational Differences in Gender Attitudes

Prior research has primarily examined gender differences in gender attitudes outside the context of family relationships (Cassidy and Warren 1996; Shearer et al. 2005). The unique design of this study allows for comparisons between men and women within the same family making it possible to consider which family members report the most and the least traditional attitudes toward family roles. Consistent with prior research, mothers and daughters reported less traditional attitudes toward childrearing roles than did fathers and sons (Cassidy and Warren; Shearer et al.). Daughters also reported less traditional marital role attitudes than sons, although mothers and fathers did not differ in these attitudes. Men in both generations may report more traditional attitudes because these attitudes perpetuate men’s advantaged position within the family. In contrast, women in both generations may report less traditional attitudes as a means of opposing these traditional values (Ferree 1990; Zinn 2000). Still, findings suggest that daughters are not only more motivated to reject norms of male authority than fathers and sons, but they do so more than mothers. It may be that daughters, more than mothers, have greater expectations for gender equality in the home that translate into daughters endorsing less traditional gender attitudes compared to all other family members. As anticipated, offspring reported less traditional attitudes than did their parents.

Variability in Generational Differences in Gender Attitudes

Consistent with our expectations, generational differences in gender attitudes varied by offspring gender and ethnicity. Findings indicated that mothers and daughters differed more than mothers and sons. This generational difference may be attributed to changes in women’s opportunities and social roles between the generations, although the cross-sectional nature of this study makes it impossible to examine these changes. Prior research examining changes in gender attitudes over the past several decades suggests that women’s gender attitudes have changed at a more rapid rate than men’s attitudes (Myers and Booth 2002). This rapid rate of change may help explain why daughters report less traditional attitudes than their mothers. In contrast, the lag in men’s attitudes may contribute to sons’ attitudes being more similar to their mothers’ attitudes (Myers and Booth). Although it is beyond the scope of the present study, daughters’ greater divergence from their parents’ attitudes may have implications for parent–offspring relationships. For example, less traditional daughters may find it difficult to receive childrearing advice from their more traditional parents because of these differences in opinion regarding how children should be socialized.

Consistent with hypothesized ethnic differences, European American parents and offspring differed more in their gender attitudes than African American dyads. The greater similarity in gender attitudes in African American families may reflect socialization processes operating from childhood into adulthood. Potentially, strong family ties, frequent contact, and an emphasis on obedience to family norms may present opportunities for ongoing gender attitude socialization in African American families (Barber 1994; Parke and Buriel 2006; Smetana and Chuang 2001).

It is important to acknowledge, however, that the effect sizes for these differences were relatively small. Gender and generational status are two of many things that influence gender attitudes. There are clearly a number of other factors, including daily experiences through an individual’s occupation or the division of household labor between partners that may have a more proximal impact on gender attitudes (Apparala et al. 2003; Cunningham 2005; Dodson and Borders 2006; Luhaorg and Zivian 1995). Still, the size of our mean differences is quite similar to the gender and ethnic differences observed in other studies using the Attitudes Toward Family Roles Scale (ATFRS; Hoffman and Kloska 1995; Shearer et al. 2005).

Potential Contributors to Offspring’s Gender Attitudes

Findings provided partial support for the role of parental socialization. Fathers’ and offspring’s marital role attitudes were positively associated. This association may reflect indirect socialization processes (Kapinus 2000), where fathers’ attitudes may have been the dominant, pervasive attitudes displayed in the family of origin, whereas offspring may have been less aware of their mothers’ attitudes. Due to the cross-sectional nature of this study, it is not possible to determine the directions of these effects. It could also be that offspring’s attitudes contribute to fathers’ attitudes. For example, a traditional father’s attitudes could become less traditional after he witnesses his daughter establish herself in a successful career. Future longitudinal studies should explore the extent to which adult offspring socialize their parents.

