Canadian Families’ Strategies for Employment and Care for Preschool Children

Abstract

Based on the 2006 Canadian Census “long form” sample of one in every five households, the authors develop a detailed typology of family strategies for employment and the care of preschool children. The analysis is restricted to opposite-sex couples with at least one child under age 6 and no older child or other adult in the household. The typology reveals the persistence of a highly gendered division of labor whereby mothers do more child care and men more paid employment. Multinomial regression reveals strong but nonmonotonic effects of age and education on a couple’s strategy. When the partners differ in age or education, the family’s choice of strategy depends on which partner is older and has more education. The authors interpret the findings as more consistent with gender theory perspectives on family models of employment and domestic work.

Keywords

Household Division of Labor Childcare Employment Gender Couples

Introduction

How couples with young children earn a living and share child care is a fundamental aspect of gender equality, the material condition of families, and labor supply. Questions about the time each family member spends on child care in the 2006 Canadian Census offer insight into partners’ joint decisions about paid employment and child care. In this article, we create a typology of household strategies for employment and child care and then examine the relationship between a family’s strategy and the number of young children in the family and the two partners’ ages and education, the two key measures of human capital.

By strategies we mean combinations of the two parents’ employment—whether they work full-time or part-time or are not employed—and how child care is shared between them. The typology does not make employment decisions the determinant of child care arrangements or vice versa, nor does it make one partner the “cause” of the other’s roles in employment and child care. This is a realistic accommodation to our cross-sectional data, which provide no empirical basis for untangling the mutual effects of decisions about employment and child care and of the partners’ roles.

Our typology is based on the time that both partners spend in paid employment and child care, reducing the thousands of combinations to a theoretically informed and usable scheme that embodies the relations between the partners’ employment and child care (Wallace, 2002). One might expect that figures showing how parents of young children juggle employment and child care would be a staple social statistic, describing a fundamental aspect of social life, with implications for gender equity and fertility as well as policy on day care and other aspects of the family. To our knowledge, however, these are the first and only such Canadian statistics. Using the typology, we provide regression estimates of the impact of the two partners’ education and age, along with the number of preschool children, on their family strategy. We consider these results in light of theoretical arguments from rational choice and gender perspectives.

Theorizing Couples’ Patterns of Work and Child Care

Rational choice arguments about family strategies rely on economic concepts, including household production, individual preferences, and human capital. The essential idea is that couples choose a strategy that maximizes their well-being. The key formulation is Gary Becker’s (1981, 1985; a fine review is Pollak, 2003) argument that the division between paid work and domestic work and child care depends on the labor market valuation of the two partners’ skills. According to Becker, the optimal strategy is to combine male market and female domestic roles, and these specialized traditional gender roles produce a better joint outcome that is more efficient than a more egalitarian arrangement.

This rational choice model assumes that couples adopt consensual, gender-neutral strategies to maximize their economic well-being. Differences in partners’ human capital determine the balance of market and household work. Departures from economic rationality, due to cultural—economists prefer the term institutional—effects of gender, racialization, and other aspects of identity are “residual.” If the value of child care is not a function of who provides it, decisions about employment precede and predict child care arrangements.

The fundamental criticisms of the neoclassical account of the household division of labor involve, first, the assumption that maximizing household utility reflects the best interests of both partners (Folbre, 2004) and, second, the exaggerated role of objective rational decision-making, relative to the “residual” effects of culture (Crow, 1989). An alternative model, still in the rational choice tradition, is the idea of noncooperative bargaining between partners. Rather than optimizing household utility, household strategies reflect differences in two partners’ preferences and relative bargaining power (Lundberg & Pollak, 1994, 1996; Schultz, 1990) and so their potentially unequal ability to pursue their preferred strategy and exit if the outcome is unsatisfactory (England & Folbre, 2005; Katz, 1997).

The outcome of bargaining, contra Becker, is that the partner with the higher rate of pay and/or more power spends more time working and less on child care and housework, which also decides the partners’ levels of market earnings. While the responsibility for domestic labor is constrained by the partners’ employment, net of this constraint, the higher earning partner uses (usually his) leverage to lower his contribution to domestic labor. The idea of bargaining makes sense only if it involves market income; otherwise idiosyncratic perceptions of individual utility, not to mention the personalities and attitudes of the two partners, could lead to almost any outcome. One could argue that rational choices involve gendered assumptions. A broader limitation of rational choice models is uncertainty about short- and long-term payoffs and difficulty of comparing long-term negative effects of labor force interruptions with short-term economic benefits of caring for a child at home. Still more complexity and variation results if we place a value on the quality of child care and the psychic costs and benefits to the caregiver.

The alternative to rational choice models is what is variously termed the gender, cultural, feminist, or sociological perspective. The argument is that the behavior of couples differs systematically from the strategy that would maximize their economic welfare, in the direction of traditional gender roles and to the disadvantage of female partners. On the structural side, the argument is that the labor market and state institutions that regulate employment, and also regulate and potentially supply or subsidize child care, create incentives favoring traditional gender roles, even for economically rational couples. On the individual side, the argument is that noneconomic aspects of culture favor men in bargaining with their female partners.

This perspective is shown in the work of Hochschild (1997), who argues that the distribution of men’s and women’s market and domestic work is not reducible to their comparative positions in the labor market. Rather, both normative gender roles and a legacy of state family and labor market policies preserve the gendered nature of household production. This gives rise to women’s “double-burden,” based on the evidence that mothers do not reduce the amount of time they spend on child care as they increase their hours of employment (Bianchi, 2000; Bryant & Zick, 1996; Hofferth & Sandberg, 2001; Sandberg & Hofferth, 2001, Sayer, Bianchi, & Robinson, 2004), even though time spent on child care tends to be lower among working mothers than nonemployed mothers (Bianchi, 2000; Noch & Kingston, 1988; Pleck, 1981; Robinson & Godbey, 1997; Sandberg & Hofferth, 2001). England and Folbre (2005, p. 742) take these findings to suggest “a special resistance to having men replace women in parenting.”

Further research demonstrates that the mother’s hours of employment do not significantly affect the time a father spends on child care (Bryant & Zick, 1996; Noch & Kingston, 1988; Pleck, 1997). Nevertheless, there is evidence that the time women spend on housework has decreased, women’s labor force participation has increased (Robinson & Godbey, 1997), and that total child care time is increasing (Bryant & Zick, 1996; Gauthier, Smeeding, & Furstenbern, 2004; Sandberg & Hoffert, 2001). These findings have been partly attributed to the increased social desirability of male parenting and of partners taking equal responsibility for child care, especially among dual-income couples (Press, 1998). Furthermore, increased rates of childlessness may contribute to more observed child care among those who are highly motivated to be parents (Gauthier et al., 2004) or a “hovering” parenting culture committed to reducing perceived “risks” of childhood (Sayer et al., 2004). An alternative to the “double-burden” argument, though leading to the same predicted outcomes, does away with the need for consensus on gendered norms of behavior. Instead, inequality arises from a maze of expectations and gendered ideas, characterized by the idea of “doing gender” (Berk, 1985; Bittman, England, Sayer, Folbre, & Matheson, 2003; Bittman & Pixley, 1997; Brines, 1994; Coltrane, 2000; West & Zimmerman, 1981).

