3. Gender role attitudes within couples, and parents' time in paid work, child care and housework
Jennifer Baxter, Australian Institute of Family Studies
The allocation of Australian parents' time to paid and unpaid work remains very gendered, with fathers usually in full-time paid employment, and mothers often employed part-time or not in employment (Baxter, 2013). Mothers also spend more time than fathers doing household work, whether that is child care or other domestic work (Craig & Mullan, 2011). Even when mothers work full-time, when there are young children in the family, mothers tend to do more of the child care and other domestic work than fathers, and gender differences such as these are apparent across many developed countries (e.g., Coltrane, 2000; Craig & Mullan, 2011; Davis & Greenstein, 2009; Hook, 2006; Sayer, 2005; Shelton & John, 1996).
These different time-use patterns are likely to be linked with gender role attitudes towards work and family and towards the distribution of household work. The focus of this chapter is on exploring gender role attitudes among Australian parents. A significant contribution of this research is being able to undertake couple-level analyses of gender role attitudes for a large sample of parents, and also being able to explore associations with each parent's time use and assessments of fairness in the ways they share child care and household work. By exploring these associations, this research provides insights on the degree to which the gendered patterns of parental time use in Australia might be related to gendered perceptions of parents' roles within the family.
The evidence to date is that although there has been considerable change in terms of attitudes toward women being employed, there is still a diversity of views concerning the division of household work (Coltrane, 2000). For example, in a survey of Australian households in 2005, a significant proportion of Australian men and women agreed that "it is better for the family if the husband is the principal breadwinner outside the home and the wife has primary responsibility for the home and children" (41% of men and 36% of women agreed with this) (Van Egmond, Baxter, Buchler, & Western, 2010). However, a majority of the men and women agreed that household work should be shared equally when both parents work. Such findings have also been observed in other studies (e.g., Coltrane, 2000).
Research from Australia and elsewhere shows that gender role attitudes vary with a number of personal and family characteristics. For example, men tend to have more conservative attitudes than women, and variation has been observed according to characteristics such as educational attainment, ethnicity, religion, employment, life stage and family composition (see Coltrane, 2000; Davis & Greenstein, 2009; Van Egmond et al., 2010).
The findings noted above that most men and women think household work should be shared equally if both parents work is somewhat at odds with the way in which household work is actually divided in many Australian families. To some extent this may be a reflection of the high rate of part-time work by mothers in Australia. This part-time work can be a compromise for those who have more traditional gender role attitudes but also wish or need to be employed. Previous research has highlighted that individuals' gender role attitudes and behaviours may not be consistent (Schober & Scott, 2012). In the case of maternal employment, this may be a reflection of mothers' constrained employment choices, such that mothers may not always be able to engage in the labour force the way in which they would like (Himmelweit & Sigala, 2004; McRae, 2003). Of course, there are many likely influences on women and on families when it comes to negotiating employment and care of children, and gender role attitudes are only part of this picture.
Despite the inequities in patterns of time use, mothers in Australia and elsewhere often report that the distribution of household tasks is fair (Baxter & Western, 1997; Blair, 1998; Thompson, 1991; Wilkie, Ferree, & Ratcliff, 1998). This may reflect that the unequal distributions of time spent on paid and unpaid work may be part of the negotiated way in which parents manage their paid and unpaid work tasks. Mothers and fathers appear to equate a "fair" division of child care or other domestic work to being when around one-third of the total parental time on these activities is done by fathers (Baxter & Smart, 2010). Assessments of fairness, however, may also vary when men and women have non-traditional rather than traditional attitudes. Some of the research in this area is discussed later in this chapter.
Decisions about how the paid and unpaid work within the family is distributed may well be a result of mothers' as well as fathers' attitudes, and so it is useful to consider this broader view rather than that of one parent alone. Previous research, for example, has highlighted the importance of fathers' as well as mothers' gender role attitudes in explaining how household work is shared (Coltrane, 2000; Davis & Greenstein, 2009). Some studies have taken this further by exploring whether there is some interaction between mothers' and fathers' gender role attitudes in predicting the allocation of parents' time to household activities (Bulanda, 2004; Greenstein, 1996b), as might be suggested by the idea of "maternal gatekeeping" by more traditional mothers, such that fathers are deterred from being involved, regardless of their own desire to be involved (Allen & Hawkins, 1999).
Analysis of gender role attitudes using LSAC allows us to explore views of mothers and fathers in a sample of parents with school-aged children, and to relate these data to parental and family characteristics, and to parents' time use. Specifically, these data allow detailed analyses of gender role attitudes, as expressed in views about the male breadwinner model and about equal sharing of child care and household work. To what extent these attitudes vary for mothers and fathers, and also according to family and personal characteristics is explored in this chapter. A particular focus is given to the relationships between attitudes and parental employment patterns. The extent to which there are more traditional gender role attitudes among parents when mothers are not employed might provide more insights into the factors leading to the non-employment of these women. Further, the amounts of time parents spend on child care and other domestic work, as well as their perceptions of fairness, are considered in relation to parents' gender role attitudes to explore whether the concept of fairness differs in families of more traditional, versus non-traditional, parents. To summarise, this chapter will seek to answer the following questions:
- What are mothers' and fathers' gender role attitudes, and how do these align with parental employment patterns?
- What key demographic factors emerge in describing which mothers and fathers have more traditional gender role attitudes (e.g., educational attainment, language spoken at home, family composition)?
- How closely are gender role attitudes matched within couples?
- To what extent are different patterns of paid and unpaid (child care and domestic) work related to different gender role attitudes of each parent?
- How are parents' perceptions of the fairness in how unpaid work is shared related to their gender role attitudes?
These questions will be explored within each of the subsections of the chapter, following a description of the data used throughout.
3.2 Data and methods
Measures of mothers' and fathers' gender role attitudes
This chapter uses Wave 5 data for the K cohort of LSAC, in which the LSAC study children were aged 12-13 years. This was the first main wave in which gender role attitudes were collected. The items were also asked of B cohort parents, and similar results are achieved if the analyses are repeated for that sample. The B cohort results have not been presented here.
The gender role items available in LSAC and analysed in this chapter are:
- "It is better for the family if the husband is the principal breadwinner outside the home and the wife has primary responsibility for the home and children"; and
- "If both husband and wife work, they should share equally in the housework and child care".
Both these items relate to the way in which paid and unpaid work should be distributed within the family. Items the same as or very much like these have been used to assess gender role attitudes in a number of studies (see Davis & Greenstein, 2009). For Australia, see in particular Van Egmond and colleagues (2010) who used data from five surveys, including the Household, Income and Labour Dynamics in Australia (HILDA) survey, to examine trends in men's and women's gender role attitudes.
Throughout this chapter the first of these statements is referred to as measuring support for the male breadwinner model. The second statement is referred to as measuring support for equal sharing when both parents work. Note that the wording of the equal sharing statement does not take account of the likely differences in paid work hours of mothers and fathers in Australia, with the majority of employed mothers working part-time hours. A presumption of this question seems to be that when both husband and wife work, they would be working the same hours and so should share equally in housework or child care. Whether this affects how parents respond to this item remains an open question.
Parents' views on these items were collected on a Likert scale in which parents indicated whether they "strongly disagree", "disagree", "neither agree nor disagree", "agree" or "strongly agree" with each statement. For most analyses in this chapter, the two categories representing some disagreement are combined, as are the two categories representing some agreement. Those who strongly agreed or agreed that the male breadwinner model is better are referred to here as being "traditional", those who were neutral (neither agreed nor disagreed) are referred to as being ambivalent or undecided, and those who strongly disagreed or disagreed are referred to as being "non-traditional".
While having only two items for analysis is somewhat limiting, the rich family information on paid and unpaid work in LSAC provides extensive opportunities for exploring how these two gender role items vary with different family circumstances.1
In Wave 5 of the K cohort, 3,956 families were interviewed, from which responses to the gender role items were available for 3,679 mothers and 2,263 fathers. The smaller number of responses for fathers reflects both the incidence of single-mother households (of the 3,679 mother responses, 580 were from single-mother households) and the non-response by fathers (within 3,256 couple families, 3,099 mothers and 2,199 fathers provided responses to these items).2 Beyond this introduction to the data, the analyses are presented only for couple families, so that gender role attitudes can be explored in the context of the sharing of paid and unpaid work in the one household. Only partnered parents where both parents provided responses are included, to maintain consistency throughout the analyses and to allow within-couple comparisons of mothers and fathers. These respondents are considered to be in-scope mothers and fathers. There is some bias in this sample; in particular, the final in-scope sample excludes the vast majority of those families with not-employed fathers, and also under-represents families with not-employed mothers.3 This is because of higher non-response to the self-completion questionnaire by fathers in these families. Because the sample is not representative of families in which fathers are not employed, these analyses are not able to explore how non-traditional patterns of parental employment (in particular, female breadwinner families) are related to different gender role attitudes of parents.
