Division of household and child-care tasks within heterosexual-parent, gay-father, and lesbian-mother families. A comparison during the first year of parenthood
Master thesis Preventive Youth care & Parenting Department of Child Development and Education
University of Amsterdam Marijke Huijzer-Engbrenghof - 10901493 First evaluator: Dr. L. van Rijn-van Gelderen Second evaluator: Prof. dr. G.J. Overbeek Amsterdam, June 2018
Abstract
The division of household and child-care tasks is a central aspect in the life of a couple who jointly rear their child(ren) but is not often studied among gay and lesbian parents with biological related children. Differences in division of tasks between heterosexual parents (n = 22), gay-fathers, (n = 20) and lesbian-mothers (n = 38) were studied. When the children were 4 and 12 months old, parents were queried about their division of household and child-care tasks. During the first period of parenthood, all parents divided household tasks quite equally, but not child-care tasks. Primary caregivers across all family types performed more child-care
tasks than the secondary caregivers, with heterosexual families having the most skewed division. After a year, primary caregivers from heterosexual-parent families still performed
more child-care tasks than their spouses. The same was true for lesbian-mother families, however their division became more egalitarian. Gay-fathers shared child-care tasks equally
at this time, but not household tasks. Finally, biological relatedness to the child did not influence the division of tasks. There can be concluded that differences between
heterosexual-parent, gay-father, and lesbian-mother families do exist. Nuclear father-mother families should therefore not be seen as representative for all family types.
Division of household and child-care tasks within heterosexual, gay-father, and lesbian-mother families
Families exist in many shapes and sizes which can be seen in the wide diversity of family forms nowadays (Furstenberg, 2014). Besides the traditional nuclear father-mother families, there are also single parent, cohabiting, and stepparent families (Cancian, Meyer, & Cook, 2011). Furthermore, there is an increasing number of children who are born and raised in same-sex families (Goldberg & Gatrell, 2014). For example, in 2002 15% of lesbian couples in the Netherlands had children under the age of eighteen. In 2010 this increased to 20% (Bos & van Gelderen, 2010). For gay couples these numbers are less clear: the Dutch Centraal Bureau voor de Statistiek (CBS) (Central Statistical Office) reported that in the Netherlands 31.000 gay-couples were living together in 2009, of which 3% had children. It is unclear whether these children were born within the gay-relationship, or in former
heterosexual relationships (Bos & van Gelderen, 2010). This diversity in family forms has received a lot of attention from researchers in their attempt to understand whether children are faring equally well across these families (e.g. Golombok, 2017).
Within this field of research an area of interest is coparenting. Coparenting is the shared activity of adults who are responsible for the child they are rearing. One aspect of coparenting is the division of both household and child-care tasks (Feinberg, 2003; Margolin, Gordis, & John, 2001; McHale & Irace, 2011). The division of household and child-care tasks and the satisfaction parents experience as a result are important predictors for child outcomes. Studies in heterosexual couples with children have found that children fare better when both parents are equally involved in household tasks, but particularly in the child-rearing
responsibilities (Johnson & O'Conner, 2002). For example, an egalitarian approach to child rearing is related to greater academic achievements for girls (Updegraff, McHale, & Crouter, 1996). Satisfaction with regard to the division of tasks also influences child outcomes. If
parents report less satisfaction about the division of tasks, they also report more externalizing problems with their child (Chan, Brooks, Raboy, & Patterson, 1998; Farr & Patterson, 2013). This result was found in heterosexual-parent families as well as in gay-father and lesbian-mother families (Farr & Patterson, 2013). When expectations regarding division of tasks are not met, it can lead to an increase in parental stress, which may interfere with warm and sensitive interaction towards the child (Feinberg, 2003). Parents themselves also fare better with an egalitarian division. If child-care tasks are equally divided between both parents, there is more satisfaction regarding the sexual relationship and a higher quality of the marital relationship (Carlson, Hanson & Fitzroy, 2016)
Most knowledge about division of tasks within families is obtained via research among heterosexual couples and their biological children (McHale & Irace, 2011). There are some studies within lesbian-mother families (e.g. Goldberg & Perry-Jenkins, 2007) and also some studies in which lesbian-mother families are compared with heterosexual families (e.g. Chan et al., 1998; Farr & Patterson, 2013). Rather few studies are held within gay-father families (Patterson & Farr, 2011). Studies with gay-father and lesbian-mother families mainly exist of parents with older children and, for gay-father families, adoptive children (Sumontha, Farr, & Patterson, 2013). Apart from the study with lesbian-mother families (Goldberg & Perry-Jenking, 2007), no additional studies have been found with a focus on the first year of parenthood. This is surprising, seeing that this is the timeframe where most transitions in parenthood are made (Durtschi, Soloski, & Kimmes, 2017), and coparent relationships are developed (Van Egeren, 2004). The increase in children who are born and raised within gay-father and lesbian-mother families and the legislation that becomes more and more favorable for these families (e.g. marriage, adoption, surrogacy) has led scholars, politicians, and citizens alike to accept that gay-father and lesbian-mother families are a viable new family form (Keuzenkamp, 2010). Therefore, this study will focus on division of household and
child-care tasks during the first year of parenthood, by comparing heterosexual parent, gay-father, and lesbian-mother families.
Up to date, there are various ideas and theories about how parents divide tasks. According to the gender theory, women have traditionally been defined by their unpaid familial role, whereas men have been defined by their status in the paid labor market.
Through this socialization, mothers are more predisposed to embody nurturing and caregiver roles, resulting in women spending more time in taking care of their children when compared to their male partners. In contrast, fathers may not incorporate the caregiver role into their self-concept to the same extent as mothers do (Hall, Walker, & Acock, 1995), while some fathers also see their involvement in paid employment as an important contribution to the caregiving of the child (Chan et al., 1998). This specialization of labor is the most pronounced in families in which the wife/mother does not participate in the paid labor market and the least in families where both partners work fulltime. Even though there is an increase in the
participation of women in the paid labor market and the specialized division of tasks
decreases to some extent, women still perform more household and child-care tasks compared to their husbands (Greenstein, 2000).
Besides this process of gender socialization, gender ideologies of both partners play a role in how couples divide tasks. For example, two men can have different ideas about what being male implies. One can believe that household labor is women’s work, whereas another may feel that being male means doing an equal share of the household tasks. Within marital and family roles, partners tend to behave in ways that will validate their identity as male or female, to display the visible aspects of their own gender ideologies (Greenstein, 2000). More progressive beliefs about the fathers' role, for instance, influence the fathers' involvement in child rearing. Interestingly it is the mothers' beliefs regarding family roles that appear to be of great influence (Schoppe-Sullivan, Brown, Cannon, Mangelsdorf, &, Sokolowski, 2008). If
she believes that the father should be directly involved in the day-to-day care, it is more likely that she will facilitate father involvement, which may help foster a mutually supportive coparenting relationship (Schoppe-Sullivan & Mangelsdorf, 2013).