Consistent with prior research, offspring’s own educational attainments were associated with their attitudes toward childrearing roles, but parents' educational attainment was not (Moen et al. 1997). Not surprisingly, more educated offspring reported less traditional attitudes toward childrearing roles (Brooks and Bolzendahl 2004; Harris and Firestone 1998). This finding suggests that childrearing attitudes are more susceptible to role-shaping experiences, such as education, whereas marital role attitudes, such as endorsing male authority, may be less susceptible to these influences for offspring in today’s adult families. The increase in dual earner households may have contributed to individuals choosing to reject norms of male authority regardless of level of education. In contrast, less traditional childrearing attitudes, such as giving a boy a doll as a toy may be endorsed more by those with more education because they have been exposed to experiences that emphasize the importance of treating boys and girls similarly. Alternatively, offspring with more traditional attitudes may not seek more education. Together, results suggest that parents’ educational attainment may determine offspring’s opportunities for advanced education, but offspring’s own educational experiences may shape their attitudes toward childrearing roles.

It is also necessary to acknowledge that parents' and offspring's gender attitudes could covary for other reasons, including genetic covariation of broad personality traits. Indeed, prior research indicates there are significant genetic influences on aspects of personality, including sex-typed behaviors (Cleveland et al. 2001; Iervolino et al. 2005; Johnson et al. 2005; Losoya et al. 1999). Future studies should consider interactions between genes and family environment to better understand the mechanisms underlying gender attitude similarity in adult families.

Study Limitations and Directions for Future Research

This study contributes to our understanding of generational differences in gender attitudes, yet is not without limitations. First, the cross-sectional nature of this study makes it impossible to describe developmental changes or causal mechanisms. Further, the sample may not represent the experiences of parents and adult offspring in general. For example, these findings represent the experiences of families living in close proximity to one another, and it is unclear whether similar results would be found with parents and offspring living at a distance from one another. Also, this sample includes only African American and European American families. Families from other ethnic backgrounds may vary in both the attitudes they hold toward family roles as well as in the extent to which the generations differ in their gender attitudes. For example, Latino American families often endorse traditional gender attitudes and emphasize obedience and respect for parental authority (Fuligni 1998). It is also necessary to recognize that variations by offspring gender reflect differences between daughters and sons from different families. It is less clear how parents’ attitudes might differ from the attitudes held by daughters and sons within the same family.

Further, African American parents and offspring in this study represent a select sample that may not reflect the experiences of other African American families. Previous research emphasizes the intersection between gender, race, and class in describing gender and ethnic variations in attitudes toward family roles (Hill 2001; Hill and Sprague 1999; Kane 2000). In our sample, the majority of African American and European American participants could be described as middle-class, making it difficult to examine further distinctions in attitudes by class. Still, our results suggest there may be unique circumstances in African American families, aside from socioeconomic status, that contribute to parents and offspring holding more similar gender attitudes into adulthood. It could be that shared experiences with racial inequality combined with frequent family contact in African American families serve to keep grown offspring’s gender attitudes similar to those of the parents, even in a select sample of African American families (Blee and Tickamyer 1995; Hill 2002; Hill and Sprague; McLoyd et al. 2000). In the future, researchers should examine ethnic variability in gender attitudes in a more socioeconomically diverse sample.

This study also explores only one attitude-shaping experience, educational attainment, which may be associated with the gender attitudes of adult offspring and their parents. Divorce contributes to less traditional gender attitudes by forcing individuals into nontraditional family roles (Bolzendahl and Myers 2004). Future longitudinal studies should consider whether changes in offspring's marital status affect not only adult offspring’s gender attitudes, but also their parents’ attitudes (Miller and Glass 1989). Further, due to the large percentage of married participants in our sample, we also acknowledge that the gender attitudes of our sample may be more conservative on the whole than we might expect across a more diverse sample of family compositions.

Conclusions

This study expands upon research on gender attitudes between parents and offspring by including the experiences of fathers and sons, family members often under-represented in intergenerational research. Results reveal that although parents report more traditional gender attitudes than their offspring, there are greater generational differences between mothers and daughters and in European American families. In a changing social context where gender roles are becoming integrated and less distinct, generational discrepancies in gender attitudes may make it more difficult for parents to understand offspring’s family experiences and for offspring to look to their parents as role models for negotiating family roles. Further, findings suggest fathers’ marital role attitudes may contribute to offspring’s attitudes, whereas offspring’s own educational attainment seems more important for their attitudes toward childrearing roles.