Chinese and South Korean Children’s Moral Reasoning Regarding the Fairness of a Gendered Household Labor Distribution

Abstract

This study explored age-related changes in Chinese and Korean children’s fairness judgments and reasoning regarding the gendered division of household labor. The majority of previous research on this issue has focused on adults’ experiences and been conducted in Western countries. Interviews were conducted with 133 children, 65 Chinese and 68 South Korean, ages of 9–11 (N=48), 12–14 (N=43), and 16–18 (N=42). Despite country differences in women’s involvement in the workforce in both countries women do most of the housework. The majority of participants across ages evaluated a hypothetical scenario in which the mother did most of the work as unfair, and endorsed as fair a hypothetical scenario in which housework was divided evenly. Developmental effects were found with younger children in both countries employing more equality justifications and adolescents providing more social convention justifications. In both countries children described their mother as doing the majority of the housework, and in contrast with their judgments about hypothetical situations, were evenly divided overall in their evaluations of whether this unequal distribution in their actual families was fair. Korean children were more likely to view their family’s division unfair than Chinese children, and were more likely to employ equality justifications in support of those judgments. Unexpectedly, across countries moral reasoning in the form of expectations of equity rather than gender stereotyping was employed to justify an unequal division of labor. This study’s findings suggest the value of investigating children’s fairness judgments and moral reasoning regarding both hypothetical and actual situations.

In most countries, mothers do the majority of their family’s housework (Fuwa, 2004; Greenstein, 2009; Lachance-Grzela & Bouchard, 2010; Qi & Dong, 2013). While much research has been dedicated to understanding why women may be taking on the bulk of the housework (Coltrane, 2000; Hochschild, 1989; Lachance-Grzela & Bouchard, 2010), and some on why many women find this fair (Greenstein, 2009; Jansen, Weber, Kraaykamp, & Verbakel, 2016), little research has investigated children’s perspectives on the gender unequal division of household labor. When the research has turned to children’s experiences of housework, the main focus has been on the amount that children engage in (Klein, Graesch & Izquierdo, 2009) and the gendered differences in their involvement (Blair, 1992; Lam, Greene, & McHale, 2016). As a result, little is known about children’s thoughts on the fairness of their family’s household labor distribution.

One way to investigate children’s thinking about the fairness of family’s household labor distribution is to determine the forms of judgments and reasoning about this issue. Social domain theorists have found strong evidence of children’s active evaluation and reasoning about parental behavior in the family context (Sinno & Killen, 2009; 2011; Smetana, 2011). According to social domain theory (SDT) both children and adults differentiate between several domains of social knowledge: moral, conventional, personal, and pragmatic (Nucci, Turiel, & Roded, 2017; Smetana, Jambon, & Ball, 2014; Turiel, 1983; 2002). The moral domain pertains to issues of justice, rights, and welfare. The conventional domain refers to agreed-upon social norms and rules that organize society and social interaction, such as social hierarchy and gender norms. The personal domain pertains to issues of individual choice and preferences, such as choices of friends or hobbies. The pragmatic domain refers to practical considerations such as feasibility, ability, or necessity. Social domain research has demonstrated that individuals coordinate their understandings of these domains in their evaluations of social issues including those within the family. Previous SDT studies of issues occurring in the family context have investigated adolescent-parent conflict regarding decision-making power (Chen-Gaddini, 2012; Smetana, 2011), young children’s negotiations with parents regarding their choice of clothing (Nucci & Weber, 1995), children’s decisions to deceive and defy their parents (Gingo, 2017), and children’s thoughts on parental gender roles (Brose, Conry-Murray & Turiel, 2013; Sinno, 2007; Sinno & Killen, 2009).

As the family is one of the principal contexts in which children learn about moral issues such as justice (Okin, 1989), it is crucial to consider how children are making sense of and evaluating their family’s situation. Investigating children’s thoughts on housework distribution is an important avenue of research because how housework is divided is an issue that involves concerns for fairness and justice, often involves the child’s participation (Thomson, 2007), and has been found to reflect and perpetuate structural race, class, and gender relations (Coltrane, 2000). Moreover, while previous cross-cultural research has found that around 41–45% of women in general find the gender unequal division of household labor fair (Braun, Lewin‐Epstein, Stier, & Baumgärtner, 2008; Greenstein, 2009; Tai & Baxter; 2018), whether children make similar judgments is an open empirical question. Ultimately, whether children accept or reject the gendered division of household labor has implications both for understanding children’s moral development as well as children’s future involvement in perpetuating or challenging society’s gender relations.