In gay and lesbian couples there is one main characteristic that distinguishes these couples from heterosexual couples: there is no gender difference between the two partners. This eliminates the option of any gender hierarchy within the relationship. Hence, in a relationship without traditional roles and less traditional ideologies, the gender theory suggests that both partners are expected to prefer a more equal division of household and child-care tasks. Several studies indeed see this greater equity within gay-father and lesbian-mother families. Lesbian-lesbian-mother families report dividing tasks more evenly compared to heterosexual families. Lesbian-mothers strive to uphold an even division, whereas
heterosexual mothers ideally want to do more than half and heterosexual fathers less than half (Patterson, Sutfin, & Fulcher, 2004). Even though the amount of research on gay-father families is considerably less, the available data also reports more equal division of tasks among gay-father families (Patterson & Farr, 2011). Gay-fathers tend to share child-care tasks evenly, but not necessarily household tasks (Johnson & O’Connor, 2002).
An aspect that also needs to be taken into account when looking at the division of household and child-care tasks, is the biological relatedness to the child. In heterosexual families, both parents usually are biologically related to their child (there are of course also families in which this is not the case, but that subject is beyond the scope of this research). Even though both parents are biologically related to the child, there is still an important difference: the mother is the one who carries the pregnancy and gives birth to the child. Becker (as cited in Goldberg & Perry-Jenkins, 2007) argues that “women . . . spend much time and energy caring for their children because they want their heavy biological investment in production to be worthwhile” (p. 23). Therefore, heterosexual couples may have a
relatively nonspecialized division before becoming parents, but once the child is born, the mother who gave birth to the child becomes most involved in feeding and caring for the infant. This can lead to a shift to a more traditional gender specialized division of labor (Chan et al., 1998; Farr & Patterson, 2013). In lesbian-mother families this can be of influence too, seeing that only one of the mothers carries the pregnancy and gives birth to the child and is usually also the parent with the biological connection. She will, therefore, have a stronger connection to their child and will generally be the one who invests most time and effort in taking care of the child (Goldberg & Perry-Jenkins, 2007). Data from empirical research indeed shows that within lesbian-mother couples, the mother who gave birth to the child performs more child-care tasks compared to her partner, but no difference was found for household tasks (Goldberg & Perry-Jenkins, 2007). For gay-father families, there appears to be no difference in the amount of household and child-care tasks when biological and non-biological gay-fathers are compared to each other (Tornello, Kruczkowksi, & Patterson, 2015).
Despite all the above, it is still often expected that gay and lesbian couples form the same traditional, gendered patterns as heterosexual couples do by placing one member in the “father role” and the other in the “mother role” (Patterson & Farr, 2011). By comparing three different family types within one study, unique insights can be gathered. These insights will contribute to the much-needed knowledge about gay-father and lesbian-mother families. There are, for instance, several parenting programs, such as Family Foundations (Feinberg & Kan, 2008) and Directed Co-parenting Intervention (DCI) (Garber, 2004), in which
coparenting skills are an important component. But none of these programs are developed with knowledge about coparent dynamics within lesbian-mother and gay-father families. Therefore, more research is needed on coparenting within different family forms, in particular on whether coparenting dynamics are influenced by factors such as parent gender and sexual
orientation. Current study
The aim of this study was to investigate whether there are differences in the way heterosexual-parent, gay-father, and lesbian-mother families divide household and child-care tasks during the first year of parenthood. To accomplish this aim, four research questions were formulated. Firstly, are there differences between heterosexual-parent, gay-father, and lesbian-mother families in the division of household and child-care tasks when the children are 4 months old? Secondly, are there differences between heterosexual-parent, gay-father, and lesbian-mother families in the division of household and child-care tasks when the children are 12 months old? Thirdly, are there changes in the division of household and child-care tasks during this first year of parenthood? Finally, does biological relatedness to the child influence the division of tasks? With those four questions we strive to unveil the development in the division of household and child-care tasks over the timespan of the first year of
parenthood and the biological relatedness as a factor in task division.
There are two main hypotheses. The first hypothesis is that the primary caregiver (the parent who is most involved on a day-to-day basis with the child) within a heterosexual family will perform more child-care tasks compared to the primary caregiver within gay-father and lesbian-mother families. This difference is expected to be most pronounced between heterosexual families and gay-father families, due to several factors. First of all, maternity leave is assumed to play a role in the division of tasks. Heterosexual mothers will give birth to their child and receive maternity leave to recover and nurture the child. The heterosexual mother will therefore automatically spend more time at home with her child, at least for the first three months, and will perform a great deal of household and child-care tasks (Rehel, 2014). In gay-father families, neither men will have maternity leave. Therefore, the fathers have to consider the division of household and child-care tasks before they receive
their child. Expected is that gay-father families will strive for an egalitarian division with regard to child-care tasks, but not so much with regard to household tasks (Johnshon & O’Connor, 2002). Second of all, the parent who is biologically related to the child is already attached to the child, simply by virtue of their biological connection and is, therefore, more likely to feel he/she is the primary caregiver. It is thus expected that the parent with the biological relatedness to the child will perform more child-care tasks (Goldberg & Perry-Jenkins, 2007). Third of all, within heterosexual families, there tend to be more set and traditional ideologies about what appropriate tasks are for who and how much of these tasks need to be performed (Greenstein, 2000). Seeing that gay- and lesbian couples are already ‘breaking' these traditional views, it is plausible that their ideologies about who should do what is also different. Based on earlier research it is expected that there will be no differences with regard to the division of household tasks (Patterson et al., 2004).
The second hypothesis is that there will be less fluctuation within heterosexual families than within gay-father and lesbian-mother families when comparing the division of tasks during the first year of parenthood. This is expected because heterosexual families are more likely to hold on to traditional ideologies when it comes to paid labor and household labor (Greenstein, 2000) and gay- and lesbian couples are more likely to make sure the division of tasks is equal (Farr & Patterson, 2013). This fluctuation is expected to be the largest within lesbian-mother families. Usually, one of the mothers gives birth to the child and, like in heterosexual families, receives maternity leave and will, therefore, perform the most child-care tasks during the first period of parenthood. Because gay- and lesbian couples are more likely to make sure division of tasks are equal, it is expected that after this first period, the division will shift from less equal to more equal. In gay-father and heterosexual-parent families it is expected to stay quite stable once the tasks are divided during this first period.