Gender role attitudes by family type and parental response
Responses on the male breadwinner item are shown in Table 3.1 for single mothers, partnered mothers with no responding partner, and the in-scope sample - partnered mothers with a responding partner. Differences in the responses on the breadwinner item by mothers who had and did not have a responding partner were not statistically significant, suggesting that on this item, there was no particular bias in using the sample of mothers with responding partners. These results show that single mothers were considerably more likely than partnered mothers to strongly disagree that a male breadwinner model is better. While disagreement with this item is typically interpreted as indicating a less traditional gender role attitude, it may simply indicate that they do not agree that having a breadwinner husband would make their situation better. Results for partnered fathers with a responding partner are also shown.4 The responses of mothers and fathers in the in-scope sample had very similar distributions on this item. Previous research on gender role attitudes has typically found more conservative or traditional attitudes for men than for women (e.g., Van Egmond et al., 2010). The lack of difference between men and women here may reflect that this sample includes only partnered parents, whose views may not be representative of views in the wider population.
|Partnered mother, no partner response (%)||Partnered mother, with partner response (%)||Partnered father, with partner response (%)|
|Notes: Chi-square tests were used to compare distributions of single compared to partnered mothers (χ2 (4, n = 3610) = 61.07, p < .001); then, within partnered mothers, those mothers with and without partner responses (χ2 (4, n = 3030) = 10.52, p > .05). Chi-square tests could not be used to compare mothers' and fathers' responses since these responses are not independent.|
|Neither agree nor disagree||33.2||38.4||33.8||37.0|
|No. of observations||580||914||2,116||2,116|
The results for the equal sharing item are shown in Table 3.2. Overall, there was very strong agreement with this statement by mothers and fathers, which is consistent with views of the wider population reported elsewhere (Van Egmond et al., 2010). For example, in reviewing literature related to household labour, Coltrane (2000) noted that the vast majority of men and women agree that family labour should be shared, despite the very gendered patterns of family labour that are evident. There were differences between single and partnered mothers on this item, with single mothers more likely than partnered mothers to strongly agree that there should be equal sharing of housework and child care. There was a small difference in the distribution of responses to this item if comparing partnered women with and without a partner response, with the in-scope sample in somewhat more agreement about equal sharing. Also, in-scope mothers and fathers differed somewhat on their responses to this item, with greater agreement (strongly agree) about equal sharing by mothers rather than fathers. For all mothers and fathers, however, very few disagreed or strongly disagreed that housework and child care should be equally shared when both parents work.
|Partnered mother, no partner response (%)||Partnered mother, with partner response (%)||Partnered father, with partner response (%)|
|Notes: Chi-square tests were used to compare distributions of single compared to partnered (χ2 (4, n = 3609) = 22.19, p < .01), and then within partnered mothers, those with and those without partner responses (χ2 (4, n = 3029) = 16.59, p < .05). Chi-square tests could not be used to compare mothers' and fathers' responses since these responses are not independent. Percentages may not total exactly 100.0% due to rounding.|
|Neither agree nor disagree||13.8||16.9||13.5||19.2|
|No. of observations||580||914||2,115||2,115|
Descriptive statistics are used to analyse the gender role attitudes in relation to mothers' and fathers' work hours in section 3.4, and other demographic characteristics in section 3.5. These demographic characteristics include parents' educational attainment, religion, speaking a language other than English at home, age, marital status, number of children aged under 15, and age of youngest child. These data are described in these later sections. In additional analyses that have been referred to but not presented, the associations between variables were examined while controlling for other demographic characteristics.5
The gender role attitudes of mothers and fathers are compared within couples in section 3.6. Following this, parents' gender role attitudes are related to the distribution of time in paid and unpaid work in section 3.7, and to parents' perceptions of the fairness of sharing household work and child care in section 3.8. These analyses of time use are descriptive only, and focus on the male breadwinner item.
In all analyses presented in this chapter, only one wave of LSAC is used, and so it is not possible to make any claims on whether gender role attitudes determine different patterns of parental time spent in paid and unpaid work. Equally, such attitudes may be a reflection of parents' allocation of time to, and sharing of, paid and unpaid work in the family (Crompton & Lyonette, 2005; Himmelweit & Sigala, 2004; McRae, 2003).
3.3 Mothers' and fathers' gender role attitudes
Mothers' and fathers' responses on the two gender role items were described in section 3.2. Overall, these data indicate that there is some diversity of views regarding the male breadwinner model, with 38% of in-scope mothers and 36% of in-scope fathers disagreeing (including strongly disagreeing) that a male breadwinner model is better, 34% of mothers and 37% of fathers being uncommitted, and 28% of mothers and 27% of fathers believing that a male breadwinner model is better.
Despite the diversity of views about the male breadwinner model, there was much more agreement on the equal sharing of unpaid work when both parents work:
- 83% of mothers and 77% of fathers agreed (or strongly agreed) that the unpaid work should be equally shared;
- 14% of mothers and 19% of fathers were uncommitted; and
- only 4% of mothers and fathers did not agree that work should be equally shared when both parents work.
These findings are consistent with the findings of broader population studies in Australia and elsewhere (Coltrane, 2000; Van Egmond et al., 2010), as discussed in the introduction.
Regardless of views about the male breadwinner model, the majority of parents believe that unpaid work should be equally shared when both parents work (see Table 3.3). Agreement with this was lowest among those who expressed no strong views regarding the male breadwinner model (that is, those who selected "neither agree nor disagree") as a relatively high percentage of these parents likewise expressed no strong views regarding the sharing of unpaid work.
|Agreement with male breadwinner model||Agreement with equal sharing when both parents work (%)||No. of observations|
|No. of observations||1,755||277||81||2,113|
|No. of observations||1,618||406||89||2,113|
These results indicate that having a view that the male breadwinner model is better is not typically associated with a view that the allocation of child care and household work at home should be gendered if both parents work. This suggests that these two measures, while used throughout this chapter to assess gender role attitudes, might be capturing two quite different concepts. We return to discuss this more fully in the final section of the chapter.
3.4 Employment patterns and gender role attitudes
The extent to which parents' gender role attitudes are matched to their actual levels of participation in paid work is examined in this section. The tables in this section show the associations first for the male breadwinner item and then for the equal sharing item. The discussion of results also references additional analyses of these associations in which other characteristics of parents and families were controlled.6
Overall, we expected to find significant associations between mothers' behaviour (maternal employment) and gender role attitudes, with less traditional views among mothers who spend longer hours in paid work, and more traditional views among mothers who are not employed. Such associations might reflect that gender role attitudes of mothers drive them to choose particular employment patterns, or that parents change their attitudes to be consistent with their behaviour (as predicted by Festinger's  cognitive dissonance theory). In fact, it appears that both arguments apply, with cross-sectional and longitudinal research finding evidence of reciprocal effects between parents' attitudes and behaviours concerning maternal employment (Himmelweit & Sigala, 2004; Kalmijn, 2005; Schober & Scott, 2012). Further, partners' gender role attitudes may be part of this picture, in that mothers' employment participation and/or attitudes may change in response to fathers' attitudes (Kalmijn, 2005; Schober & Scott, 2012).