On the structural side, a mass of scholarship describes the institutional mechanisms favoring traditional gender roles. One example is Ursel’s (1991) study of how the development of Canadian family policy contributed to separate spheres of male market work and female domestic work, protecting the breadwinning role of men. A variety of policies affect gender inequality in households and employment. For instance, public education and subsidized child care socialize the cost of child care (Folbre, 2004; Jacobs & Gornick, 2002; Joesch & Spiess, 2002), which could be expected to support alternatives to the traditional division of labor. A higher rate of marginal taxation on second wage earners, however, lowers female labor force participation and augments the supply of unpaid child care labor (Apps, 2002). For dual-earner couples, Stier and Mandel (2009) find that the presence of more child care facilities increases, while longer maternity leaves and part-time employment decrease, the female contribution to household income. In Switzerland, Stähli, LeGoff, Levy, and Widmer (2009) show that mother’s work and child care arrangements have little to do with their preferences than with structural factors, such as the family’s economic resources, available child care, and policy support for mothers’ labor force participation.

The theoretical perspectives describe tendencies in what is known to be considerable variation in couples’ employment and child care strategies (Beaujot & Ravanera, 2009). Similarities in perspectives exist. For instance, both predict that the demands of younger children and having more children lead families to adopt more specialized, less egalitarian roles, in part because of the cost and difficulty in obtaining infant child care and the cost of care for two or more preschool children.

The approaches differ in a number of respects, however. First, the gendered approach emphasizes, and there is copious empirical research to show, that even after accounting for human capital, women do more child care and are less likely to be employed than men. This is attributed both to gender differentials in power from socioeconomic differences and from culture (Bianchi, 2000; Hochschild & Machung, 1989; Robinson & Godbey, 1997).

Second, Hochschild demonstrates a gendered asymmetry in the relationship between partners’ individual attributes and their time use. Notably, the negative relationship between the time spent on paid employment and the time spent caring for children is stronger for men than for women, so that men are better able to trade-off employment and domestic labor, while women experience the double-burden. The presence of children, especially young children, increases the time spent on child care in a highly gender-specific manner, affecting women’s domestic labor far more than men’s (McFarlane, Beaujot, & Haddad, 2000; Presser, 1994; Shelton, 1992). Also, male partners’ wage rate much more strongly (positively) affects women’s child care time than the reverse (Blair & Lichter, 1991; Glass, 1998; Kamo, 1994; Presser, 1986). While women who work part-time mitigate the effect of the double-burden (Michelson, 1985), this is possible only when the lost income does not severely compromise the family financially. We therefore expect to find asymmetric effects of differences in the human capital on family strategy such that differences favoring men will have larger effects on the choice of family strategy. Still we expect that women with more education relative to their male partners will have more paid employment and do less child care given that additional human capital may realize labor force returns and contribute to bargaining power.

Third, the gender theoretical approach emphasizes the role of education in increasing the likelihood of an egalitarian division of labor between partners (Brines, 1993; Geerken & Gove, 1983; Haddad, 1994; Kalleberg & Rosenfeld, 1990; Presser, 1994) and that women with more education do less housework and child care (Brines, 1993; South & Spitz, 1994; Strober & Chan, 1998). Without this cultural emphasis, the rational choice perspective leads only to the prediction that partners with more similar human capital are more likely to share employment and child care more equally.

Rational choice and gender theory both suggest that when partners differ in age or education, the younger, less educated parent will do more child care and work fewer hours. But, whereas the rational choice perspective treats as equal the human capital effects of the two partners, gender theory predicts asymmetric effects of inequality in age and education, such that women with more human capital, either education or age, are less likely than men with the same characteristics to substitute employment for child care. That is, human capital has less of a return to mothers than to fathers.

We aim to investigate how couples’ human capital and the demand for parents’ time, created by the number of children within the household, affects the household strategy. In three respects rational choice and gender theory make quite different predictions that are central hypotheses in our investigation. First, after accounting for the human capital of both partners, gender theory predicts that women do much more child care. Second, gender theory, but not rational choice, predicts that the human capital of men has a greater effect on their contributions to employment and child care than the human capital of women—this would lead to asymmetric effects in families where the two partners differ in age or education. And, third, because human capital is a monotonic, if not necessarily linear, function of age, rational choice theory leads us to expect monotonic effects on the choice of family strategy; gender theory suggests that these effects are nonlinear and might not be monotonic, due to effects of higher education and the changing life course on how people think about and act out the division of labor.

Method and Data

For empirical research on time use, the natural choice is a survey employing exact time diaries. Since 2000, Merz (2009) counts more than 40 such studies internationally. Like most time use surveys, however, the Canadian GSS measures time use for only one person in a selected household. This allows comparisons between women and men (Marshall, 2006) but provides no information about joint patterns of work and child care between couples, though some national diary-based surveys gather data for both partners in a household, including Sweden (Hallberg & Klevmarken, 2003) and France (Bloemen & Stancanelli, 2008). Cruder measures of time use for all adults in a family are also available from some household panel studies, including the US Panel Study of Income Dynamics (Boca & Flinn, 2006) and the German Socio-Economic Panel (Merz, 2009).

Robinson (1985, 1999) shows that time use data collected using diary methods are superior to the retrospective or “stylized” measures of time use available from the Census and from some household panel studies. Also, there is evidence that parents tend to overestimate their time spent with children, especially in socially desirable activities such as reading aloud to children (Hofferth, 1999). While critical when investigating specific activities or producing precise estimates of time spent on child care, these measurement issues are less critical when researching the magnitude of the difference couples spend caring for children. Due to the ordinal categorization, the child care responsibility of couples who we describe as sharing child care “equally” are only approximately equal—the two responses are in the same general category.

The 2006 Canadian Census measures exact hours of employment and time spent on child care, as a primary activity, in six categories—0, 1 to 4, 5 to 14, 15 to 29, 30 to 59, and 60 hours or more—and additionally includes rich information on each family member’s age, earnings, occupations, and education. The Census sample is very large—one in every five Canadian households, 6.3 million persons in all. More than 100,000 households fit our criteria for household composition and the age of children, thereby allowing us to finely differentiate families according to the number and exact ages of children 5 years old and younger.