Methods & Materials Participants
This study was part of an international research project. Researchers from the
Netherlands, United Kingdom, and France recruited 41 heterosexual families, 38 gay-father families and 61 lesbian-mother families (N = 140 families). Unfortunately, due to restrictions from French research institutes (no permission for master students to use their data),
participants from this country had to be removed from the dataset. This led to an analytic sample of 22 heterosexual families, 20 gay-father families, and 38 lesbian-mother families (N = 80 families).
Ten percent of the families had twins and 50% of the 88 children were female. The age of the parents ranged from 22 to 52 years old (Mage = 34.8, SD = 5.07). The
lesbian-mothers were significantly younger compared to gay-father and heterosexual families, F(2,77) = 20.11, p < .001. The duration of the relationship between the parents varied between 2 and 21 years (M = 8.01, SD = 3.89). The gay-fathers and lesbian-mothers differed significantly from each other: lesbian-mothers had a longer relationship compared to the gay-fathers, F(2,77) = 9.03, p < .001. Almost 75% of the parents were married or in civil partnership. The parents were mainly White (96.2%). Thirty-seven families (46%) lived in medium-sized cities, 25 families (31%) in small cities and 13 families (16%) in large cities. The remaining 5 families (6%) lived in rural areas. Most families (80%) had an annual household income of more than 42,365 US dollars. Table 1 presents demographics for each family type.
For this study, it was important to determine which of the parents was the primary caregiver and which was the secondary caregiver. The primary caregiver being the most involved parent on a day-to-day basis with the child. For this identification six items on the “Who does what” instrument (Cowan & Cowan, 1990) were used. At T1 both parents
up, during breakfast, and when dressing the child, (2) during the day from 9.00 a.m. to 1.00 p.m., (3) during the day from 1.00 p.m. to 5.00 p.m., (4) when having dinner, during playtime, at bedtime, (5) in the evening until midnight, and (6) when the child needed care in the middle of the night. Response options ranged from 1 (“I do it all”) to 9 (“Partner does it all”). The primary caregiver was, therefore, the parent with the lower average score on these six items. If the parents had the same average score, the question “During the past week, who spent the most time with [name infant(s)]?" (asked by the research assistant when arranging the home visit) was used to identify the parent with the primary caregiver role.
Another important factor in this study was to determine which parent was biologically related to the child. Both primary and secondary caregivers of heterosexual families were biological related to their child(ren). Of the 40 gay-fathers, 28 disclosed whether or not they were biologically related to their child. Of the primary caregivers, 57% were biologically related to the child. Of the 76 lesbian-mothers, 97% of the primary caregivers were biological related to their child.
Procedure
In each country, the recruitment of the participants was with the aid of lawyers with an expertise in surrogacy (for the gay couples) and through parenting support groups, forums, and magazines. Lesbian and heterosexual parents were also recruited at fertility clinics. Inclusion criteria concerning methods of conception were: gay-father families had to have used surrogate carriers, lesbian-mother families had to have used sperm donors, and heterosexual-parent families had to have used IVF without sperm or egg donation. All participants gave a written informed consent.
The gathering of the data occurred on two occasions. Seeing that this study is part of a larger research project, a lot of different data was gathered. Only the data relevant to this study will be described. The first data collection (T1) was collected during a house visit when
the children were around 4 months old (within two weeks before and after a child turned 4 months). Before this visit, the parents filled in an online questionnaire. This questionnaire gathered information about several demographics, such as age, marital status, living location and so forth. During the first home visit, both parents completed, separately from each other and among other queries, the Who Does What questionnaire. When the child was around 12 months old (within two weeks before and after a child turned 12 months) (T2) the families were invited to visit a research lab. During this visit, both parents repeated the same questionnaires. Traveling expenses were compensated.
Study measures Division of tasks. To determine how parents divide household and child-care tasks, the Who Does What Questionnaire (WDW) was used (Cowan & Cowan, 1990). The WDW is designed to assess parents' reports about the division of household tasks, decision making, and child-care tasks. For this study, the subscales division of household tasks and division of child-care tasks were used. On a scale ranging from 1 (I do it all) to 5 (we do it equally) ending at 9 (my partner does it all), the parents reported the relative frequency of who performs each of the 12 tasks for each subscale. Examples of tasks were ‘Planning and preparing meals’, ‘Bathing the baby’ and ‘Arranging day-care or a
babysitter’. For each task, the parents indicated the division at the current moment and what their ideal division would be. For this study, only the current division was used. The
Cronbach's alpha and Spearman-Brown's split-half reliabilities for all subscales ranged from .92 to .99, which indicates high levels of reliability (Cowan & Cowan, 1990). For the sample of this study, the Cronbach's alpha for household tasks ranged from .23 to .53. For child-care tasks, the Cronbach's alpha ranged from .77 to .86.
Mean scores for household and child-care tasks were calculated separately for primary and secondary caregivers. This was done for the data of T1 and T2. To determine whether there were changes between T1 and T2, difference scores were calculated by subtracting the
scores of T2 with the scores of T1. This was done for the scores on household and child-care tasks, separately for primary and secondary caregivers.
Analysis plan
The data was nested because the two parents in each family were not independent of each other in their reports regarding the division of household and child-care tasks and therefore Repeated Measures Analyses was performed.
To test whether there were differences in the way heterosexual-parent, gay-father, and lesbian-mother families divide household and child-care tasks when the children were 4 months old, a repeated measures MANOVA was performed. Primary and secondary caregivers were compared to each other on the division of household and child-care tasks (within-subject factors) and also across the three family types (between-subject factors). Subsequent T-tests and/or ANOVA’s were performed to explain the differences that were found. To test whether there were differences in the way heterosexual-parent, gay-father, and lesbian-mother families divide household and child-care tasks when the children were 12 months old, a second repeated measures MANOVA was run.
To determine whether there were significant changes between T1 and T2 a repeated measures MANOVA was performed with use of the difference scores. Primary and secondary caregivers were yet again compared to each other on the division of household and child-care tasks (within-subject factors) and across the three family types (between-subject factors), based on the mean difference scores. If significant differences were found, subsequent T-tests and/or ANOVA’s were performed to explain these differences.
Results Descriptive analyses
Between T1 (N = 160) and T2 (N = 137), twenty-three parents dropped out of the study and were therefore not included in the analyses. Analyses were run as if no data was missing.