First, looking separately at mothers' and fathers' views on the male breadwinner model, Table 3.4 shows how views vary with differences in mothers' and fathers' usual weekly work hours. Mothers' usual hours in employment were classified as full-time (35 hours or more per week), part-time (1-34 hours per week) or not employed. Overall, 31% of mothers were employed full-time, 53% part-time and 17% not employed.
|Mothers (%)||Fathers (%)|
|Agree (traditional)||Neutral||Disagree (non-traditional)||Total||Agree (traditional)||Neutral||Disagree (non-traditional)||Total|
|Notes: Chi-square tests used to compare distribution of responses to the male breadwinner item: mothers' views by mothers' employment: χ2 (4, n = 2,116) = 344.22, p < .001; mothers' views by fathers' employment: χ2 (2, n = 2,116) = 11.7, p > .05 (not significant [n. s.]); fathers' views by mothers' employment: χ2 (4, n = 2,116) = 270.65, p < .001; fathers' views by fathers' employment: χ2 (2, n = 2,116) = 13.45, p > .05 (n. s.). Percentages may not total exactly 100.0% due to rounding.|
|Not employed or part-time employed||33.6||24.9||41.6||100.0||23.5||30.6||46.0||100.0|
|All mothers and fathers||28.1||33.8||38.1||100.0||26.9||37.0||36.1||100.0|
|No. of observations||567||710||839||2,116||544||779||793||2,116|
The gender role attitudes of mothers and fathers vary significantly with mothers' work hours, with greater support of the male breadwinner model when mothers were not employed. There were, however, significant numbers of parents for whom the reality did not appear to reflect their general views of what is better for the family. In particular:
- among couple families in which mothers worked full-time hours, around 16% of mothers and fathers agreed that the male breadwinner model is better; and
- among couple families in which mothers were not employed, one fifth of mothers and fathers (18%) did not agree that the male breadwinner model is better.
We return to this later in this section to explore the alignment between maternal employment and support of the male breadwinner model.
While parents' gender role attitudes may also matter in explaining the employment participation of fathers, this sample has few families with non-traditional employment arrangements; in particular, there were almost no families in which the father was not employed. This is likely to limit the potential for detecting the importance of gender role attitudes in explaining different patterns of paternal employment. To analyse these data by fathers' work hours, usual work hours were classified as full-time hours (35 hours or more per week, 93% of fathers), or part-time or not employed (including being on leave) (7% of fathers). There was somewhat less support for the male breadwinner model by fathers who worked part-time hours or were not employed, although this difference did not reach statistical significance. Mothers' support for the male breadwinner model did not vary significantly by this classification of fathers' hours.
When the associations between parental work hours and responses on the male breadwinner item were examined after controlling for other personal and family characteristics, the associations with mothers' work hours remained statistically significant. As in the analyses presented in Table 3.4, however, fathers' employment was not significant in explaining differences in mothers' views about the male breadwinner model when other variables were controlled. In these additional analyses, it did emerge that fathers who were part-time employed or not employed were significantly less likely to support a male breadwinner model compared to fathers who were full-time employed.
The links between gender role attitudes and time spent in paid as well as unpaid work are explored further in section 3.7.
We return to examine these associations for the equal sharing item after some further analyses related to views about the male breadwinner model. Table 3.4 shows that mothers' views on this item did not always align with their actual employment patterns. That is, some mothers who did not support a male breadwinner model were not employed, whereas some mothers who did support a male breadwinner model were employed. It is not especially surprising that these attitudes and employment patterns do not align perfectly. In fact, this dissonance has been observed as likely to reflect that choices about employment are to some extent constrained (Crompton & Lyonette, 2005; Himmelweit & Sigala, 2004; McRae, 2003). For example, some mothers may prefer to be working, but are unable to find a suitable job. Others may be persuaded to remain out of employment by a competing concern over the value of caring for children. Some mothers, on the other hand, may prefer that they were caring for children instead of working, but are motivated to be employed by financial concerns, by other personal rewards that they gain from employment, or by their partner's expectations.
Of course it is also important to reflect on the nature of the item to which parents were responding. Agreement that "it is better if the husband is the principal breadwinner and the wife has responsibility for the home and children" does not necessarily mean parents believe the wife should not be employed at all. Agreement indicates that earning is primarily the husband's job and looking after the home primarily the wife's job, but there remains the possibility that those who strongly agree with this might also strongly believe that there is value in the wife also spending some time in paid work.
For the not-employed mothers who did not support a male breadwinner model, it is of interest to consider why these mothers were not employed, and whether or not decisions about employment were likely to be motivated by their gender role attitudes. Focusing on not-employed mothers, Table 3.5 first presents mothers' reasons for not being in paid work, according to their male breadwinner model views. This shows:
- Mothers who supported a male breadwinner model were considerably more likely to say that they were not working because of family-related reasons (83%), when compared to those who did not support a male breadwinner model (51%).
- A relatively high proportion of those who did not prefer the male breadwinner model said that they were not employed because of job-related reasons (17%, compared to 2% for those who prefer the male breadwinner model) and "other" reasons (34%, compared to 12% for those who prefer the male breadwinner model).
- Among mothers who were not employed overall, these other (i.e., non-job related) reasons reflected a diversity of things, but ill health or disability and studying were commonly cited factors.7
- Similarly, mothers who preferred a male breadwinner model were more likely than those who did not to say they left their last job because of family or caring reasons, while factors related to health and disability were more common in the "less traditional" group, as were "other" reasons.
|Agree (traditional) (%)||Neutral (%)||Disagree (non-traditional) (%)||All not-employed mothers (%)|
|Notes: a More than one reason could be chosen, so percentages add to more than 100%. b Includes: "prefers to look after own child(ren) themselves", "too busy with family", "wants to continue breastfeeding" and "have had another baby". c Includes: "partner earns enough to support them", "it's not worthwhile with child care costs" and "would lose government benefits if worked". d Includes: "no jobs available", "can't find a job that interests" and "can't find a job with enough flexibility". No one gave the reason "can't get suitable child care". e Percentages may not total exactly 100.0% due to rounding. f Includes: "pregnant/to have children", "looking after family members or ageing parents", "looking after children" and "child care too expensive, unsuitable, unavailable". g Includes: "lost job (retrenched, made redundant, employer went out of business, dismissed, no work available)", "job ended/temporary/seasonal", "unsatisfactory work arrangements" and "self-employed business closed down for economic reasons (went broke, liquidated, no work, no supply or demand)". h Includes: "studying, returning to study", "moved to another location" and "other".
Chi-square tests used to compare distributions according to not-employed mothers' support of male breadwinner model: family reasons (χ2 (2, n = 317) = 234.45, p < .001); not worthwhile (χ2 (2, n = 317) = 38.40, p > .05, n. s.); no suitable jobs (χ2 (2, n = 317) = 170.08, p < .001); other (χ2 (2, n = 317) = 159.44, p < .01); main reason stopped working (χ2 (6, n = 301) = 293.94, p < .01). *** p < .001; ** p < .01.
|All reasons not in paid work a|
|Family reasons b||82.9||73.6||51.0||74.1 ***|
|Not worthwhile/lose benefits/partner earnings c||5.9||11.7||5.1||7.8|
|No suitable jobs d||2.2||13.4||16.9||8.7 ***|
|Main reason stopped working e|
|Family/caring reasons f||68.1||75.1||43.5||66.2|
|Own ill health/injury/disability||5.5||6.1||15.4||7.4|
|Other (studying, moved location, other) h||7.3||4.4||22.6||9.0|
|No. of observations||147||112||58||317|
From these data it appears that constraints form part of the explanation for mothers' non-employment, and more so among those who have less traditional gender role attitudes.
For employed mothers whose gender role attitudes suggest that they may prefer to have a secondary rather than primary role as earner in the family, it is relevant to consider how they perceive their involvement in paid work. Ideally we would examine mothers' reasons for employment, but as this information is not available, the analyses here focus on mothers' reports of whether work has a positive effect on her children and whether working makes her a better parent (Table 3.6). This information gives some insights on whether there are different views on positive outcomes of employment according to differences in views about the male breadwinner model.
|Agree (traditional) (%)||Neutral (%)||Disagree (non-traditional) (%)||All employed mothers (%)|
|Notes: Chi-square tests used to compare distributions according to employed mothers' support of male breadwinner model: positive effect on children (χ2 (4, n = 1817) = 52.45, p < .001); makes me a better parent (χ2 (4, n = 1817) = 105.48, p < .001). Percentages may not total exactly 100.0% due to rounding.|
|My working has a positive effect on my child(ren)|
|Strongly agree or agree||65.2||64.9||76.6||70.0|
|Neither agree nor disagree||24.8||29.1||18.5||23.5|
|Strongly disagree or disagree||10.0||6.0||4.9||6.5|
|The fact that I work makes me a better parent|
|Strongly agree or agree||45.9||46.5||62.5||53.1|
|Neither agree nor disagree||33.4||42.8||28.3||34.4|
|Strongly disagree or disagree||20.7||10.6||9.2||12.5|
|No. of observations||421||603||793||1,817|
There were significant differences on the two items examined, with employed mothers who did not support a male breadwinner model being more likely to feel that their work had a positive effect on children and on their parenting. Nevertheless, what is important to note is that many mothers who supported the male breadwinner model saw positive aspects in their working, with a minority of employed mothers, even those who support a male breadwinner model, feeling that their work did not have a positive effect on their children, or did not make them a better parent. For example, 65% of employed mothers with traditional views agreed or strongly agreed that their working has a positive effect on their children, while 10% disagreed or strongly disagreed. These findings no doubt reflect that having an overall view that a male breadwinner model might be better often co-exists with an appreciation of the positive effects of employment. Gender role attitudes would be one of many factors contributing to decision-making about maternal employment. Also, as noted above, valuing a male breadwinner model does not necessarily mean that mothers themselves (or their partners) prefer that mothers were not employed at all, but suggests they might prefer to take a secondary role as income earners, and thus might prefer to work part-time rather than full-time hours.