To address this gap in the literature, the present study investigated Chinese and Korean children’s evaluations and reasoning regarding the gendered division of household labor. While China and South Korea have both been historically influenced by Confucianism (Park & Cho, 1995; Tu, 1998), and in the past forty years have undergone swift economic development (Hu, 2018; Ji, Wu, Sun, & He, 2017; Kim, 2017), they have distinct organizations of gender and labor (OECD, 2008/2014). Furthermore, previous scholarship has shown that geographical generalizations and global dichotomies are insufficient for understanding and capturing cultural heterogeneity and complexity (Mohanty, 2003; Gutiérrez & Rogoff, 2003; Turiel, 2002). Therefore, a comparison of Chinese and Korean children’s social and moral reasoning allowed for both extending previous moral developmental research that has been primarily focused on Western samples, as well as a more nuanced investigation into the role of culture and social context on social reasoning beyond generalized comparisons of “West” and “East” or “progressive” and “traditional.”

Children’s Reasoning about Parental Gender Roles

The majority of prior research on children’s and adolescents’ thoughts on parental gender practices has focused on the division of childcare responsibilities (Brose et al., 2013; Sinno & Killen, 2009). Previous studies have found that American children and adolescents are supportive of mothers serving as the family’s primary caretaker. Sinno (2007) found that children (10-year-olds), adolescents (13-year-olds), and young adults (19-year-olds) prefer the mother to be the primary caretaker regardless of work arrangement. Sinno and Killen (2009) found that 8- and 10-year-olds considered it acceptable for both mothers and fathers to be breadwinners, however they were more likely to find it acceptable for mothers to be homemakers than fathers. Another study reported that most 10- and 13-year-old’s considered it better for mothers to take on the double shift of child care and work, as mothers were believed to be more nurturing (Sinno & Killen, 2011). Similarly, Brose, Conry-Murray, and Turiel (2013) found that 16-year-olds thought mothers, rather than fathers, should stay home and care for their infant based on the assumption that mothers take better care of infants.

American children’s family structure and larger community context has been found to influence children’s evaluations and reasoning regarding family roles (Sinno & Killen, 2009; 2011; Sinno, Schuette, & Hellriegel, 2017). For example, Sinno and Killen (2009) found that 7- and 10-year-olds who came from traditional homes (i.e. where the mother was mainly a homemaker) were more likely to hold gender stereotypic beliefs, and consider it unacceptable for a father to be a homemaker. In a related study, Sinno and Killen (2011) found that 10- and 13-years-olds from families in which mothers worked full- or part-time were more likely to consider hypothetical scenarios in which the mother did the second-shift parenting as unfair and to use moral reasoning. On the other hand, children who came from traditional homes (i.e. where only the father worked) were more likely to consider it acceptable for the mother to do the majority of the caretaking, and to employ social conventional reasoning.

In summary, American children and adolescents appear to still hold gendered expectations regarding caretaker roles, believing mothers are better suited for such a role (Brose et al., 2013; Sinno, 2007; Sinno & Killen, 2011). On the other hand, to date, no study has directly explored how children and adolescents evaluate and reason about situations involving how parents divide the housework. Previous research has suggested that boys may be more likely to endorse traditional gender differences than girls (Brose et al., 2013; Gere & Helwig, 2012; Schuette & Killen, 2009), that children and adolescents are more likely to employ social conventional reasoning when evaluating mothers’ gender roles than fathers’, and that family structure, community context, and culture may influence moral evaluations and reasoning. However, the majority of studies investigating children’s moral reasoning regarding the gendered division of labor have been conducted in the North American (US or Canada) context, where compared to other countries, men are more actively involved in doing housework and childcare (e.g., US men reported doing 37.4% of the family’s daily unpaid labor; OECD, 2017).

Considering Culture & Family Structure

As found in most countries around the world, both Chinese and South Korean women report doing the majority of their family’s household labor (Oshio, Nozaki, & Kobayashi, 2013; Qian & Sayer, 2016). Korean women report spending 215 minutes a day on housework, while Chinese women report spending 234 minutes (OECD, 2008/2014). However, men’s reported daily time spent on housework in each country markedly differs. Korean men are reported to spend a total of 49 minutes a day on housework, while Chinese men report spending 91 minutes (OECD, 2008/2014). As a result, Chinese women do on average 72% of their family’s housework, while Korean women do 81% of their family’s housework.