The mean score for the primary caregivers on the division of household tasks when the children were 4 months old (T1) was 4.83 (SD = 0.74) and for the division of child-care tasks, this was 3.95 (SD = 1.03). At T1, secondary caregivers scored on average 4.92 (SD = 0.84) for the division of household tasks and 5.83 (SD = 0.93) for the division of child-care tasks. When the children were 12 months old (T2), primary caregivers scored on average 4.63 (SD = 0.70) for the division of household tasks and 4.33 (SD=1.04) for the division of child-care tasks. The mean scores for the secondary caregivers at T2 were 5.08 (SD = 0.79) and 5.47 (SD = 0.96) for the division of household tasks and division of child-care tasks, respectively. These mean scores imply that primary caregivers feel that they perform on average more household and child-care tasks than their spouses. Secondary caregivers feel they perform half or slightly less than half of the household and child-care tasks. Table 2 presents mean scores on household and child-care tasks for primary and secondary caregivers for each family type. Additionally, the difference score between T1 and T2 was included in the table. When checking for assumptions that are required for a repeated measures MANOVA, the test for equality of covariance was significant (p < .001). Therefore, the Pillai’s criterion was used to ensure robustness and to reduce the chance of a type II error (Tabachnick & Fidell, 2007).
Biological relatedness
Hypothesized was that biological relatedness to the child is of influence in the division of tasks. To include biological relatedness as a covariate in the analyses, assumptions were
checked. One of the assumptions entails homogeneity of regression slopes, in which there is assumed that the relation between the dependent variable and the covariate is true for all groups (Field, 2013). When the assumption of homogeneity of regression slopes is met, there should be an equal number of primary caregivers who were biologically related to their child in each family type. This should also be true for the secondary caregivers. With Chi-square tests (one for the primary caregivers and one for the secondary caregivers) we tested whether the covariate biological relatedness (yes, no) was equally distributed across all three family types (heterosexual, gay-father, and lesbian-mother families). The biological relatedness for primary caregivers appeared to be significant, !" (2) = 22.6, p < .001, just as for secondary
caregivers, !" (2) = 47.68, p < .001. These results show that not all primary caregivers had the
same kinship across the family types. The same is true for the secondary caregivers.
Therefore, the assumption was not met, and biological relatedness turned out to be unfit to be included as a covariate.
To be able to determine whether biological relatedness with the child is of influence on the division of household and child-care tasks, further analyses were performed. Using an ANOVA, biological and non-biological related primary caregivers were compared on the division of household tasks. Results were non-significant, F(1,72) = 0.03, p = .86. Also, biological and non-biological related primary caregivers were compared on the division of child-care tasks. These appeared to be non-significant as well, F(1,72) = 0.02, p = .88. The ANOVA was repeated for the secondary caregivers. When comparing biological and biological related secondary caregivers on the division of household tasks, there was a non-significant result, F(1,70) = .31, p = .58. For the secondary biological and non-biological related caregivers, there did appear to be a significant difference in the division of child-care tasks, F(1,70) = 19.38, p < .001. Secondary caregivers with a biological relation to the child performed more child-care tasks compared to secondary caregivers without a biological
relation. However, when controlling for family type, the difference between biological and non-biological related secondary caregivers disappeared, F(1) = 3.34, p = .07. Based on these results it seems that biological relatedness does not influence the division of household and child-care tasks.
Division of household and child-care tasks at 4 months
The repeated measures MANOVA for analyzing whether families differed in the division of household and child-care tasks when their child was 4 months old showed that there was no main effect for family type, F(4) = 0.15, p = 0.96, but there was a main effect for the caregiver role, F(2) = 55.69, p < .001. Furthermore, there was an interaction effect
between family type and the division of tasks, F(4) = 6.87, p < .001.
The main effect for the caregiver role appeared to be significant for the division of child-care tasks, F(1) = 112.28, p < .001. Parents did divide child-care tasks differently from each other, but they did not with regard to the division of household tasks. To investigate the direction of the main effect on caregiver role and the division of child-care tasks, a Paired-samples T-test was performed. This showed that the difference between primary and
secondary caregivers was significant for all family types. The difference was most substantial for the heterosexual parents, t(21) = -10.29, p < .001, d = 2.19. Table 3 presents the difference between primary and secondary caregivers on the division of child-care tasks for each family type.
The interaction effect between family type and division of tasks also appeared significant on the division of child-care tasks, F(2) = 13.52, p < .001, and not for household tasks, F(2) = 1.69, p = .19. To investigate the interaction effect further, an ANOVA was performed. This analysis showed that primary caregivers across family types differed
significantly in the division of child-care tasks, F(2,77) = 8.64, p < .001. Post-hoc tests using the Bonferroni correction (see also table 4) showed that both gay-father and lesbian-mother
families differed from heterosexual families, .26 ± 1.68, p = .004 and .37 ± 1.60, p = .001, respectively, in which primary caregivers in heterosexual families performed more child-care tasks compared to primary caregivers in gay-father and lesbian-mother families (M = 3.24 vs M = 4.20 and M = 4.22). There was no difference between gay-father families and lesbian-mother families. When the ANOVA was repeated with the secondary caregivers there were also significant results, F(2,76) = 15.17, p < .001. Post-hoc tests showed that secondary caregivers in heterosexual families performed significantly less child-care tasks compared to secondary caregivers of gayfather and lesbianmother families, 1.74 ± .52, p < .001 and -1.60 ± -.56, p < .001.
In sum, at 4 months there were no differences with regard to the division of household task, but there were with regard to the division of child-care tasks. Primary caregivers across all family types performed more child-care tasks compared to the secondary caregivers, with the difference being most substantial between primary and secondary caregivers of
heterosexual families. Further, analysis showed that primary caregivers in heterosexual families performed more child-care tasks compared to primary caregivers in gay-father and lesbian-mother families. The same was true for secondary caregivers: secondary caregivers in heterosexual families performed significantly less child-care tasks compared to secondary caregivers of gay-father and lesbian-mother families.
Division of household and child-care tasks at 12 months
Before testing whether there were changes in the division of tasks during the first year of parenthood, the repeated measures MANOVA was replicated to see how families divide tasks when the children were 12 months old. Similarly to T1, there was no main effect for family type, F(4) = 1.07, p = .37, but there was a main effect for the caregiver role, F(2) = 14.91, p < .001. There was also an interaction effect between family type and caregiver role, F(4) = 2.85, p = .03.
At T2 there is not only a difference in how parents divide child-care tasks, F(1) = 28.01, p < .001, but also in how parents divide household tasks, F(1) = 7.90, p = .007. Paired samples T-tests showed that only gay-father families divided household tasks differently, in which the primary caregivers performed more tasks compared to the secondary caregivers, t(16) = -3.06, p = .007, d = 0.74. With regard to child-care tasks, there were also significant differences between primary and secondary caregivers in lesbian mother families, t(33) = -2.89, p = .007, d = .50, and heterosexual families, t(14) = -4.01, p = .001, d = 1.03. Both primary caregivers performed more child-care tasks compared to the secondary caregivers. Table 3 presents the difference between primary and secondary caregivers on the division of child-care tasks for each family type.