Table 3.7 shows the relationships between beliefs about the equal sharing of housework/child care and parents' work hours. There were actually no statistically significant differences in the distribution of responses to this item according to mothers' or fathers' work hours. When other characteristics were controlled, some significant differences emerged for families in which mothers worked part-time hours. Compared to families in which mothers were working full-time hours or were not employed, if mothers worked part-time hours, fathers expressed somewhat less agreement that there should be equal sharing of housework and child care. While not statistically significant, the same finding was apparent for mothers. This may reflect a view by some that mothers' part-time work should not necessarily involve equal sharing at home, given that fathers are doing a disproportionate share of the paid work, though the majority of mothers working part-time hours, and fathers with partners working part-time hours, agreed or strongly agreed with this statement.
|Mothers (%)||Fathers (%)|
|Notes: Chi-square tests used to compare distributions: mothers' views by mothers' employment: χ2 (4, n = 2115) = 16.56, p > .05 (n. s.); mothers' views by fathers' employment: χ2 (2, n = 2115) = 4.47, p > .05 (n. s.); fathers' views by mothers' employment: χ2 (4, n = 2115) = 16.69, p > .05 (n. s.); fathers' views by fathers' employment: χ2 (2, n = 2115) = 1.34, p > .05 (n. s.). Percentages may not total exactly 100.0% due to rounding.|
|Not employed or part-time employed||83.0||11.1||5.9||100.0||75.0||19.6||5.4||100.0|
|All mothers and fathers||82.7||13.4||3.9||100.0||76.7||19.2||4.1||100.0|
|No. of observations||1,756||277||82||2,115||1,619||406||90||2,115|
3.5 Socio-demographic characteristics and gender roles
This section turns to the research question concerning the key demographic factors related to mothers and fathers having more traditional gender role attitudes.
Demographic characteristics examined include parents' own characteristics (educational attainment, religion, main language spoken at home, and age) and family characteristics (marital status, age of youngest child, and number of children aged under 15 years in the family). These variables were selected for analyses given findings from previous research that have found differences across these characteristics (e.g., Van Egmond et al., 2010; see also reviews by Coltrane, 2000; Davis & Greenstein, 2009). The distributions of each variable are shown in Table 3.8. As with all analyses in this chapter, only partnered parents' responses are included.
|Characteristics||Mothers (%)||Fathers (%)|
|Bachelor or higher||35.2||34.7|
|Main language spoken at home|
|Under 40 years||17.0||10.7|
|50 years or older||10.4||22.6|
|Number of children aged < 15 years|
|3 or more||26.3||26.3|
|Age of youngest child|
|No. of observations||2,118||2,118|
As in the earlier section that examined parental employment, the analyses presented here are based on the relationships presented in Table 3.9 for the male breadwinner model and Table 3.10 for equal sharing, but references are also made to additional analyses in which other characteristics were controlled.8
The findings discussed refer to associations with views about the male breadwinner model, and about equal sharing, for mothers and fathers. These findings are presented in this section, taking each of the characteristics examined one at a time.
- Mothers' gender role attitudes varied significantly with educational attainment, whether looking at support of the male breadwinner model or equal sharing of unpaid work (albeit a much weaker finding for this item). Higher education was associated with less traditional views on both items. Fathers' views on these items did not vary significantly with his education in the independent analyses presented in Tables 3.9 and 3.10. However, when other characteristics were controlled, fathers with a bachelor degree or higher were less likely to support the male breadwinner model and were more likely to agree with equal sharing of household work when compared to fathers with incomplete secondary education.
- Differences in attitudes toward the male breadwinner model were relatively large when compared for mothers and fathers who did and did not identify with a religion: those who identified with a religion were more likely to support a male breadwinner model. In the analyses presented in Table 3.10, mothers' and fathers' views about equal sharing were not significantly related to religion, but after controlling for other characteristics, those who identified with a religion were less likely to agree about equal sharing.9
- Mothers and fathers who spoke a language other than English at home were significantly more likely to support a male breadwinner model than mothers and fathers mainly speaking English at home. Mothers' and fathers' views about equal sharing of child care and housework when both parents work did not vary significantly according to whether or not the main language spoken at home was English.
- No significant differences in gender role attitudes emerged according to these categories of ages as presented in Tables 3.9 and 3.10; however, when other variables were controlled, older mothers (aged 50 years or over) were a little more likely to support equal sharing of child care and housework compared to the youngest mothers (aged under 40 years).
- Mothers in married, rather than cohabiting, relationships were more supportive of the male breadwinner model, and this was also significant when other characteristics were controlled. No other differences according to marital status were apparent.
- Fathers' (but not mothers') views about the male breadwinner model varied according to the number of children in the family, with more children associated with greater support of the male breadwinner model. For both parents, views about equal sharing did not vary significantly by number of children. When other characteristics were controlled, no statistically significant differences by number of children were apparent for either of these items. Note that this count of children is based on children aged under 15 years living at home. Many families have older children who were not included in these counts.
- Fathers were more likely to support a male breadwinner model when there was a child aged under 7 years in the family, but these views did not vary by age of youngest child once other characteristics were controlled. In the more detailed analyses including other characteristics, mothers with a youngest child aged 7-11 years were more likely than those with a youngest child aged 0-6 years to agree that there should be equal sharing of household work when both parents work.
|Characteristics||Mothers (%)||Fathers (%)|
|Agree (traditional)||Neutral||Disagree (non-traditional)||Total||Agree (traditional)||Neutral||Disagree (non-traditional)||Total|
|Notes: Significance of differences tested using chi-square. *** p < .001; ** p < .01; * p < .05; n. s. p > .05. Percentages may not total exactly 100.0% due to rounding.|
|Highest qualification||***||n. s.|
|Bachelor or higher||24.1||30.6||45.4||100.0||26.3||35.4||38.2||100.0|
|Main language spoken at home||**||***|
|Age||n. s.||n. s.|
|Under 40 years||29.1||32.8||38.1||100.0||26.4||40.4||33.2||100.0|
|50 years or older||28.5||31.4||40.1||100.0||29.8||35.3||34.9||100.0|
|Marital status||*||n. s.|
|Number of children aged < 15 years||n. s.||*|
|3 or more||31.6||33.7||34.7||100.0||31.6||38.0||30.4||100.0|
|Age of youngest child||n. s.||*|
|Characteristics||Mothers (%)||Fathers (%)|
|Notes: Significance of differences tested using chi-square. *** p < .001; ** p < .01; * p < .05; n. s. p > .05. Percentages may not total exactly 100.0% due to rounding.|
|Highest qualification||**||n. s.|
|Bachelor or higher||86.7||9.7||3.6||100.0||80.6||15.7||3.7||100.0|
|Religion||n. s.||n. s.|
|Main language spoken at home||n. s.||n. s.|
|Age||n. s.||n. s.|
|Under 40 years||79.7||13.9||6.4||100.0||80.9||15.3||3.8||100.0|
|50 years or older||82.1||13.7||4.3||100.0||77.8||19.2||2.9||100.0|
|Marital status||n. s.||n. s.|
|Number of children aged < 15 years||n. s.||n. s.|
|3 or more||83.6||12.3||4.2||100.0||77.4||18.8||3.8||100.0|
|Age of youngest child||n. s.||n. s.|
The above findings are generally consistent with the wider literature on predictors of gender role attitudes. The lack of significant findings for most variables in regard to equal sharing reflects the widespread agreement with this item. The lack of differences for some of the variables that are observed to be related to different gender role attitudes in the wider population possibly reflects that this LSAC sample is less diverse than the wider population, especially in terms of ages of respondents and family composition. Also, the lack of variation in responses to the item concerning equal sharing when both parents work makes this item less discriminating with regard to gender role attitudes, and so making it difficult to differentiate responses across groups in the population.