In terms of participation in the labor force, however, Chinese women reported spending more hours on paid work than all other women in OECD countries and non-OECD countries such as India and South Africa (OECD, 2008). This may be the result of the historical and continued high participation of women in the Chinese workforce (Hu, 2018), where 72% of mothers with children ages six and under were employed in 2010 (National Bureau of Statistics in China, 2011). On the other hand, South Korea has a 50% female labor force participation rate (Qian & Sayer, 2016), and women ages 30–45 (child-bearing and rearing age) are even less likely to be employed than their younger and older counterparts (Statistics Korea, 2017). Therefore, in terms of family structure, the majority of Chinese children have working mothers, while at least half of Korean children are likely to have mothers who are primarily homemakers.

Prior research investigating South Korean and Chinese children’s reasoning regarding parental gender roles and housework is scarce. Despite research suggesting that Korean parents hold more traditional gender beliefs than American parents (Conry-Murray, Kim, & Turiel, 2015) and Japanese parents (Sagara & Kang, 1998), Korean children (9- and 12-years-old) have been found to judge a parent’s decision to exclude a child in engaging in counter-stereotypic activities based on gender as wrong (Park, Lee-Kim, Killen, Park, & Kim, 2012). While Korean children were not directly asked to consider parental gender roles, past research suggests that despite strong cultural gender norms, Korean children are critical of gender-based inequalities (Noh & Midgette, 2018). To the author’s knowledge, no study has directly explored Chinese children’s moral reasoning regarding gender inequalities in the home.

The Present Study

While the majority of Chinese and Korean children are growing up exposed to distinctly gendered divisions of household labor, no prior research has explored their evaluations of whether this division is fair. Furthermore, few studies on the distribution of household labor have specifically focused on comparisons between children’s evaluations and reasoning for actual and hypothetical situations. Given that the present study was conducted with an underrepresented sample, it was an open question whether children’s interpretations of hypothetical scenarios would be similar or different to their own family’s division of labor. Therefore present study investigated children’s judgments and reasoning regarding both hypothetical household labor divisions as well as their own family’s current division. This study had several hypotheses:

• H1. It was hypothesized that the majority of participants would judge a gendered division of labor (actual or hypothetical) as fair, and employ mostly gender normative/conventional justifications, such as stating that women are better suited to take care of the home (Brose et al., 2013; Conry-Murray, 2015; Schuette & Killen, 2009; Sinno & Killen, 2011; Sinno, 2007).

• H2. It was hypothesized that in keeping with the literature, boys, younger children, children from a traditional family structure (i.e. a family in which the mother did the majority of the housework), and Korean children who come from a community in which fathers do very little housework (OECD, 2014) would be more likely than their counterparts (i.e. girls, older adolescents, children from non-traditional families, and Chinese children) to find a gendered division fair and to employ conventional reasoning (Park et al., 2012; Sinno et al., 2017; Sinno & Killen, 2009; 2011).

• H3. It was expected that when presented a scenario in which there was a perfectly equal division between parents, the majority of participants would be supportive of such a division and find it fair based on principles of equality (Gere & Helwig, 2012; Thomson, 2007).

Method

Participants

A total of 133 children and adolescents, 65 Chinese and 68 Korean, participated in this study. In order to investigate possible developmental trends, children from across three age groups, 9–11, 12–14, and 16–18 were recruited during the year 2017. These ages were selected based on previous research suggesting that these ages were associated with important developmental shifts in social and moral reasoning (Nucci et al., 2017; Smetana et al., 2014). Chinese children were distributed among the three age groups as follows: 9–11-years (11 boys, 11 girls: M_age=10.42, SD= .72), 12–14-years (10 boys, 13 girls: M_age= 12.9, SD=.98), and 16–18-years (10 boys, 10 girls: M_age=16.52, SD=.49). South Korean children were distributed as follows: 9–11 years (14 boys, 12 girls: M_age=10.50; SD=.73), 12–14-years (10 boys, 10 girls: M_age=13.35, SD=.54), 16–18 years (12 boys, 10 girls: M_age=17.14, SD=.46). The total number of subjects for each age group was: 9–11-years (n=48), 12–14-years (n=43), and 16–18 years (n=42). A within-subject design was employed and participants were presented with the same scenarios. Between-subject variables included age, family structure, country, and fairness judgment. Power analysis conducted in G*Power 3.1 revealed that the sample size for this study was sufficient for repeated-measures ANOVA analysis for 24 groups, with an alpha of .05, a power of .80, and an effect size of f=.25 (Faul, Erdfelder, Lang, & Buchner, 2007).