Just as in T1, there was also an interaction effect for child-care tasks and family type at T2, F(2) = 3.86, p = .03. An ANOVA showed that primary caregivers from different family types differed from each other, F(2,65) = 4.42, p = .02. Post-hoc tests with Bonferroni
corrections (see also table 4) showed that primary caregivers of heterosexual-parent families performed more child-care tasks compared to primary caregivers of lesbian-mother families, .13 ± 1.60, p = .02., M = 3.69 vs M = 4.55. As to secondary caregivers there were no
significant differences across family type, F(2,66) = 3.08, p = .05.
In sum, at 12 months only primary caregivers of gay-father families performed more household tasks compared to the secondary caregivers. Primary and secondary caregivers of lesbian-mother and heterosexual-parent families divided child-care tasks differently, in which the primary caregiver performed more tasks compared to the secondary caregivers.
Additionally, primary caregivers of heterosexual-parent families performed more child-care tasks compared to primary caregivers of lesbian-mother families.
Changes between T1 and T2
For the third research question (are there changes in the division of household and child-care tasks during the first year of parenthood), a repeated measures MANOVA showed that there is a main effect for caregiver role, F(2) = 7.82, p = .001. On both household tasks, F(1) = 5.77, p = .001, and child-care tasks, F(1) = 8.81, p = .004, there is a shift in the
division. Paired-samples T-test showed that there are no significant changes within gay-father families. There are changes within lesbian-mother families with regard to child-care tasks, t(33) = -2.83, p = .008, but not for household tasks. When comparing the average score of lesbian-mothers on child-care tasks at T1 with the average score at T2, their scores differed less (see table 2). Heterosexual-parent families had changes with regard to household tasks, t(14) = 2.18, p = .05, but not for child-care tasks. When comparing heterosexual parents’ average scores of household tasks at T1 with the average scores at T2, their scores differed more (see table 2).
Discussion
This study was the first to investigate the division of household and child-care tasks in heterosexual-parent families, gay-father families, and lesbian-mother families, in which at least one parent was biologically related to the child. With regard to household tasks (e.g cooking, cleaning, groceries) the results when the children were 4 months old showed that all parents from all three family types divided the household tasks evenly, with each partner doing about half the work. When looking at child-care tasks (e.g. bathing, feeding, dressing) at 4 months, results showed that primary caregivers in all family types spent more time on these tasks than the secondary caregivers did. When the children were 12 months old there were only differences found in gay-father families, with the primary caregivers performing more household tasks than their spouses. With regard to child-care tasks, primary caregivers of lesbian-mother families and heterosexual-parent families still performed more tasks
compared to their spouses. Heterosexual families had at 4 months as well as at 12 months the most skewed division of child-care tasks, with the primary caregiver spending the most time on child-care tasks in comparison with the other primary caregivers. During this first year of parenthood, lesbian-mother families shifted to a more egalitarian division in child-care tasks, but heterosexual-parent families shifted to an even less egalitarian division with regard to household tasks. None of these differences were explained by whether or not the parent had a biological connection to the child.
When the children were 4 months old there were no differences in the division of household tasks between heterosexual-parent, gay-father, and lesbian-mother families. In all three family types each partner performed about half the work. This is in line with earlier research, which reported an even division as well (Patterson et al., 2004). This research also showed that heterosexual couples tend to specialize more in the division of tasks, whereas lesbian couples were likely to share responsibility for every different task. The specialization of tasks in heterosexual-parent families may have a traditional origin, in which the division of household tasks is based on sex. Women do “women’s-work” (e.g. cleaning, laundry) and men do “male-tasks” (yard work, maintenance) (Greenstein, 2000). This traditional division may seem less egalitarian at first sight but evens out at the bottom line. Future research should determine whether specialization or shared responsibility indeed occurred within the sample of this study.
With regard to the division of child-care tasks, heterosexual families have the most skewed division. Primary caregivers, which were always the mothers, performed more child-care tasks compared to the secondary child-caregivers, the fathers, when the children were 4 months old. This result is consistent with findings of earlier research (Hall et al., 1995; Patterson et al., 2004; Patterson & Farr, 2011), and is also in line with the gender theory in which mothers are more predisposed to embody nurturing and caregiving roles (Hall et al., 1995). Contrary
to earlier research, there were also differences between primary and secondary caregivers in lesbian-mother and gay-father families. In both family types, the division was more
egalitarian compared to heterosexual families, but during this first period of parenthood, primary caregivers still performed more child-care tasks than their spouses. Based on the gender theory it was expected that gay-father and lesbian-mother families would be more likely to achieve an egalitarian division of tasks because the sex of the two parents would not determine a predisposed role as with heterosexual parents (Goldberg & Perry-Jenkins, 2007). Where heterosexual parents are influenced by processes of gender socialization and gender ideologies, gay-father and lesbian-mother parents were expected to be more ‘free' to make rational choices in how they want to parent. Interestingly, there were differences between the two parents within gay-father and lesbian-mother families in this study, with the primary caregiver performing more child-care tasks compared to the secondary caregiver. Perhaps the societal demands push lesbian-mothers and gay-fathers in the same traditional roles. Society does have certain expectations of appropriate parenting behavior. This might make it difficult for gay-father and lesbian-mother families to create new definitions of parenting roles (Ciano-Boyce & Shelley-Sireci, 2003), which might result in similar skewed divisions in child-care task as in heterosexual families.
When the children were 12 months old, there were differences found in the division of household tasks between primary and secondary caregivers in gay-father families, but not for heterosexual-parent families and lesbian-mother families. Primary caregivers in gay-father families performed more household tasks than their spouses. Johnson and O’Connor (2002) found similar results. They report that gay-father families do share child-care tasks, but not household tasks. A possible explanation could be that gay-father and lesbian-mother families make truly intentional choices in how to parent their children (Golombok, 2000) because becoming a parent was such a strong desire and the path to becoming a parent such a struggle
(Johnson & O'Connor, 2002). These intentional parenting choices can have a direct influence on the division of child-care tasks, which could result in a more egalitarian division.
Household tasks, on the other hand, may be considered to be less important to discuss and more likely to be influenced by personal preferences, which could result in a skewed division, as was found in this study.