3.6 Couples' gender role attitudes
One of the strengths of these LSAC data is having a large sample of couple-parent families, in which gender role attitudes (and other characteristics) are available for both parents. Of particular interest is the extent to which there is alignment within couples on these items. We would expect some alignment, with the theory of assortative mating predicting that individuals choose partners with similar characteristics to themselves. Typically, this is considered in terms of demographic characteristics, but it may also extend to views about gender roles, especially if such views have implications for how tasks would be allocated within the immediate family. Kalmijn (2005), for example, examined this using a sample of couples in the Netherlands. The gender role attitudes of husbands and wives were significantly correlated at a point in time in this research, and partners' attitudes became more aligned over time. While such attitudes may be correlated, there is also likely to be diversity across families. For example, Marks, Lam, and McHale (2009) found in a sample of middle-class US families that gender role attitudes were often, but not always, shared by husbands and wives. Their analyses, which also incorporated information on the gender role attitudes of adolescents in the family, found families clustered into more traditional families, more egalitarian families and divergent families. In the traditional families, mothers tended to be more traditional in their gender role attitudes than the fathers, and in the divergent families, fathers tended to be more traditional than the mothers.
Table 3.11 shows the within-couple associations between mothers' and fathers' responses on the male breadwinner item using the LSAC data. Just as was evident from individual-level responses, when explored from the perspective of couples, there was also considerable diversity:
- Overall, in 19% of couples both parents were classified as non-traditional, with both disagreeing that the male breadwinner model was better.
- In another 13%, both were classified as traditional, agreeing that the male breadwinner model was better.
- It was similarly not common for mothers and fathers to have completely opposite views (in only 6% of couples did the mother disagree that the male breadwinner model was better while the father agreed, and in another 6% of couples the mother agreed that the male breadwinner was better but the father did not agree).
- There were numerous combinations in between when also taking into account the possibility of one or both parents having ambivalent or undecided views.
|Mothers||Fathers (overall %)||Fathers (as % of mothers)|
|Agree (traditional)||Neutral||Disagree (non-traditional)||Total||Agree (traditional)||Neutral||Disagree (non-traditional)||Total|
|Note: Percentages may not total exactly 100.0% due to rounding.|
|No. of observations||550||790||806||2,146||550||790||806||2,146|
The degree of correspondence between mothers and fathers is evident when fathers' responses are examined for each of the groupings of mothers' responses:
- Between 43% and 50% of fathers gave the same (grouped) rating as their partner.
- In families of non-traditional mothers (who did not agree with the male breadwinner model), approximately 16% of fathers were distinctly more traditional, agreeing that the breadwinner model was better.
- Of fathers who had a more traditional partner, 21% were themselves non-traditional, disagreeing that the male breadwinner model was better.
As previously discussed, most mothers and fathers agreed that child care and other household work should be equally shared when both parents work. When analysed at the couple level, Table 3.12 shows that:
- in 65% of couples both parents agreed with this;
- in another 24% of couples one parent agreed but the other was ambivalent or undecided; and
- there were small numbers showing other combinations, including 3% of couples in which mothers thought the household work should be equally shared but fathers did not; and another 3% with the opposite situation.
|Mothers||Fathers (overall %)||Fathers (as % of mothers)|
|Note: Percentages may not total exactly 100.0% due to rounding.|
|No. of observations||1,643||410||92||2,145||1,643||410||92||2,145|
Table 3.12 also shows that if fathers' views are examined in relation to mothers' views, a relatively high proportion of fathers agreed that there should be equal sharing of household work among those whose partner believed this. Fathers were somewhat more ambivalent or undecided when mothers were ambivalent or undecided or disagreed with this item, although still a large majority believed in equal sharing of household work in these families.
An interesting direction for future research with these data might be to explore the demographic characteristics of families according to the degree of correspondence between parents in their gender role attitudes. It would also be interesting to explore whether relationship quality and other aspects of family wellbeing vary when parents have discordant gender role attitudes.
3.7 Couples' gender role attitudes and sharing of paid and unpaid work
Extending the above analyses, the gender role attitudes can be related to information collected in LSAC on the time parents spend doing child care and household work, along with time spent on paid work. As noted previously, being cross-sectional in nature, these analyses cannot inform on whether different gender role attitudes lead to particular time-use patterns, or whether they are a reflection of them. The analyses presented here provide a first look at these associations, with more detailed analyses possible to explore how parental and family characteristics also contribute to differences in parents' time use in the context of different gender role attitudes.
For these analyses, we consider time spent in child care and in other domestic work separately, given that these activities offer very different rewards to parents, and the amount of time spent on these activities might be determined by different factors (Bulanda, 2004; Deutsch, Lussier, & Servis, 1993). For example, looking at predictors of fathers' child care time and housework time using Waves 2 and 3 of LSAC, fathers with better mental health undertook more child care, but less housework (Baxter & Smart, 2010). Better relationship quality was related to fathers doing more child care, but was unrelated to the amount of housework done. Parents might have quite different motivations and processes for sharing child care compared to that of sharing housework, such that gender role attitudes might matter more to one type of household work than the other (Bulanda, 2004).
Previous research has highlighted how gender role attitudes are linked with patterns of participation in paid work and in unpaid household work and child care. In particular, fathers who have less traditional, or more egalitarian, gender role attitudes are more often involved in these activities in the home, leading to more equal sharing of housework (Greenstein, 1996b) and child care (Bulanda, 2004; Jacobs & Kelley, 2006). Women's gender role attitudes are also important in considering how unpaid family work is shared, with more egalitarian views likely to be linked with more equal sharing in the home. In contrast, if women have traditionally gendered views, fathers may actually be discouraged from their involvement, as is especially noted in respect of mothers' "gate-keeping" of child care activities and other household tasks (Allen & Hawkins, 1999). See also Coltrane (2000) and Davis and Greenstein (2009) for reviews of the extensive literature in this field.
No doubt there is a complex interplay of attitudes and behaviours in families, especially when mothers and fathers have divergent gender role attitudes. In this section, parents' time use is related to mothers' gender role attitudes, and also within-couple gender role attitudes. This research especially extends previous research in this area by making use of up-to-date couple-level data. The focus throughout this section is on views about the male breadwinner model, as this item provides some differentiation between families that is not so apparent with the item on equal sharing, given the widespread agreement by parents on this item.
Information on time spent doing unpaid child care was collected from each parent with the question, "How much time per week do you personally spend playing with your children, helping them with personal care, teaching, coaching or actively supervising them, getting them to child care, school or other activities?" For other household work, parents were asked "How much time per week do you personally spend on domestic tasks such as housework, home maintenance, shopping and cooking?" We refer to these two estimates as time spent in child care and in housework respectively. Information on parents' usual hours spent in employment is included in these analyses, and referred to as time in paid work. Among the in-scope sample:
- mothers spent an average of 24 hours per week in paid work, 19 hours per week doing child care, and 20 hours per week doing housework; and
- fathers spent, on average, 45 hours doing paid work, 10 hours doing child care, and 10 hours doing housework per week.