This analysis is confined to opposite-sex couples with at least one school child less than 6 years old (though it is not strictly true, we call them “preschool” children) and no older child. There is strong evidence that without a live-in relative or paid caregiver the demands of a young child strongly affect labor force participation, especially of women (P. E. Becker & Moen 1999). So the families we analyze have no older child or adult household member to help with child care and potentially affect the distribution of caregiving between partners.

To account for slight age asymmetry in the ages of partners in families with preschool children, we exclude couples where the mother was 50 or older or the father was 60 or older. Finally, we excluded families in which both partners reported less than 5 hours child care per week. We lack additional information, such as care provided by extended out of household family members that might explain such improbably low levels of child care in a family with no other adult, though there is some degree of measurement error.¹ Finally, we exclude households in which neither partner was employed for more than 15 hours per week. All analyses were weighted to yield unbiased population estimates.

Results

Child Care and Employment

Table 1 gives the combinations of mothers’ and fathers’ hours of child care, as a percentage of all families, along with row and column totals. It is no surprise that child care is highly gendered: 23.9% of mothers reported 30 to 59 hours of child care and 54.3% reported 60 hours or more, compared with figures of 19.9% and 15.0% for fathers, respectively. Despite the restriction to two-parent households with no other adult or older child to help with child care, 4.5% of the fathers of preschool children report that they do no child care at all, 7.6% report 1 to 4 hours of child care per week, and 25.0% report 5 to 14 hours; the corresponding figures for mothers are just 2.4%, 1.0%, and 5.3%. In 61.1% of families women do more child care than men, whereas in just 3.0% of families men do more child care, and in 35.9% of families child care is about equal.

Table 1.

Combinations of Mother’s and Father’s Child Care Hours (Percentage of All Families)

	Mother’s Hours/Week
Father’s Hours/Week	Zero	Less Than 4	5-9	15-29	30-59	60 or More	Total
Zero	2.2	0.2	0.2	0.2	0.4	1.4	4.5
Less than 4	0.0	0.5	1.0	1.0	1.3	3.6	7.6
5-9	0.0	0.1	3.4	4.0	5.4	12.0	25.0
15-29	0.0	0.1	0.4	6.7	6.6	14.1	27.9
30-59	0.0	0.0	0.2	0.7	9.4	9.5	19.9
60 or more	0.0	0.0	0.1	0.4	0.7	13.7	15.0
Total	2.4	1.0	5.3	13.1	23.9	54.3	100.0

Note: Unweighted n = 131,600; weighted n = 653,440.

Totals are affected by rounding error. Bold indicates approximately equal contributions of mother and father.

Source: 2006 Canadian Census, one in five sample.

The combinations of male and female market hours of paid employment are in Table 2, removing about 3% of the families in which neither partner does more than 4 hours of child care per week. Strikingly, 48.1% of mothers report no paid employment and 38.1% of couples have a traditional arrangement in which the father works 35 or more hours per week and the mother has no paid employment. In just over one quarter of families, 26.3%, women and men work about the same hours. In 28.6% of families both partners are employed for 35 or more hours per week. In families with one part-time and one full-time worker, the latter are predominantly men and by a margin of 10.1% to 0.8% (of all families), the father is employed for 35 hours or more per week and the mother for 1 to 24 hours rather than the reverse.

Table 2.

Combinations of Mother’s and Father’s Employment (Percentage of All Families)

	Mother’s Hours/Week
Father’s Hours/Week	Zero	Less Than 15	15-24	25-34	35-44	45-54	55 or More	Total
Zero	7.7	0.2	0.4	0.5	2.2	0.4	0.1	11.5
Less than 15	0.5	0.1	0.1	0.1	0.2	0.1	0.0	1.0
15-24	0.7	0.1	0.1	0.1	0.4	0.1	0.0	1.5
25-34	1.1	0.1	0.2	0.4	0.6	0.1	0.0	2.6
35-44	20.2	1.9	3.1	3.6	15.4	1.6	0.3	46.2
45-54	11.3	1.3	1.8	1.9	5.4	2.0	0.2	23.8
55 or more	6.6	0.9	1.1	1.0	2.4	0.7	0.6	13.3
Total	48.1	4.7	6.8	7.5	26.7	4.9	1.3	100.0

Note: Unweighted n = 127,690; weighted n = 633,860.

Totals are affected by rounding error. Bold indicates approximately equal contributions of mother and father.

Source: 2006 Canadian Census, one in five sample.

Ten Family Strategies for Child Care and Employment

The conceptual and methodological challenge is to reduce to manageable categories the more than 1,700 possible combinations of families’ child care time, the 6 × 6 matrix in Table 1, and employment, the 7 × 7 matrix in Table 2. This we accomplished in two steps. First, Table 3 shows the cross-classification of seven combinations of employment, denoted a to g, and eight combinations of child care, denoted 1 to 8. Child care is measured by the total amount of care and how it is shared, while employment is classified in four categories, none, 1 to 14, 15 to 34, and 35 or more hours per week. The 56 combinations of employment by child care in Table 3 were reduced to 10 “strategies.” We excluded 8.5% of all couples (the top row of Table 3), in which neither partner was employed for 15 hours per week or more, on the grounds that without external support or wealth this strategy is not economically sustainable.

Table 3.

Cross-Classification of Combination of Work Hours and Child Care Arrangement

		Child Care Arrangement
			2	3	4	5	6	7
		1	Equal Amounts		Mother Does More		Mother Does Much More		8
Combination of Work Hours		Father Does More	Total Is Low	Total Is High	Total Is Low	Total Is High	Total Is Low	Total Is High	Mother Does Far More	Total
a	Both zero or less than 15 hours	0.3	0.7	3.0	0.4	1.3	0.5	1.0	1.4	8.5
b	Mother much more	1.0	0.4	1.0	0.1	0.2	0.1	0.1	0.1	3.0
c	Mother somewhat more	0.4	0.3	0.9	0.1	0.2	0.1	0.1	0.1	2.3
d	Father somewhat more	0.0	0.1	0.4	0.1	0.4	0.1	0.4	0.5	2.0
e	Father full-time, mother not employed or less than 15 hours	0.3	1.2	4.1	1.5	7.8	3.3	10.3	13.9	42.3
f	Father full-time, mother high part-time	0.1	1.2	2.9	0.8	2.9	1.1	1.6	1.7	12.4
g	Both full-time	0.9	6.3	11.7	2.2	3.8	1.7	1.2	1.7	29.4
	Total	3.0	10.3	23.9	5.2	16.6	6.8	14.6	19.5	100.0

Source: 2006 Canadian Census, one in five sample.