At 12 months, primary caregivers of heterosexual-parent families and lesbian-mother families still performed more child-care tasks than their spouses. The difference between primary and secondary caregivers in gay-father families for child-care tasks disappeared at this time. An explanation for the differences between primary and secondary caregivers could be related to hours spent working in paid employment. Patterson et al. (2004) found in their study among heterosexual and lesbian-mother families that the number of hours spent in paid employment influenced the involvement in child-care tasks. The lesbian-mothers spent the same number of hours in paid employment, but heterosexual fathers spent twice as many hours in paid employment as did their wives. Lesbian-mothers were equally involved in child-care, but heterosexual mothers were more intensively involved in child-care than their
husbands. However, in this study, only data was gathered on whether parents worked part-time (12 to 35 hours per week) or full-part-time (36 hours or more per week). Possibly, primary caregivers of heterosexual-parent families and lesbian-mother families indeed spent fewer hours in paid employment and were, therefore, more involved in child-care. Gay-father families might share hours spent in paid employment more evenly. Unfortunately, without accurate data, this cannot be checked. Future research with more precise data with regard to hours spent in paid employment should investigate whether this indeed influences the involvement in child-care.
Hypothesized was that there would be fluctuations within lesbian-mother families and not so much in heterosexual-parent and gay-father families when comparing the division of
tasks during the first year of parenthood. Within lesbian-mother families there was expected that the division would shift from less equal to more equal. There were indeed fluctuations within lesbian-mother families: When comparing child-care tasks, lesbian-mothers shifted to a more egalitarian division. This confirms the hypothesis in which lesbian-mother families will strive for an egalitarian division (Farr & Patterson, 2013). In contrast, primary caregivers of heterosexual families performed more household tasks at 12 months compared to 4 months. This was not necessarily expected but does fit with the traditional ideologies heterosexual families tend to hold on to (Greenstein, 2000). Based on the results of when the children were 4 and 12 months old, it seemed that gay-fathers also shifted in their division of tasks. With regard to the division of child-care tasks, gay-father seemed to have shifted to a more
egalitarian division, while shifting to a less egalitarian division of household tasks. However, the difference test did not detect this shift. Possibly, the number of gay-fathers that dropped out of the study was the reason the test did not detect an effect. Future research with a larger sample should investigate this matter further, given that the results of this study are an immediate cause to believe that shifts do occur.
Furthermore, there was a notable change in heterosexual-parent families and gay-father families: At 4 months primary caregivers in heterosexual families performed more child-care tasks compared to the primary caregivers in gay-father and lesbian-mother families. This difference was as hypothesized the most pronounced between heterosexual and gay-father families. At 12 months, however, the large difference between heterosexual-parent families and gay-father families disappeared. With this result, there is a strong support for the hypothesis that this difference is due to maternity leave (Rehel, 2014). Heterosexual families received maternity leave during the first period of parenthood, but neither men in gay-father families did and had to divide child-care tasks right from the start. This explains the big difference between heterosexual and gay-father families at 4 months. At 12 months there is no
question of maternity leave anymore for heterosexual-parent families, and the large difference between heterosexual-parent families and gay-father families disappears. However, the
difference between primary caregivers of heterosexual-parent families and the primary caregivers of lesbian-mother families cannot be explained by maternity leave.
What was also expected was that the biological relatedness to the child would
influence the division of tasks and that differences between primary and secondary caregivers in gay-father and lesbian-mother families could be explained by whether or not a parent had a biological connection with the child (Goldberg & Perry-Jenkins, 2007). This appeared not to be the case. When a primary caregiver performed more child-care tasks, this was not related to the fact that he or she was biological related to the child. When looking at a study in which none of the parents are biologically related and the child is adopted, it is interesting to see that similar patterns in the division of tasks are found (Farr & Patterson, 2013). Among the 104 adoptive heterosexual-parent, gay-father and lesbian-mother families there was also a more egalitarian division within the gay-father and lesbian-mother families compared to the heterosexual-parent families. Additionally, there were also differences between the two parents, in which one parent performed more tasks compared to the other parent. This implies that the division of tasks is influenced by other factors than biological relatedness. Another explanation might be the fact that there is one parent without a biological connection to the child, which may actually serve as a motivation for this parent to have a high degree of involvement with the child. He or she may feel that more work is needed to establish a
meaningful relationship with the child. The biologically related parent, on the other hand, may be particularly sensitive to the partners' position and may put an effort in supporting the relationship between the child and the non-biological related parent (Johnson & O'Connor, 2002).
At this point, several strengths and limitations of this study need to be acknowledged, starting with the strengths. Very few studies have included families from three different types (heterosexual-parent, gay-father and lesbian-mother families). Therefore, this study makes an important contribution to the knowledge about similarities and differences across these families. A unique aspect of this study is the fact that at least one parent had a biological connection to the child. Most studies with gay-father and lesbian-mother families are with adoptive and mostly older children. Combined with the longitudinal design, in which insight is gathered during the first year of parenthood, this study explores new grounds.
On the other hand, the relatively high socio-economic status (SES) and the mainly White sample limits the generalizability to the whole population of first-time parents from heterosexual, gay-father, and lesbian-mother families. Yet, these characteristics are very common in lesbian-mother and gay-father couples who use IVF or surrogacy to become parents. Surrogacy, for instance, is a very costly affair (between $90,000 and more than $120,000 in the US) (Thompson & Dodge, 2018) and therefore only an option for couples with high incomes. Also, the nonprobability techniques that were used to recruit the families hamper generalizability: The preset criteria for participating in the study might have caused a heterogeneous sample, which influenced the results (Bryman, 2012). Another limitation comes with the small sample size. A power analysis showed that there was enough power to detect large effects, but not necessarily smaller ones. This implies that small differences were possibly not discovered due to the small sample size. Finally, the use of self-rapport brings along several limitations, such as the risk of missing data, social desirability and the gap between stated and actual behavior (Bryman, 2012). Earlier research found a discrepancy between the responses of two partners (Shechory & Ziv, 2007). In this study, the perception of the concept of "equality" in the division appeared not the same between each partner. This may also have happened in our study, which could imply that conclusions based on the
self-rapports can be misleading.
The division of household and child-care tasks is an important aspect of coparenting but also just one aspect. Coparenting should, in fact, be placed in a broader perspective. Feinberg (2003) states that coparenting (he marked four core coparenting components: Support/undermining, Childrearing agreement, Division of labor and Joint family
management) is directly and indirectly influenced by several factors (e.g. parent and child characteristics, child adjustment, interparental relationship, parenting). With the Ecological Model of Coparenting Feinberg created, he shows that coparenting is a complex concept (Feinberg, 2003). Future research should therefore also focus on gathering data regarding parent and child characteristics, parent and child adjustments, interparental relationships, and the (social) support families receive. This can extend the knowledge about how coparenting relationships are formed and how they develop within families. Seeing the wide diversity of family forms nowadays (Furstenberg, 2014), it only makes sense to include gay-father and lesbian-mother families, just as nuclear father-mother families in these studies. This inclusion is necessary because even though this study focused only on one aspect of coparenting, it still showed that there are differences between heterosexual-parent, gay-father, and lesbian-mother families. The knowledge this study provided should also be a cause for inducing a discussion whether parenting programs should be adapted to diverse family types. Parenting programs with coparent components are usually based on knowledge that is gathered through research in nuclear father-mother families only. Assumptions are made that these components are also suitable for gay-father and lesbian-mother families. These assumptions could just as well be incorrect.