These estimates are comparable to those obtained using HILDA.10
How do these gendered time-use patterns vary by mothers' gender role attitudes? We saw above that mothers who spent few (or no) hours in paid work had more traditional gender role attitudes (as assessed on views about the male breadwinner model) than those working full-time. Not surprisingly, then, Table 3.13 shows that mothers with more traditional gender role attitudes spent the least hours in paid work (average of 18 hours per week) compared to those with non-traditional views (average of 31 hours per week).
|Notes: Parents who were not employed were recorded as spending zero hours in paid work. The averages were compared across responses to the male breadwinner model item using analysis of variance: mothers' paid work (F = 111.19, df = 2, p < .001); mothers' child care (F = 8.47, df = 2, p < .001); mothers' housework (F = 39.05, df = 2, p < .001); fathers' paid work (F = 3.04, df = 2, p < .05); fathers' child care (F = 5.21, df = 2, p < .01); fathers' housework (F = 11.42, df = 2, p < .001); mothers' per cent of parental paid work (F = 99.32, df = 2, p < .001); mothers' per cent of parental child care (F = 11.15, df = 2, p < .001); mothers' per cent of parental housework (F = 27.80, df = 2, p < .001). *** p < .001; ** p < .01; * p < .05.|
|Mothers, average hours per week|
|Paid work||17.5||23.0||30.5||24.3 ***|
|Child care||20.9||19.3||18.3||19.3 *|
|Fathers, average hours per week|
|Paid work||44.7||46.5||44.8||45.3 *|
|Child care||9.2||9.9||10.8||10.1 **|
|Mothers, average % of total parent time|
|Paid work||24.7||30.1||38.6||31.8 ***|
|Child care||66.3||64.2||61.9||63.9 **|
|No. of observations||567||710||839||2,116|
Mothers with more traditional views spent more hours per week doing child care and doing other housework. There is a difference of 2.6 hours per week on child care and 5.2 hours per week on housework if comparing the mothers with views that are more traditional to those with non-traditional views. The difference in paid work hours across these groups was much larger, at 13 hours per week. It is interesting to note that compared to time in paid work and time doing housework, mothers' time spent caring for their children varied less according to mothers' gender role attitudes. This may reflect that child care is valued by mothers, regardless of their gender role attitudes, with time for child care perhaps protected by mothers varying the time they spend in paid work and housework, or reducing time for other activities.
When mothers had more traditional views, fathers did somewhat less child care (a difference of 1.6 hours per week, comparing the traditional and non-traditional mothers) and somewhat less housework (2 hours difference). While the paid work hours of fathers also varied across the groups, it was actually highest when mothers were ambivalent or undecided about the male breadwinner model.
One way of summarising these data is to calculate (in percentage terms) how much of the total parental paid work, child care and other household work is done by each parent. Overall, mothers did 32% of the paid work, 64% of the child care and 67% of the household work. Reflecting the patterns described above, mothers whose views were more traditional did less of the parental paid work and proportionately more of the parental child care and housework. That is, mothers did the majority of child care and housework, regardless of their views about the male breadwinner model, but they did a greater share when they had more conservative views on this.
This can be extended to examine how these time-use patterns vary for the different combinations of mothers' and fathers' gender role attitudes - again, focusing on their views on the male breadwinner model (Figure 3.1). For simplicity, the percentage of child care, housework and paid work done by mothers is shown. To describe the findings, results for ambivalent or undecided mothers and fathers are put aside to concentrate on findings related to the non-traditional and traditional mothers and fathers.
Figure 3.1: Mothers' share of parental time in paid work, child care and housework, by couple-level agreement on male breadwinner model
There are different patterns when comparing one extreme to the other; that is, comparing families in which both parents have non-traditional views to those in which both parents have more traditional views. For example, when both parents have traditional views, mothers do 69% of the child care and 75% of the other household work. In contrast, when both parents have non-traditional views, mothers do less (60% of the child care and 59% of the housework). Parents' time in paid work also varies in line with what would be expected. These data indicate that parents' time allocation is related to both parents' gender role attitudes, rather than one parent's alone. That is, regardless of mothers' own gender role attitudes, the percentage of the parental child care and housework done by them is lower when fathers' gender role attitudes are non-traditional. Of course, as discussed previously, given these are cross-sectional associations, it is not appropriate to assume that attitudes have caused these time-use patterns. For example, it may be that attitudes reflect the established time-use patterns of parents - as predicted by cognitive dissonance theory (Festinger, 1957).
Greenstein's (1996b) study found that mothers' gender role attitudes to some extent superseded those of fathers when explaining variation in parents' share of housework. When mothers were more traditional, it did not matter whether or not fathers had non-traditional or traditional views - fathers' share of housework remained low. This suggested some degree of gate-keeping of this household work by mothers who had especially traditional views about gender roles. This finding is not apparent here, perhaps reflecting the more contemporary nature of these data compared to the 1987-88 data used in Greenstein's study. The findings from the LSAC analyses are not unique, with several studies finding that fathers' gender role attitudes matter in explaining variation in fathers' participation in housework or child care. Looking at child care involvement, for example, Bulanda (2004) found that fathers' involvement was more strongly predicted by fathers' gender role attitudes than by mothers', with no evidence that fathers' involvement was lower when mothers had more traditional attitudes, as might be suggested if these mothers were "gate-keeping" these child care activities.
Overall, these percentages suggest that when both parents have traditional views, then the sharing of paid and unpaid work is also more "traditional" or less equal, and when both parents have non-traditional views, there is more equal sharing of paid and unpaid work. Even so, the sharing of paid and unpaid work in these non-traditional families approaches but does not achieve equality. These percentages of course fail to highlight the underlying variation in mothers' and fathers' time spent on activities, since variation in percentages may reflect variation in mothers' time alone, in fathers' time alone, or in both. These data offer the opportunity examine these time-use patterns more fully, also in the context of different parental and family characteristics.
3.8 Perceived fairness of child care and housework time and gender role attitudes
It is frequently observed that despite the uneven gender distribution of housework and child care within couples, parents often report that the allocation of these activities between themselves is fair (Baxter, 2000; Blair, 1998; Blair & Johnson, 1992; Coltrane, 2000; Greenstein, 1996a; Thompson, 1991).
In earlier analyses of the LSAC data, for example, Baxter and Smart (2010) found that 44% of mothers reported that they did their fair share and 55% more than their fair share of the child care. Mothers who reported that the distribution of child care between themselves and their partner was fair did an average of 64% of the parental child care compared to 71% for those doing more than their fair share. Also, 38% of mothers reported doing their fair share of housework and 60% reported doing more than their fair share. Those who reported that the distribution of housework was fair did 67% of the housework, on average, and those who reported that they did more than their fair share did 76% of the housework, on average. Baxter and Smart's analyses drew upon both cohorts and earlier waves of LSAC, and so children were somewhat younger than those analysed in this chapter.
For this final analytical section, the time-use data and gender role attitudes are examined in relation to mothers' reports of fairness of the parental sharing of housework and child care. For simplicity, here we refer only to mothers' gender role attitudes and mothers' reports of the fairness of sharing of housework and child care. Clearly, extending this to take account of fathers' views is a direction for further research. For this subsection, as with the above subsection, the analyses have not yet been extended to consider how associations vary when other family or parental characteristics are taken into account.
The information about fairness is first examined without consideration of gender role attitudes. Overall it is expected that, like Baxter and Smart's earlier analyses, perceptions of fairness will be more likely when the sharing of tasks is more equal. However, as discussed in many of the studies on this topic, perceptions of fairness appear to be formed within broader frameworks and contexts than this, such that the division of household work between partners will not always predict parents' perceptions of fairness, and it does not take a 50/50 split to be seen as "fair". (See, for example, Baxter, 2000; Blair & Johnson, 1992; Thompson, 1991).
Looking first at child care:
- when mothers reported that they did their fair share of child care (53% of mothers), they did 62% of the child care; and
- when they reported doing more than or much more than their fair share (46% of mothers), they did 67% of the child care.11
The differences in the average amount of time mothers spent doing child care for these groups is quite small, with 19.1 hours per week child care done by mothers who said they did their fair share and 19.8 hours per week by mothers who did more than their fair share.
The amount of time fathers spent on child care in these two groups contributes to the difference in percentages, with fathers doing 10.9 hours per week when mothers reported that they did their fair share of child care, and fathers doing 9.0 hours per week when mothers reported themselves doing more than their fair share.
With respect to housework:
- mothers reporting doing less than their fair share (4% of mothers) did 46% of the housework;
- mothers reporting doing their fair share (43%) did 62% of the housework; and
- mothers reporting doing more than their fair share (53%), did 72% of the housework.
Differences between the "fair" and "more than fair" groups reflect different amounts of time mothers spend doing housework (19.3 hours and 21.8 hours per week for these two groups respectively), as well as time that fathers spend doing housework (11.2 hours and 8.1 hours per week respectively).
Introducing gender role attitudes into these analyses, it is first worth noting that "traditional" mothers were more likely to say they did more than their fair share of child care (56%) and housework (63%), compared to non-traditional mothers (40% and 48% respectively). Of course, if mothers make assessments about fairness based on their relative contribution to these tasks, this may simply reflect that mothers with more traditional gender role attitudes do a higher percentage of each of these types of household labour.