When the mother is employed for much longer hours, in 35% of families the father does more child care, in 45% child care is shared about equally, and in 20% the mother does more child care.² In 87% of families where the father is employed for much longer hours, the mother does more child care, in 12% of those families child care is shared about equally, and in 0.7% the father does more child care. This remarkable asymmetry is consistent with the gender theory, but certainly not with traditional ideas of rational choice; nor does it seem plausible that income differences are so extreme that this could be the result of noncooperative bargaining. More ambiguous, as it must reflect the typically greater earning power of male partners, is the finding that families in which only the father is employed or he is employed for much longer hours are about eight times more common than families where the mother plays that role.

The 10 employment–child care strategies are shown in Table 4, which also indicates how the cells in Table 3 were aggregated to create the typology. The 10 strategies can be divided according to the partners’ relative contributions of child care. Three of the four most common family strategies involve a gendered division of child care and employment. Families in which the mother does more child care account for 4 of the 10 strategies:

Table 4.

Child Care and Employment Strategy by Number and Ages of Children (Percentage Distribution)

			Number of Children Under Age 1		One			Two or More		None
“Cells” From Table 3	Comparative Hours of Child Care	Employment	Number of Children Between 1 and 5	Total	None	One	Two or More	None	One or More	One	Two	Three or More
g2 g3	Equal	Both full-time		19.6	3.7	4.1	2.2	2.4	5.1	28.5	23.8	13.9
d2 d3 f2 f3		Father full-time, mother part-time		5.1	1.9	1.9	2.3	1.6	0.0	6.2	6.9	5.3
e2 e3		Father full-time, mother not employed		5.8	8.9	9.4	10.7	12.9	12.8	4.3	4.0	6.4
b2-c3		Mother more hours employment		2.8	1.3	1.1	1.0	—^a	—^a	4.0	2.8	2.6
g4-g8	Mother more	Both full-time		11.6	3.2	4.0	3.8	2.9	1.3	16.5	13.1	7.7
d4-d8 f4-f8		Father full-time, mother high part-time		10.5	6.4	6.2	6.8	4.8	3.4	11.3	14.2	12.0
e4-e8		Father full-time, mother not employed or employed less than 15 hours/week		40.1	72.1	71.3	70.7	71.3	74.8	23.2	30.8	48.3
b4-c8		Mother more hours employment		1.4	0.7	0.6	0.6	—^a	—^a	2.0	1.5	0.9
d1-g1	Father more	Equal or father more hours employment		1.4	0.7	0.5	1.0^b	0.8	0.0	2.0	1.3	1.1
b1 c1		Mother more hours employment		1.6	1.1	1.0	0.0	—^a	—^a	2.0	1.6	1.8
Total				100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0
Percentage of all families^c				100.0	15.9	11.7	1.9	0.3	0.2	43.7	24.4	1.9
Estimated number of families in the Canadian population^c				580,050	92,065	67,840	11,300	1,865	1,170	253,265	141,795	10,735

Less than 1%.

Estimated.

Estimates are rounded, as a result the total in the first column is not exactly equal to the component parts.

In 40.1% of all families the father is employed full-time, while the mother is not employed or works less than 15 hours per week—the classic “specialization” of roles

In 11.6% of families both partners are employed full-time; this strategy—the “double-burden” wherein the mother works as much as the father while doing more child care—is much less common than the traditional division of labor in the first strategy

In 10.5% of families the father is employed full-time and the mother is employed 15 to 34 hours per week

In 1.4% of families both partners are employed, with the mother working longer hours

The one third of families where child care is shared about equally pursue one of four alternative strategies:

19.6% of all families with both parents employed full-time

5.1% of families where the father is employed full-time and the mother part-time

5.8% of families where the father is employed full-time and the mother is not employed

2.8% of families with both partners employed and the mother working longer hours

Last, the 3.0% of families in which the father does more child care are divided between

1.4% of families where the parents are both employed full-time or the father works longer hours

1.6% of families in which the mother works longer hours

Gender specialization, with the mother doing more child care and the father working longer hours, is more than three times more common than the strategy resulting in the “double-burden,” whereby both parents are employed full-time but the mother does more child care.

Family Strategy and the Ages and Number of Preschool Children

Before examining the effects of the parents’ age and education on family strategy, we need to look at the impact of the number and ages of children, in Table 4—to determine whether it is appropriate to include all families in a single model. There is a remarkable difference between families with and without a child under the age of 1 year (an “infant”). In 72% of families with such a very young child the mother does more child care and is not employed or works less than 15 hours per week, while the father is employed for at least 35 hours—a figures that is largely unaffected by the presence of older preschool children. In comparison, in only 23% of families without a child under the age of 1 year does the mother do more child care and is not employed or works less than 15 hours per week, while the father is employed for at least 35 hours; and the corresponding figures are 31% and 48% for families with two or three children between 1 and 5 years of age and no younger child. This difference reflects the needs of infants, the high cost of infant child care, and government and employer policies providing support to their parents.

The radical difference between parents with a child under the age of 1 year and parents with older preschool children only identifies a transition made by most couples. Around the time their youngest child becomes 1 year old, about 45% of couples switch away from the strategy whereby the father is employed full-time and the mother is not employed or works for fewer than 15 hours per week. While most couples experience a period characterized by a very traditional division of labor, usually it last only a year or so.³

There is much more variation in the strategies of families without a child under the age of 1 year and one or more children between 1 and 5 years of age. In 28.5% of those families with one child, both parents are employed full-time and share child care equally, and the corresponding figures are 24% and 15% for families with two and with three or more preschool children. Depending on the number of children, only 2% to 5% of families with a child under the age of 1 year share child care equally and work full-time, suggesting that the demands of care at this time make a more equal household strategy more difficult to achieve.

Out of concern that the distinction between families with and without a child under the age of 1 year was not oversimplified, we also investigated whether the number of children in a family at each age between 1 and 5 years could also result in major discontinuities in the child care strategy and confirmed that the fundamental distinction is between families with and without a child under the age of 1 year.

Regression Methods

The multinomial regression in Table 5 shows the probability of adopting the nine other strategies, compared with the (“omitted”) strategy, of a more “traditional” strategy whereby the father is employed full-time and the mother does more child care and is not employed or is employed for less than 15 hours per week. In addition, the right-hand column gives results of a binary logistic regression comparing that omitted category with all other strategies. The latter is helpful, but redundant, since with 10 outcomes there are only nine degrees of freedom. The highly nonlinear effects revealed in the regression prevent our saying that the likelihood of any particular family strategy simply increases or decreases with age or with education. This summary identifies the most important patterns, given limitations of space, and focuses on the most common family strategies.