An increasing number of children are born and raised within gay-father and lesbian-mother families. Many studies have focused on coparenting with regard to the division of household and child-care tasks, but these studies are mainly within heterosexual families. This
study made an important contribution to the knowledge about similarities and differences across heterosexual, gay-father and lesbian-mother families. Differences do exist between heterosexual-parent, gay-father and lesbian-mother families in the division of household and child-care tasks. During the first period of parenthood (4 months), all parents divided
household tasks evenly among each other, but not child-care tasks. Primary caregivers across all family types performed more child-care tasks than the secondary caregivers did, with heterosexual families having the most skewed division. After a year of parenthood (12 months), primary caregivers from heterosexual-parent families still performed more child-care tasks than their spouses. The same was true for lesbian-mother families, however the division between primary and secondary caregivers became more egalitarian. Gay-fathers shared child-care tasks equally at this time, but not household tasks. Finally, biological relatedness to the child did not influence the division of tasks. Based on these results it can be concluded that differences between heterosexual-parent, gay-father, and lesbian-mother families do exist. Nuclear father-mother families should therefore not be seen as
References
Bos, H. & van Gelderen, L. (2010) Homo- en lesbisch ouderschap in Nederland. In Keuzenkamp, S. (Ed.), Steeds gewoner, nooit gewoon (pp. 104-117). Den Haag: Sociaal en Cultureel Planbureau, Retrieved from
https://www.scp.nl/Publicaties/Alle_publicaties/Publicaties_2010/Steeds_gewoner_no oit_gewoon
Bryman, A. (2012). Social research methods (4th edition). Oxford: Oxford University Press Cancian, M., Meyer, D. R., & Cook, S. T. (2011). The evolution of family complexity from the perspective of nonmarital children. Demography, 48, 957–
982. doi: 10.1007/s13524-011-0041-4
Carlson, D.L., Hanson, S., & Fitzroy, A. (2016). The division of child-care, sexual intimacy and relationship quality in couples. Gender & Society, 30, 442-446. doi:
10.1177/0891243215626709
Chan, R.W., Brooks, R.C., Raboy, B., Patterson, C.J., & Parke, R.D. (1998). Division of labor among lesbian and heterosexual parents: Associations with children’s adjustment. Journal of Family Psychology, 12, 402-419. doi: 10.1037/0893- 3200.12.3.402
Ciano-Boyce, C. & Skelley-Sireci, L. (2003). Who is mommy tonight. Journal of Homosexuality, 43, 1-13. doi: 10.1300/J082v43n02_01
Cowan. C.P. & Cowan, P.A.. Who does what? In Touliatos J, Perlmutter BF, Straus MA (eds) Handbook of family measurement techniques. 1990. Sage, Beverly Hills, California, US, 447 – 448.
Durtschi, J.A., Soloski, K.L., & Kimmes, J. (2017). The dyadic effects of supportive coparenting and parental stress on relationship quality across the transition to parenthood. Journal of Marital and Family Therapy 43, 308–321. doi: 10.1111/jmft.12194
Farr, R.H. & Patterson, C.J. (2013). Coparenting among lesbian, gay, and heterosexual couples: Associations with adopted children’s outcomes. Child Development, 84, 1226–1240. doi: 10.1111/cdev.12046
Feinberg, M.E. (2003). The internal structure and ecological context of coparenting: A framework for research and intervention. Parenting, 3, 95-131. doi: 10.1207/ S15327922PAR0302_01
Feinberg, M.E., Kan, M.L., & Kazak, A.E. (editor) (2008). Establishing family foundations: intervention effects on coparenting, parent/infant well-being, and parent-child
relations. Journal of Family Psychology, 22, 253-263. doi 10.1037/0893- 3200.22.2.253
Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th edition). London:
SAGE Publications Ltd
Furstenberg, F. (2014). Fifty years of family change. The ANNALS of the American Academy of Political and Social Science, 654, 12-30. doi:
10.1177/0002716214524521
Garber, B. D. (2004). Directed co-parenting intervention: Conducting child-centered interventions in parallel with highly conflicted co-parents. Professional Psychology:
Research and Practice, 35, 55-64. doi: 10.1037/0735-7028.35.1.55
Thompson, A.D. & Dodge, D. (2018) The cost of becoming a gay dad: The ultimate guide [Web log post]. Retrieved from https://www.gayswithkids.com/the-cost-of-gay- becoming-a-gay-dad-2465969199.html
Goldberg, A.E., & Gatrell, N.K. (2014). LGB-parent families-chapter three: The current state of the research and directions for the Future. Advances in Child Development and Behavior, 46, 57-88. doi: 10.1016/B978-0-12-800285-8.00003-0
Goldberg, A.E., & Perry-Jenking, M. (2007). The division of labor and perceptions of parental roles: Lesbian couples across the transition to parenthood. Journal of Social and Personal Relationships, 24, 297–318. doi: 10.1177/0265407507075415
Golombok, S. (2000). Parenting. What really counts? London: Routledge
Golombok, S. (2017). Parenting in new family forms. Current. Opinion in Psychology, 15, 76-80. doi: 10.1016/j.copsyc.2017.02.004
Greenstein, T.H. (2000). Economic Dependence, gender, and the division of labor in the home: A replication and extension. Journal of Marriage and Family, 62, 322- 335. doi:10.1111/j.1741-3737.2000.00322.x
Hall, L. D., Walker, A. J., & Acock, A. C. (1995). Gender and family work in one-parent households. Journal of Marriage and the Family, 57, 685-692. doi:10.2307/353923 Johnson, S. M., & O’Connor, E. (2002). The gay baby boom: The psychology of gay parenthood. New York, NY: New York University Press.
Keuzekamp, S. (2010) Houding van Nederlandse bevolking. In Keuzenkamp, S. (Ed.), Steeds gewoner, nooit gewoon (pp. 31-52). Den Haag: Sociaal en Cultureel Planbureau, Retrieved from https://www.scp.nl/Publicaties/Alle_publicaties/Publicaties_2010/ Steeds_gewoner_nooit_gewoon
Margolin, G., Gordis, E. B., & John, R. S. (2001). Coparenting: A link between marital conflict and parenting in two-parent families. Journal of Family Psychology, 15, 3–21. doi:10.1037/0893-3200.15.1.3
McHale, J. P., & Irace, K. (2011). Coparenting in diverse family systems. In J. McHale & K. Lindahl (Ed.), Coparenting: A conceptual and clinical examination of family systems (pp. 15–37). Washington, DC: American Psychological Association.