The aim of this analysis is to explore whether reports of fairness are to some extent moderated by gender role attitudes (see Greenstein, 1996a, for a paper on this topic). That is, when mothers have more traditional views, do they have lesser expectations of their partner in terms of how much of the child care or housework they should do, such that a "fair" share of these activities occurs when mothers are doing a relatively high proportion of them?
Figure 3.2 shows that a "fair" housework share is 58% of parental housework for non-traditional mothers and 68% of parental housework for traditional mothers. This compares to the average share of housework for mothers who say they are doing more than their fair share of 70% when mothers are non-traditional, compared to 75% when mothers are more traditional.
Figure 3.2: Mothers' share of child care and housework, by fairness and gender role attitudes of mothers
Figure 3.2 also shows the fairness data in relation to child care, for which differences are not as marked as they were for housework. For child care, a "fair" share is 60% of parental child care for non-traditional mothers and just a little higher at 63% for traditional mothers, while mothers saying they do more than their fair share of child care, on average, do 65% of the child care when mothers are non-traditional and 69% when mothers are more traditional.
This quite simple analysis suggests that standards by which fairness is assessed vary according to mothers' gender role attitudes. But as mothers' gender role attitudes vary with a number of parental and family characteristics, and especially with mothers' own time in paid work, it would be important in future research to examine whether these findings are explained more by these varying contextual factors, rather than the gender role attitudes themselves. This first view of these relationships provides some insights that can be explored in future research with these data.
3.9 Discussion and conclusion
This chapter provides a first exploration of the gender role attitudes of mothers and fathers in the K cohort of LSAC, as collected in Wave 5 when the study children were aged 12-13 years. While this sample is somewhat specific when compared to larger population-based surveys, it allows insights on the gender role attitudes of parents that are raising young adolescent children, who are themselves likely to be forming their own gender role attitudes. The large sample size and the availability of couple-level data made it possible to explore within-family attitudes and time use to gain further insights on associations between gender role attitudes of parents and gendered time use patterns within the home.
These analyses showed that parents had varied views regarding the male breadwinner model, but there was widespread agreement that household work should be shared when both parents work. That is, while some parents expressed a view that "it is better for the family if the husband is the principal breadwinner outside the home and the wife has primary responsibility for the home and the children", most nevertheless believed that, should both parents be working, the housework and child care should be shared equally. These findings are consistent with broader studies of gender role attitudes in Australia (e.g., Van Egmond et al., 2010) and elsewhere (see reviews by Coltrane, 2000; Davis & Greenstein, 2009).
The different findings for the two items likely reflect that each captures different perspectives on gender roles. According to Davis and Greenstein's (2009) analysis of commonly used gender role ideology questions, the male breadwinner item captures perspectives on "belief in gendered separate spheres", while the equal sharing item captures perspectives on "household utility". The analysis here suggests that while we can classify parents as being "traditional" or "non-traditional" in their gender role attitudes using views about the male breadwinner model, within the family the attitude is more one of equality, even in families of "traditional" parents.
In these analyses, because the majority of parents agreed that household work should be shared, there were not particularly significant findings regarding the characteristics of mothers and fathers who responded differently against this item. However, with regard to views on the male breadwinner model, differences according to parents' educational attainment (although only for mothers), religion and ethnicity emerged. Mothers with higher educational attainment were less likely to have "traditional" gender role attitudes. Mothers and fathers who identified with a religion were more likely to have traditional gender role attitudes, as were those who mainly spoke a language other than English at home. There were more minor differences according to age and family composition.
Not surprisingly, there were significant associations between mothers' employment and views about the male breadwinner model, with "non-traditional" mothers more often employed (and more often employed full-time), compared to "traditional" mothers. However, we are unable to say with these data whether mothers' employment contributes to, or is a consequence of, their gender role attitudes. Analyses of mothers whose attitudes do not align with their employment participation revealed that "non-traditional" mothers who are not employed seem to be particularly constrained in their employment behaviour by issues of disability or ill health, as well as other barriers beyond those relating to caring for children. On the other hand, analyses of perceptions about positive aspects of work among the more "traditional" mothers who are employed revealed that, while these mothers are less likely than other mothers to perceive positive effects on children and parenting of their working, still a majority perceived there to be positive effects. For parents, choices about work and caring are not always straightforward, and these data remind us that while mothers might have more traditional gender role attitudes, they might also personally value their employment. Such complexities are often noted in qualitative research on motherhood ideals and maternal decision-making about employment and child care (e.g., Crompton & Lyonette; Hand & Baxter, 2012; Himmelweit & Sigala, 2004; McRae, 2003). Being able to see these complexities in a large-scale survey of parents helps to contextualise the rich qualitative research.
Decisions about work and caring are of course made within a family context in the case of couple families, and so these data proved especially useful in being able to examine within-couple gender role attitudes, adding to a limited range of research on couple-level gender role attitudes (see Kalmijn, 2005; Marks, Lam, & McHale, 2009). Partners did not often have opposing views on either of the items examined, but there were some families in which this occurred (in 12% of families, partners had opposing views regarding the male breadwinner model and in 6% of families, partners had opposing views about equal sharing). Overall there was considerable diversity of views within couples.
Australian mothers very often work part-time rather than full-time hours, which for some may be a way they can maintain a more "traditional" allocation of time; permitting time to be spent in paid work as well as allowing time for caring for children. Mothers working part-time rather than full-time may therefore take on more of the housework and child care. The analyses of LSAC data showed that parents were less likely to agree that there should be equal sharing in the home when mothers worked part-time hours, presumably because the allocation of their time to paid work is not seen to be "equal".
The gender role attitudes (as measured against views of the male breadwinner model) of mothers as well as fathers seemed to matter when exploring how the child care and housework was shared within couples, as has been observed elsewhere (see review by Davis & Greenstein, 2009). Overall, mothers did a disproportionate share of the household work, and that share tended to be more uneven - with more done by mothers - when either mothers or fathers had more traditional gender role attitudes. That is, the most equitable sharing of child care and housework was apparent when both parents expressed non-traditional views, and the least equitable sharing was apparent when both parents expressed more traditional views, consistent with those studies that report both mothers' and fathers' gender role attitudes explain variation in unpaid household work (e.g., Bulanda, 2004).
The division of parental time on housework was more sensitive to these gender role attitudes than was the division of parental time on child care, with mothers' time on child care varying a relatively small amount according to gender role attitudes. This may reflect that child care is an activity that is more rewarding and enjoyable than is doing housework, and so "traditional" and "non-traditional" mothers alike may protect their time with children, instead sacrificing time for housework, paid work or other activities.
These data were also used to explore associations between parents' time use, mothers' sense of fairness of child care and housework time, and gender role attitudes of mothers. This is especially of interest, in order to understand why many mothers perceive the gendered division of household tasks to be fair, despite the inequity of time spent on those tasks (see reviews by Coltrane, 2000 and Davis & Greenstein, 2009). Does the gender role attitude of mothers make a difference to mothers' perceptions of fairness? These data provided some evidence that a "fair" distribution of child care and housework involves mothers doing a greater share of the parental child care when mothers are more "traditional" versus "non-traditional", perhaps suggesting that gender role attitudes shift the reference point at which mothers consider the distribution of these tasks to be fair or not. This, however, should be read as an early finding, as it would be important to consider these relationships more fully, being mindful of other parental and family characteristics that may relate to perceptions of fairness.
This analysis is not without limitations. The main one is the reliance on one or two items to capture gender role attitudes, as described above. The other main limitation is the use of cross-sectional rather than longitudinal data to explore these relationships, which was a constraint given gender role attitude questions were not asked at previous main waves of LSAC. However, the great strength of these data is in having a large sample of perspectives of both parents, albeit for a subset of couple families in the study. As shown here, having couple data allows more insights on family processes than may be gained with the perspective of only one parent.
Exploring these gender role attitudes among parents is particularly insightful for a number of reasons, not just because these family contexts may be shaping the attitudes of children in those families. Beyond this, understanding these attitudes provides some insights on family decision-making around work and family, bringing in the attitudes of fathers as well as mothers. This chapter has provided an overview of some ways in which attitudes are linked to behaviours. In the future, it will be interesting to determine whether parental wellbeing and family functioning vary at all with these different attitudes, or when there is conflict between parents' attitudes and behaviours, or between parents' attitudes within couples. In the longer term, we will be able to see whether (and if so, how) these attitudes flow through to the later attitudes, aspirations and employment of children.