Table 5.

Multinomial Regression of Child Care/Employment Strategy on Education and Ages of Parents for Two-Parent Families With No Child Under Age 1 Year

		Multinomial Regression, Compared to Category at Right									Logistic Regression
	Child Care	Equal				Mother More			Father More		Mother More
N = 81,249	Employment	Both Full-Time	Father Full-Time, Mother Part-Time	Father Full-Time, Mother Not Employed	Mother More Hours Employment	Both Full-Time	Father Full-Time, Mother Part-Time	Mother More Hours Employment	Father More Hours Employment	Mother More Hours Employment	Father Full-Time, Mother Not Employed or Employed Less Than 15 hours/week
Number of children, compared to one child
2 children ages 1 to 5		−0.532	−0.189	−0.320	−0.598	−0.564	−0.087	−0.559	−0.483	−0.675	0.418
3 or more children ages 1 to 5		−1.482	−0.887	−0.338	−1.104	−1.514	−0.678	−1.694	−0.737	−1.267	1.127
Education
Both less than high school		−0.608	−0.737	0.091	−0.352	−0.593	−0.567	−0.672	−0.259	−0.286	0.485
Both high school graduates—The omitted category		—	—	—	—	—	—	—	—	—	—
Both college		0.465	0.183	−0.035	−0.004	0.282	0.211	0.291	−0.046	−0.170	−0.277
Both university		0.311	−0.098	−0.341	−0.288	−0.172	−0.002	0.101	0.173	−0.689	−0.037
Both graduate degree		0.184	−0.097	−0.238	0.051	−0.156	−0.003	0.240	−0.175	−0.203	−0.011
High school and not high school		−0.296	−0.253	−0.092	−0.120	−0.309	−0.215	−0.191	0.070	−0.155	0.218
College and high school or less		0.120	−0.032	−0.119	−0.038	0.147	0.068	0.153	0.046	−0.127	−0.077
University degree and less than college		0.076	−0.201	−0.349	−0.237	0.063	0.026	0.096	0.162	−0.439	−0.008
University degree and college		0.300	0.105	−0.251	0.054	0.066	0.064	0.130	0.128	−0.187	−0.138
Graduate degree and less than university degree		−0.125	−0.176	0.052	0.010	−0.179	−0.071	0.333	0.692	−0.091	0.036
Graduate degree and university degree		0.065	−0.266	−0.421	−0.038	−0.271	−0.008	−0.014	0.243	−0.616	0.082
Educational difference (positive if mother has more education)
High school and not high school		0.248	0.043	−0.073	0.514	0.286	0.243	0.181	0.418	0.641	−0.232
College and high school or less		0.367	0.228	0.059	0.477	0.456	0.262	0.235	0.431	0.640	−0.345
University degree and less than college		0.596	0.364	0.054	0.847	0.671	0.337	0.273	0.909	0.849	−0.532
University degree and college		0.322	0.036	0.065	0.371	0.310	0.053	0.281	0.497	0.402	−0.242
Graduate degree and less than university degree		0.902	0.576	0.205	0.887	0.700	0.296	0.535	0.849	0.574	−0.648
Graduate degree and university degree		0.355	0.308	0.189	0.243	0.242	0.137	0.260	0.680	0.229	−0.282
Mean age
Under 25		−0.428	−0.111	0.054	0.015	−0.414	−0.064	−0.303	0.411	0.140	0.198
25-29—The omitted category		—	—	—	—	—	—	—	—	—	—
30-34		0.305	0.036	−0.153	−0.124	0.267	0.189	0.155	0.155	0.110	−0.186
35-39		0.334	0.036	−0.242	0.035	0.312	0.185	0.113	0.378	0.147	−0.211
40-44		0.195	−0.123	−0.307	−0.016	0.227	0.154	0.213	0.549	0.419	−0.133
45 or more		0.166	0.114	−0.195	0.307	0.239	0.112	0.343	1.023	0.545	−0.196
Age difference
Mother 5 or more years older		−0.160	−0.063	0.041	0.283	−0.090	−0.140	0.004	0.242	0.079	0.076
Less than 5 years difference—The omitted category		—	—	—	—	—	—	—	—	—	—
Father 5-9 years older		−0.196	−0.114	0.103	0.051	−0.132	−0.102	−0.086	−0.083	0.166	0.113
Father 10-14 years older		−0.482	−0.229	0.245	−0.203	−0.383	−0.224	0.118	−0.239	0.113	0.284
Father 15 or more years older		−0.630	−0.301	0.413	0.231	−0.430	−0.293	0.081	0.130	0.058	0.295
Constant		−0.066	−1.216	−1.453	−1.760	−0.499	−0.816	−2.646	−2.917	−2.508	−1.573

Note: Coefficients in boldface significant at p > .05.

Source: 2006 Canadian Census, one in five sample.

The specification of the indicators of the two parents’ education and age captures effects of their sum and difference. Education is measured with 10 indicator variables comparing families in which both parents are high school graduates with families where neither parent is a high school graduate, both are college graduates, both are university graduates, and both have a graduate or professional degree, and with six categories of families where the two parents have different levels of education, without regard to which partner has more education. One of six categories, for example, is for families where one parent is a high school graduate and the other has not completed high school. Table 5 shows all the categories. For each of the six unequal combinations of education, in addition, “contrast” variables compare families where the mother has more education to those where the father has more education. These contrasts are coded +1 if the mother has more education, −1 if the father has more education, and 0 otherwise. Age is measured by their average age of the parents, in 5-year intervals and with four variables comparing couples whose ages are within 4 years to those with a bigger age difference.

Results: Regression Analysis of 10 Family Strategies

The first two rows of Table 5 show very substantial effects of the number of children on the household strategy. Compared with a family with one preschool child, the presence of a second child increases by 52% the likelihood (b = .418) that the family strategy has the father employed full-time and the mother doing more child care while not employed or working for less than 15 hours per week; and there is a corresponding, approximately 60%, decrease in the likelihood that both parents are employed full-time, with the child care shared equally or the mother doing more care (b = −.532 and −.564, respectively).

Families with three or more children between 1 and 5 years of age are strikingly more likely to “specialize.” Compared with families with one child, they are 3.1 times (b = 1.127) more likely to have a father employed full-time and the mother not employed or employed less than 15 hours per week and doing more child care, and there are equally large decreases in the strategies involving full-time employment of both partners with child care shared equally or the mother doing more (b = −1.482 and −1.514, respectively) and in strategies where the mother is employed for more hours than the father. The logic is that there are major economies of scale in a parent caring for two or more young children, but there is little or no discount purchasing care for more than one child outside the home.