Patterson, C.J., & Farr, R.H. (2011). Coparenting among lesbian and gay couples in Coparenting: A conceptual en clinical examination of family systems. American Psychological Association, 127-146. Doi 10.1037/12328-006
Patterson, C., Sutfin, E., & Fulcher, M. (2004). Division of labor among lesbian and heterosexual parenting couples: Correlates of specialized versus shared patterns. Journal of Adult Development,11, 179-189
doi: 10.1023/B:JADE.0000035626.90331.47
Rehel, E.M. (2014). When dad stays home too: Paternity leave, gender, and parenting. Gender & Society, 29,110-132. doi: 10.1177/0891243213503900
Schoppe-Sullivan, S.J., Brown, G.L., Cannon, E.A., Mangelsdorf, S.C., Sokolowski, M.S., & Kazak, A.E. (editor) (2008). Maternal gatekeeping, coparenting quality, and
fathering behavior in families with infants. Journal of Family Psychology, 22, 389-398. doi: 10.1037/0893-3200.22.3.389
Schoppe-Sullivan, S.J. & Mangelsdorf, S.C. (2013). Parent characteristics and early coparenting behavior at the transition to parenthood. Social Development, 22, 363-383. doi: 10.1111/sode.12014
Shechory, M. & Ziv, R. (2007). Relationships between gender role attitudes, role division, and perception of equity among heterosexual, gay and lesbian couples. Sex
Roles,56, 629-638. doi: 10.1007/s11199-007-9207-3
Sumontha, J., Farr, R.H., & Patterson, C.J. (2016). Social support and coparenting among lesbian, gay, and heterosexual adoptive parents. Journal of Family Psychology, 30, 987-996. doi: 10.1037/fam0000253
Tabachnick, B.B., & Fidell, L.S. (2007). Using multivariate statistics. Boston: Pearson Education, Inc.
Tornello, S.L., Kruczkowski, S.M., & Patterson, C.J. (2015) Division of labor and relationship quality among male same-sex couples who became fathers via surrogacy. Journal of GLBT Family Studies, 11, 375-394, doi:
10.1080/1550428X.2015.1018471
Updegraff, K., McHale, S. & Crouter, A. (1996). Gender roles in marriage: What do they mean for girls’ and boys’ school achievement? Journal of Youth and Adolescence, 25, 73-88. doi: 10.1007/BF01537381
Van Egeren, L. A. (2004). The development of the coparenting relationship over the transition to parenthood. Infant Mental Health Journal, 25, 453–477. doi:10.1002/ imhj.20019
Table 1
Demographic information about families
Heterosexual parents (n=22) Lesbian mothers (n=38) Gay fathers (n=20) ANOVA or x2 test Parents (n=160)
Mean age (in years at T1) 36.1(4) 32.8 (3) 39.8 (5) F = 20.11*
Ethnicity (% White) 95% 95% 93% PC = x2
= 5.5 SC = x2= 5.9 Education (% College or higher) 75% 87% 80% PC = x2
= 5.7 SC = x2= 3.7
Work status (% Full time) 68% 38% 43% PC = x2
= 5.8 SC = x2= 7.5 Annual family income
(% over $ 56,2565)
86% 68% 95% x2
= 6.78
Relationship duration (in years) 8.7(3) 6.4 (3) 10.4 (5) F = 9,03*** Marital status (% married/civil
partnership)
64% 92% 50% x2
= 13.60***
Living location Large sized city Medium sized city Small sized city Rural area 3% 11% 10% 4% 5% 24% 16% 3% 9% 11% 5% 0% x2 = 10.37
Table 2
Mean scores of child-care and household tasks reported among gay, lesbian, and heterosexual parents Gay-fathersa Lesbian-mothersb Heterosexual parentsc
Primary Secondary Primary Secondary Primary Secondary
T1 Child-care Household 4.21 (1.25) 4.62 (.71) 5.49 (.97) 5.08 (.81) 4.22 (.77) 4.82 (.67) 5.54 (.59) 4.97 (.71) 3.24 (.89) 5.06 (.86) 6.63 (.93) 4.72 (1.05) T2 Child-care Household 4.48 (.99) 4.45 (.67) 5.34 (1.08) 5.41 (.84) 4.55 (.97) 4.76 (.72) 5.32 (.69) 4.94 (.78) 3.69 (1.04) 4.56 (.66) 6.00 (1.22) 5.00 (.67) Difference T1 T2 Child-care Household -.17(.69) .14 (.47) .01 (.85) -.33 (.90) -.36 (.87) .03 (.65) .23 (.56) .04 (.53) -.44 (.84) .43 (.67) .57 (1.51) -.22 (.80) Note. a at T1 n = 40; at T2 n = 36; b at T1 n = 76; at T2 n = 70; c at T1 n = 44; at T2 n = 31
Table 3
Difference between primary and secondary caregivers in child-care tasks at T1 and T2
Primary Secondary 95% CI
Family type M SD M SD t-test p LL UL Cohen’s d
T1 Gay-father families 4.22 1.28 5.49 .96 -2.56(18) .02 -2.31 -.23 .58 Lesbian-mother families 4.22 .77 5.54 .59 -6.36(37) <.001 -1.74 -.90 1.03 Heterosexual families 3.24 .89 6.63 .93 -10.29(21) <.001 -4.07 -2.70 2.19 T2 Gay-father families 4.55 .97 5.31 1.04 -1.63(15) .12 -1.75 .23 .41 Lesbian-mother families 4.55 .97 5.33 .71 -2.89(33) .007 -1.33 -.23 .50 Heterosexual families 3.76 1.03 6.00 1.22 -4.01(14) .001 -3.44 -1.04 1.03 Note. CI = confidence interval; LL = lower limit; UL = upper limit
Table 4
Bonferroni comparison for family type on the division of child-care tasks at T1 and T2 95% CI
Comparison Mean difference Std. Error LB UB
T1 Primary caregivers Hetero vs gay .97* .29 .26 1.68 Hetero vs lesbian -.99* .25 -1.60 -.37 Gay vs lesbian .02 .26 -.62 .65 Secondary caregivers Hetero vs gay -1.13* .25 -1.74 -.52 Hetero vs lesbian 1.08* .21 .56 1.60 Gay vs lesbian -.05 .22 -.59 .50 T2 Primary caregivers Hetero vs gay .79 .34 -.04 1.63 Hetero vs lesbian .86* .30 .13 1.60 Gay vs lesbian .07 .29 -.64 .78 Secondary caregivers Hetero vs gay -.66 .33 -1.46 .14 Hetero vs lesbian -.68 .29 -1.39 .02 Gay vs lesbian .02 .27 -.64 .69 Note. * p < .05