- Allen, S. M., & Hawkins, A. J. (1999). Maternal gatekeeping: Mothers' beliefs and behaviors that inhibit greater father involvement in family work. Journal of Marriage and Family, 61(1), 199-212.
- Baxter, J. A. (2013). Parents working out work (Australian Family Trends No. 1). Melbourne: Australian Institute of Family Studies.
- Baxter, J. A., & Smart, D. (2010). Fathering in Australia among couple families with young children (FaHCSIA Occasional Paper No. 37). Canberra: FaHCSIA.
- Baxter, J. H. (2000). The joys and justice of housework. Sociology, 34(4), 609-631.
- Baxter, J. H., & Western, M. (1997). Women's satisfaction with the domestic division of labour. Family Matters, 47, 16-20.
- Blair, S. L. (1998). Work roles, domestic roles, and marital quality: Perceptions of fairness among dual-earner couples. Social Justice Research, 11(3), 313-335.
- Blair, S. L., & Johnson, M. P. (1992). Wives' perceptions of the fairness of the division of household labor: The intersection of housework and ideology. Journal of Marriage and the Family, 570-581.
- Bulanda, R. E. (2004). Paternal involvement with children: The influence of gender ideologies. Journal of Marriage and Family, 66(1), 40-45.
- Coltrane, S. (2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family, 62(4), 1208-1233.
- Craig, L., & Mullan, K. (2011). How mothers and fathers share childcare. American Sociological Review, 76(6), 834-861.
- Crompton, R., & Lyonette, C. (2005). The new gender essentialism: Domestic and family "choices" and their relation to attitudes. The British Journal of Sociology, 56(4), 601-620.
- Davis, S. N., & Greenstein, T. N. (2009). Gender ideology: Components, predictors, and consequences. Annual Review of Sociology, 35, 87-105.
- Deutsch, F. M., Lussier, J. B., & Servis, L. J. (1993). Husbands at home: Predictors of paternal participation in childcare and housework. Journal of Personality and Social Psychology, 65, 1154-1166.
- Festinger, L. (1957). A theory of cognitive dissonance. Evanston, Illinois: Row, Peterson.
- Greenstein, T. N. (1996a). Gender ideology and perceptions of the fairness of the division of household labor: Effects on marital quality. Social Forces, 74(3), 1029-1042.
- Greenstein, T. N. (1996b). Husbands' participation in domestic labor: Interactive effects of wives' and husbands' gender ideologies. Journal of Marriage and the Family, 58(3), 585-595.
- Hand, K., & Baxter, J. (2013). Maternal employment and the care of school-aged children. Australian Journal of Labour Economics, 16(3), 329-349.
- Himmelweit, S., & Sigala, M. (2004). Choice and the relationship between identities and behaviour for mothers with pre-school children: Some implications for policy from a UK study. Journal of Social Policy, 33(3), 455-478.
- Hook, J. L. (2006). Care in context: Men's unpaid work in 20 countries, 1965-2003. American Sociological Review, 71(4), 639-660.
- Jacobs, J. N., & Kelley, M. L. (2006). Predictors of paternal involvement in childcare in dual-earner families with young children. Fathering: A Journal of Theory, Research, and Practice about Men as Fathers, 4(1), 23-47.
- Kalmijn, M. (2005). Attitude alignment in marriage and cohabitation: The case of sex-role attitudes. Personal Relationships, 12(4), 521-535.
- Marks, J. L., Lam, C. B., & McHale, S. M. (2009). Family patterns of gender role attitudes. Sex Roles, 61(3-4), 221-234.
- McRae, S. (2003). Constraints and choices in mothers' employment careers: A consideration of hakim's preference theory. British Journal of Sociology, 54(3), 317-338.
- Sayer, L. C. (2005). Gender, time and inequality: Trends in women's and men's paid work, unpaid work and free time. Social Forces, 84(1), 285-303.
- Schober, P., & Scott, J. (2012). Maternal employment and gender role attitudes: Dissonance among British men and women in the transition to parenthood. Work, Employment & Society, 26(3), 514-530.
- Shelton, B. A., & John, D. (1996). The division of household labor. Annual Review of Sociology, 22(1), 299-322.
- Stier, H., & Lewin-Epstein, N. (2000). Women's part-time employment and gender inequality in the family. Journal of Family Issues, 21(3), 390-410.
- Thompson, L. (1991). Family work women's sense of fairness. Journal of Family Issues, 12(2), 181-196.
- Van Egmond, M., Baxter, J., Buchler, S., & Western, M. (2010). A stalled revolution? Gender role attitudes in Australia, 1986-2005. Journal of Population Research, 27(3), 147-168.
- Wilkie, J. R., Ferree, M. M., & Ratcliff, K. S. (1998). Gender and fairness: Marital satisfaction in two-earner couples. Journal of Marriage and the Family, 60(3), 577-594.
1 An additional item collected was: "Ideally, there should be as many women as men in important positions in government and business". As this item broaches a somewhat different topic to that of the other items, having a focus on the workplace rather than the family, it is not used in this chapter.
2 Non-response usually reflects non-completion of the self-complete survey by Parent 2.
3 Among fathers within the in-scope sample, 7% were in part-time work and 93% in full-time work, with less than 1% not employed. For the sample without a partner response, 18% of fathers were not employed, 7% were part-time employed and 75% were full-time employed. This partner employment information was provided by the responding parent in the main interview. Among mothers within the in-scope sample (both partners responding), 17% were not employed, 53% were in part-time work and 31% were in full-time work, while within partnered families with no partner response, 30% of mothers were not employed, 41% were part-time employed and 30% were full-time employed.
4 In these numbers, "mothers" and "fathers" include single and partnered parents who are biological, adoptive, step- and foster parents of the LSAC study child. There were only 77 responding partnered fathers who had no partner response, so there were insufficient data to compare responses of this group to that of the in-scope sample.
5 Specifically, ordered logistic regression models were estimated on the ungrouped responses to each of the gender role items, for mothers and fathers. Variables entered in the models were the same as those used in the descriptive analyses: categories of mothers' and fathers' work hours were included in all models. Parents' own educational attainment, religion, speaking a language other than English at home and age were included, along with family characteristics of cohabiting versus married, number of children aged under 15 and age of youngest child.
7 Based on special data request for table of "Other, specify" responses for Parent 1.
8 For methods, refer to Footnote 5. The detailed results have not been presented here, but the most important variables in explaining variation in mothers' and fathers' views about the male breadwinner model were mothers' work hours, religion and English-language proficiency. Also, for mothers, higher educational attainment and cohabiting predicted less traditional views. For fathers, employment status was an important predictor. For mothers, those who were more highly educated, without a religion, older, and with children aged 7-11 years were most likely to agree about equal sharing, while those working part-time hours were least likely to agree. For fathers, those with higher educational attainment were more likely to agree about equal sharing, and those with a part-time employed partner were less likely to agree.
9 Information about religion was not available in Wave 5 and was carried over from Wave 4 of LSAC. If a classification of different forms of religion was used, the most "traditional" views were found among those in a category of "Other Christian" (including Uniting Church, Presbyterian, Greek Orthodox, Baptist, Lutheran and Other Christian religion), and "Other religion" (including Hinduism, Judaism, Islam, Buddhism and Other religion), rather than Catholic, Anglican or no religion. Mothers who identified as belonging to Anglican or "Other Christian" religions were significantly less likely to agree with equal sharing in the home when both parents work, when compared to those who identified as having no religion.
10 To derive comparable estimates from HILDA, partnered parents in HILDA Wave 12 were examined, and the sample weights were adjusted to give an age of youngest child distribution that was the same as that of the Wave 5 K cohort. From these re-weighted data it was estimated that mothers spent an average of 23 hours per week in paid work, 19 hours doing child care and 20 hours doing housework. Fathers spent an average of 42 hours per week in paid work, 10 hours doing child care and 7 hours doing housework. Because these time-use patterns do vary by age of youngest child, if we instead derive these estimates of the HILDA sample with children aged up to 13 years using the original sample weights, mothers spent an average of 18 hours per week in paid work, 31 hours doing child care and 21 hours doing housework, and fathers spent an average of 42 hours per week in paid work, 14 hours doing child care and 7 hours doing housework.
11 Throughout these analyses, mothers classified as "more than fair share" include those who said they did much more than their fair share. Fewer than 1% of mothers reported doing less or much less than their fair share of child care, and so these mothers have been excluded from this analysis.