The effect of education on family strategy is nonlinear, with the biggest difference between families where neither parent has finished high school and those with at least one high school graduate. Compared with families where both parents are high school graduates, families in which neither parent has completed high school are about 40% less likely to have both parents employed full-time, with the mother doing more child care or child care being shared equally and 62% more likely to have the father employed full-time and the mother doing more child care and not employed or employed less than 15 hours per week. This is consistent with the argument that families with the lowest level of human capital also have the most difficulty finding employment and paying for child care. Here it would be very interesting to have information on child care provided by family members and friends, through exchange with other parents and from low-cost, unlicensed caregivers.

The next largest difference is between families with two high school graduates and two college graduates. Families with two college graduates are 24% less likely to have the father employed full-time and the mother doing more child care and not employed or employed for less than 15 hours per week, while they are 59% more likely to choose the arrangement of having both parents employed full-time and the sharing child care about equally and 33% more likely to have both parents employed full-time with the mother providing more child care.

We might expect that still more educated parents, with university or graduate or professional degrees, would be still more likely to have both partners being employed full-time and less likely to have fathers employed while mothers do child care. But university graduate couples actually fall midway between high school and college graduate couples. For families where both parents are employed full-time, child care is most likely to be shared equally by couples with university degrees or with graduate or professional degrees. This comparison of couples with approximately equal hours of employment shows that higher education produces more egalitarian family arrangements, consistent with gender theory.

There are other interesting patterns, such as the finding that the unusual arrangement of the mother doing less child care than the father and working longer hours being most common for high school graduate couples and least common for couples with two university graduates. When the education of the mother and father are not the same, the education of both parents has approximately equal, additive effects on the family strategy. For example, the probability that a couple with one non–high school graduate and both have full-time jobs and share child care equally is about midway between the probabilities for a couple made up of two non–high school graduates and a couple with two high school graduates.

Now consider the six variables measuring whether the strategy adopted by a couple with different levels of education is affected by which partner has more schooling. For all nine strategies in the multinomial regression and in the logistic regression, each set of six coefficients has the same sign (with one exception, where the coefficient is small and not significant). This indicates that the effect of inequality between partners operates in the same manner for all the six combinations of education, though the coefficients are largest in absolute value for the two categories denoting the greatest educational disparity, couples with a graduate degree and less than a university degree and couples with a university degree and less than college.

In the logistic regression all six coefficients are negative, while in the multinomial regression 53 of the 54 coefficients are positive. So when the mother has more education, the couple is less likely to adopt the traditional strategy whereby the father is employed full-time and the mother does more child care and is not employed or employed for less than 15 hours per week. If the mother has more education all nine other scenarios are more probable.

For example, consider the probability that couple with a graduate degree and an undergraduate university degree both work full-time and share child care about equally. The coefficient for the main effect is just 0.065, so on average this highly educated couple is about equally likely to adopt this strategy, compared with the “reference group” of two parents, both with high school diplomas. The coefficient for the gender contrast, however, is +.355, indicating a 40% greater probability of equal child care if the mother is the one with the graduate degree and the father has the undergraduate degree.

Like education, age has nonlinear effects on family strategy. Compared with couples whose average age is 25 to 29 years, younger couples are 35% less likely to be working full-time and sharing child care about equally. Couples whose mean ages are 30 to 34, 35 to 39, 40 to 44, and 45 years or more are more likely to be employed full-time and share child care about equally, by 36%, 40%, 22%, and 18%, respectively. The figures are very similar for the strategy where both parents work full-time and the mother does more child care. The strategy whereby the father is employed full-time and the mother does more child care and is not employed or works less than 15 hours per week exhibits the opposite pattern. It is most common among the youngest parents, less common for the 25 to 29 years age group and least common for couples whose average age is 30 years or more, but there is little variation beyond average age of 30 years.

Families in which the mother is older by 5 years or more are similar to those with partners within 4 years of age, except they are about 25% more likely to choose the strategies in which there is equal child care and the mother is employed more hours and where the father does more child care and is employed more hours, and about 20% less likely to have both parents employed full-time and child care shared equally.

Compared with couples about the same age, fathers who are 5 to 9, 10 to 14, and at least 15 years older than their partners are, respectively, 11%, 33%, and 34% more likely to work full-time while the mother does more child care and is not employed or works less than 15 hours per week. Compared with couples within 4 years of age, the strategy whereby both parents are employed full-time and share child care about equally is 15% less likely to be chosen if the mother is 5 or more years older than her partner, and 18%, 32%, and 47% less likely if the father is 5 to 9, 10 to 14, and 15 or more years older, respectively. Interestingly, parents whose ages differ by at least 5 years are more likely to share child care about equally, when the father is employed full-time and the mother is not employed or works less than 15 hours a week, and those couples are less likely to have both parents employed full-time while the mother does more child care.

There are relatively linear effects of the age difference between the father and mother that involve a lower likelihood of the two strategies where both partners are employed full-time, with equal child care or the mother doing more care. Also decreased are the strategies whereby the mother does more child care while the father is employed full-time and mother employed higher part-time (15-34) hours per week and the strategy of equal child care with the father employed full-time and the mother part-time.

Parents’ ages have complex and nonlinear effects on family strategy and for the most common strategies the effect of the difference in partners’ ages is as big as the main effects of age. Up to about age 35, there is a fairly linear increase in the proportion of families where both parents are employed full-time and a corresponding decrease in the proportion where the father is employed while the mother does more child care and is not employed or works less than 15 hours per week. Beyond age 35, however, there is no further change or the pattern reverses itself. Because the changes do not correspond to the monotonic, presumably fairly linear increase of human capital with age, these effects cannot be understood only in terms of human capital, which does increase monotonically with age.

Discussion and Conclusions

To a striking degree our analysis reveals a traditional gendered division of labor. In three out of five families with a preschool child, the mother does more child care than the father; in just a few percent of families the father does more child care; and one third of couples share child care about equally. In about four fifths of families where the mother does more child care, she works fewer hours than her partner or is not employed. Even if these represent equitable arrangements in the short term, are consensual, and optimize the family’s financial well-being, women disproportionately bear the long-term, negative consequence of career interruptions (Budig & England, 2001; Shelton, 1992; Waldfogel, 1997).

Our findings are consistent with the interesting qualitative work by Moen and Sweet (2003) and others (Arendell, 2001; Forsberg, 2009; Hyman, Scholarios, & Baldry, 2005) and demonstrate the value of identifying household strategies. Work and child care arrangement are shaped by family contexts, the presence and age of children, and the gains to marriage that each member of a couple can realize in the labor market. These strategies are specific to a couple’s family circumstance, human capital, and stage in the life course. While human capital, especially education (Brines, 1993; Geerken & Gove, 1983; Haddad, 1994; Kalleberg & Rosenfeld, 1990; Presser, 1994), plays an important role in how child care is shared, the highly nonlinear effects we observe suggest that more education has a noneconomic positive effect on the likelihood of more equal sharing of child care, and the same is true for age. At the same time, in substantial numbers of families, supporting the work of Hochschild and Machung (1989), mothers experience the “double burden” of being employed full-time and doing more child care.

Interestingly, given the sample size and demographic detail in the census, we found that any difference in age between partners results in a less egalitarian arrangement, generally, even when the mother was older. Consistent with gender theory, however, an equal household strategy with the mother engaged in more market work is more likely when the mother is older than her male partner, thereby demonstrating the potential returns of greater female labor force participation and human capital, in the form of age. Previous research (Hank & Jürges, 2007) did not find an effect for differences in couples’ ages; however, this work investigated the division of household labor among a more homogeneous sample of older couples, while other work has found that the greater the age disparity between couples, the less likely domestic work is shared more equally (Bozon, 1991; Presser, 1994). Consistent with the noncooperative bargaining model, the older partner typically holds a strategic advantage, though our findings suggest greater returns when male partners are the senior within couples.

Child care arrangements are strongly affected by the needs of children, as measured by the number and ages of children in the household. Though the number of children is important, the key distinction is between families with a child under the age of 1 year and those with only older children. In roughly 70% of those families, the mother does more child care and is not employed at all or works less than 15 hours per week. We note that the federal Employment Insurance program and many employers support 1 year of maternity leave, which helps explain the large consequence of a child under the age of 1 year on couples’ household strategy.

The overwhelmingly greater responsibility of women for the care of young children and asymmetry of men’s “specialization” in paid employment is not compatible with the idea that the way that the families of young children live their lives is the product of “gender-neutral” and consensual agreements between parents. Even if, in some way, we could show parental strategies were entirely the product of rational choice, the gender differentiation that we observe describes a society with a profoundly gendered vision of the care of young children. The other two predictions of gender theory were also borne out. We found asymmetric effects on family strategy of differences in the two parents’ ages and education—so the impact of having more education and being older is much greater if the father is better educated or older. Also, we found highly nonlinear effects of education on strategy, and particularly that university graduates were distinctly more egalitarian in dividing child care. Consequently, we strongly favor gender theory, because it provides a better explanation of the nonlinearity, gendered asymmetry, and overwhelmingly greater responsibility for child care we observe that is not reducible to utility maximization.

Our aim in this article is to present an analysis of the key household strategies of the parents of young children, making innovative use the Canadian Census. For technical reasons and on methodological grounds, we acknowledge the limitations of our empirical analysis as a test of alternative theories: the data are cross-sectional, we do not measure attitudes and values, and we are also missing some structural variables, such as the cost of child care and opportunities in the local labor market. Nevertheless, for the parents of young children, our typology conceptualizes the key intersection of gender, work, and care while demonstrating the value of studying household strategies, thereby yielding insight into both labor force involvement of mothers and gender equality.

Footnotes

Authors’ Note

This article was jointly authored.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Apps

(2002). Gender, time use, and models of the household (Unpublished manuscript). Faculty of Law, University of Sydney, Sydney, Australia.

Arendell

(2001). The new care work of middle class mothers: Managing childrearing, employment, and time. In Daly

K. J.

(Ed.), Minding the time in family experience: Emerging perspectives (pp. 163-204). Amsterdam, Netherlands: JAI Press.

Beaujot

Ravanera

(2009). Models of earning and caring: Determinants of the division of work. Canadian Studies in Population, 35(1), 74-101.

Becker

G. S.

(1981). A treatise on the family (expanded edition, 1991). Cambridge, MA: Harvard University Press.

Becker

G. S.

(1985). Human capital, effort, and the sexual division of labour. Journal of Labour Economics, 3, 33-58.

Becker

P. E.

Moen

(1999). Scaling back: Dual-earner couples work-family strategies. Journal of Marriage and the Family, 61, 995-1007.

Berk

S. F.

(1985). The gender factory: The apportionment of work in American households. New York, NY: Plenum.

Bianchi

(2000). Maternal employment and time with children: Dramatic change or surprising continuity? Demography, 37, 401-414.

Bittman

England

Sayer

Folbre

Matheson

(2003). When does gender trump money? Bargaining and time in household work. American Journal of Sociology, 109, 186-214.

10.

Bittman

Pixley

(1997). The double life of the family: Myth, hope, and experience. Sydney, Australia: Allen & Unwin.

11.

Blair

S. L.

Lichter

(1991). Measuring the division of household labor. Journal of Family Issues, 12, 91-113.

12.

Bloemen

H. G.

Stancanelli

E. G. F.

(2008). How do parents allocate time? The effects of wages and income (Discussion Paper No. 3679). Bonn, Germany: Institute for the Study of Labour.

13.

Boca

Flinn

C. J.

(2006). Rationalizing child-support decisions. American Economic Review, 85, 1241-1262.

14.

Bozon

(1991). Women and the age gap between spouses: An accepted domination? Population: An English Selection, 3, 113-148.

15.

Brines

(1993). The exchange value of housework. Rational Society, 5, 302-340.

16.

Brines

(1994). Economic dependency, gender, and the division of labor at home. American Journal of Sociology, 100, 652-688.

17.

Bryant

W. K.

Zick

C. D.

(1996). Are we investing less in the next generation? Historical trends in time spent caring for children. Journal of Family and Economic Issues, 17, 385-392.

18.

Budig

England

(2001). The wage penalty for motherhood. American Sociological Review, 66, 204-225.

19.

Coltrane

(2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family, 62, 1208-1233.

20.

Crow

(1989). The use of the concept of “strategy” in recent sociological literature. Sociology, 23, 1-24.

21.

England

Folbre

(2005). Gender and economic sociology. In Smelser

N. J.

Swedborg

(Eds.), The handbook of economic sociology (2nd ed., pp. 727-749). Princeton, NJ: Princeton University Press.

22.

Folbre

(2004). A theory of the misallocation of time. In Folbre

Bittman

(Eds.), Family time: The social organization of care (pp. 7-24). London, England: Routledge.

23.

Forsberg

(2009). Managing time and childcare in dual-earner families: Unforeseen consequences. Acta Sociologica, 52, 162-176.

24.

Gauthier

A. H.

Smeeding

T. M.

Furstenbern

F. F.

Jr. (2004). Are parents investing less time in children? Trends in selected industrialized countries. Population and Development Review, 30, 647-671.

25.

Geerken

Gove

(1983). At home and at work. Beverly Hills, CA: Sage.

26.

Glass

(1998). Gender liberation, economic squeeze, or fear of strangers: Why fathers provide infant care in dual-earner families. Journal of Marriage and the Family, 60, 821-834.

27.

Haddad

(1994). Men’s contribution to family work: A re-examination of “time availability.” International Journal of Sociology of the Family, 24, 87-111.

28.

Hallberg

Klevmarken

(2003). Time for children: A study of parent’s time allocation. Journal of Population Economics, 16, 205-226.

29.

Hank

Jürges

(2007). Gender and the division of household labor in older couples: A European perspective. Journal of Family Issues, 28, 399-421.

30.

Hochschild

A. R.

(1997). The time bind: When work becomes home and home becomes work. New York, NY: Henry Holt.

31.

Hochschild

A. R.

Machung

(1989). The second shift: Working parents and the revolution at home. New York, NY: Viking.

32.

Hofferth

(1999). Family reading to young children: Social desirability and cultural biases in reporting. Paper presented at the Workshop on Measurement of and Research on Time Use, Committee on National Statistics, National Research Council, Washington, DC.

33.

Hofferth

Sandberg

(2001). How American children spend their time. Journal of Marriage and Family, 63, 295-308.

34.

Hyman

Scholarios

Baldry

(2005). Getting on or getting by? Employee flexibility and coping strategies for home and work. Work, Employment and Society, 19, 705-725.

35.

Jacobs

J. A.

Gornick

J. C.

(2002). Hours of paid work in dual-earner couples: The U.S. in cross-national perspectives. Sociological Focus, 35, 169-187.

36.

Joesch

J. M.

Spiess

C. K.

(2002). European mothers’ time spent looking after children: Differences and similarities across nine countries (EPAG Working Paper No. 31). Essex, England: European Panel Analysis Group.

37.

Kalleberg

A. L.

Rosenfeld

R. A.

(1990). Work in the family and in the labor market: A cross-national, reciprocal analysis. Journal of Marriage and the Family, 52, 331-346.

38.

Kamo

(1994). Division of household work in the United States and Japan. Journal of Family Issues, 15, 348-378.

39.

Katz

(1997). The intra-household economics of voice and exit. Feminist Economics, 3(3), 25-46.

40.

Lundberg

Pollak

R. A.

(1994). Noncooperative bargaining models of marriage. American Economic Review, 84, 132-137.

41.

Lundberg

Pollak

R. A.

(1996). Bargaining and distribution in marriage. Journal of Economic Perspectives, 10(4), 139-158.

42.

Marshall

(2006). Converging gender roles. Perspectives on Labour and Income, 7(7), 5-17.

43.

McFarlane

Beaujot

Haddad

(2000). Time constraints and relative resources as determinants of the sexual division of domestic work. Canadian Journal of Sociology, 25(1), 61-82.

44.

Merz

(2009, August). Time use and time budgets: Improvements, future challenges and recommendations (IZA Discussion Paper Series No. 4358). Bonn, Germany: Institute for the Study of Labor.

45.

Michelson

(1985). From sun to sun: Daily obligations and community structure in the lives of employed women and their families. Totowa, NJ: Rowman & Allenheld.

46.

Moen

Sweet

(2003). Time clocks: Work hour strategies. In Moen

(Ed.), It’s about time: Couples and careers (pp. 17-34). Ithaca, NY: Cornell University Press.

47.

Noch

Kingston

(1988). Time with children: The impact of couples’ work time commitments. Social Forces, 67, 59-85.

48.

Pleck

H. J.

(1981). Wives’ employment, role demands and adjustment: Final report. Wellesley, MA: Wellesley College Center for Research on Women.

49.

Pleck

H. J.

(1997). Parental involvement: Levels, sources, and consequences. In Lamb

M. E.

(Ed.), The role of father in child development (3rd ed., pp. 66-103). New York, NY: John Wiley.

50.

Pollak

(2003). Gary Becker’s contributions to family and household economics. Review of Economics of the Household, 1(1), 111-141.

51.

Press

(1998). Wives’ and husbands’ housework reporting: Gender, class, and social desirability. Gender and Society, 12, 188-218.

52.

Presser

(1986). Shift work among American women and child care. Journal of Marriage and the Family, 48, 551-563.

53.

Presser

(1994). Employment schedules among dual-earner spouses and the division of household labor by gender. American Sociological Review, 59, 348-364.

54.

Robinson

J. P.

(1985). The validity and reliability of diaries versus alternative time-use measures. In Juster

R. T.

Stafford

F. P.

(Eds.), Time, goods, and well-being (pp. 33-62). Ann Arbor: University of Michigan, Institute for Social Research.

55.

Robinson

J. P.

(1999). The time-diary method: Structures and uses. In Pentland

W. E.

Harvey

Lawton

M. P.

(Eds.), Time-use research in the social sciences (pp. 47-89). New York, NY: McColl, Kluwer Academic/Plenum.

56.

Robinson

J. P.

Godbey

(1997). Time for life: The surprising ways Americans use their time. University Park: Pennsylvania State University Press.

57.

Sandberg

J. F.

Hoffert

S. L.

(2001). Changes in children’s time-use with parents: United States, 1981-1997. Demography, 38, 423-436.

58.

Sayer

L. C.

Bianchi

S. M.

Robinson

J. P.

(2004). Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American Journal of Sociology, 110, 1-43.

59.

Schultz

T. P.

(1990). Testing the neoclassical model of family labor supply and fertility. Journal of Human Resources, 25, 599-634.

60.

Shelton

B. A.

(1992). Women, men and time: Gender differences in paid work, housework, and leisure. Westport, CT: Greenwood Press.

61.

South

S. J.

Spitz

(1994). Housework in marital and non-marital households. American Sociological Review, 59, 327-347.

62.

Stähli

M. E.

LeGoff

Levy

Widmer

(2009). Wishes or constraints? Mothers’ labour force participation and its motivation in Switzerland. European Sociological Review, 25, 333-348.

63.

Stier

Mandel

(2009). Inequality in the family: The institutional aspects of women’s earning contribution. Social Science Research, 38, 594-608.

64.

Strober

M. H.

Chan

A. M. K.

(1998). Husbands, wives, and housework: Graduates of Stanford and Tokyo Universities. Feminist Economics, 4(3), 97-127.

65.

Ursel

(1991). Private lives public policy: 100 Years of state intervention in the family. Toronto, Ontario, Canada: Women’s Press.

66.

Waldfogel

(1997). The effects of children on women’s wages. American Sociological Review, 62, 209-217.

67.

Wallace

(2002). Household strategies: Their conceptual relevance and analytical scope in social research. Sociology, 36, 275-292.

68.

West

Zimmerman

D. H.

(1981). Doing gender. Gender and Society, 1, 125-151.