Participants were recruited through teacher recommendation in local schools in Changchun, China, and Seoul, South Korea. To control for social class, only participants from middle-class families (based on parents’ profession) were recruited. All of the Korean participants were ethnically Korean, while 86.2% of Chinese participants were Han, the ethnic majority group of the Chinese population. Nine Chinese participants (13.8%) reported belonging to an ethnic minority group (4 Manchu, 2 Huizu, 1Chaozu, 1 Mengu, and 1 Miaozu). In order to standardize across participants and to investigate children’s reasoning regarding the division of labor between two-parents, only children from two-parent households were recruited for this study.

Procedure and Design

Parental consent and children’s assent was obtained following procedures approved by the institutional review board (IRB) at the University of California, Berkeley under the project titled “Gendered Household Labor Distribution & Morality: Social & Moral Reasoning about Household Chores in China & South Korea,” Protocol ID: 2016-10-9244. Chinese parents provided verbal consent for their children’s participation, while in South Korea, as is the local practice, the child’s school principal provided verbal approval in lieu of parental consent. In both countries children provided verbal assent. The majority of interviews were completed at the child’s school in a private room, such as the teacher’s office or break room. A few of the interviews were completed at the child’s home, or at the author’s office at the local university. All children completed a demographic questionnaire (i.e. gender, ethnicity, age, parental occupation, etc), a daily routine survey of the time they and other family members spent at school, work, leisure time, and doing specific housework tasks, and a 30–45 minute interview. The present analysis mainly focuses on the interview outcomes. The interview protocol was translated and back-translated by bilingual researchers, who were native speakers of either Chinese or Korean.

Interviews were conducted in the child’s native language. In China, the author, who is fluent in Mandarin, completed all interviews. In South Korea, the author accompanied a trained female Korean research assistant fluent in both Korean and English. The interviews were audio recorded. Chinese children were offered 50RMB (~$8), for completing the study. In South Korea children were offered a local snack or a lollypop for their participation.

Daily Routine Survey Measure

The survey presented a series of questions in which participants were asked to report the time spent on each activity in time ranges from: 0–10 minutes, 20–30 minutes, 45 min-1 hour, 1–2 hours, and 3+ hours. The survey asked participants to circle the daily time they and other family members spent on routine-activities, such as attending school, making breakfast, doing homework, working at home, doing laundry, etc. At the end of the survey participants were asked to circle the average time each family member spent on housework daily. While participants filled out the survey either a research assistant or the author was present and available for any questions. On average participants spent five to ten minutes filling out the three-page survey. None of the participants reported any difficulty in filing out the survey. Participants’ responses were converted into minutes and averages calculated for each family member role (i.e. mother, father, child, and grandparents).

Interview Protocol

The interview was constructed following methods from previous investigations of children’s moral reasoning (Nucci et al., 2017; Sinno & Killen, 2009; 2011; Sinno et al., 2017; Smetana et al., 2014). It was composed of semi-structured questions that allowed for investigation of participants’ fairness evaluations and justifications of the gendered nature of housework as well as their own family’s housework distribution. During the interview participants were asked to make judgments and provide reasons about the following: 1) One hypothetical scenario in which there is a gender unequal division of labor, 2) A second hypothetical scenario in which there is an equal division of household labor and childcare, and 3) Participants’ own family division. (Refer to Appendix for hypothetical scenarios).

Participants were asked to paraphrase each story, to assess their understanding, and then asked whether they found the story situation fair or unfair and their reasons for their judgments of the situation as fair or unfair. In addition to being presented hypothetical scenarios, participants were also interviewed about their own family’s household labor division. Participants were asked how their family divided the housework, whether one family member did more than the others, and the reason for their family’s current household labor division. To investigate their fairness evaluations and reasoning regarding, participants were also asked to judge whether their family’s current division was fair or unfair, and if so why.

Coding & Reliability

All interview recordings were transcribed verbatim by Chinese and Korean native speakers. These transcriptions were coded in their original language. Responses to each hypothetical scenario were coded for initial judgment: fair or not fair, and the following response to the question: Why or Why not?” were coded as a justification. When coding children’s initial fairness judgment (fair or unfair) of their own family’s division, an additional code of “reasonable” was added. This was as a result of a minority of participants (N=3) being unwilling to judge their family’ division as either fair or unfair. Instead, they stated that they considered their family’s division as reasonable. Justification types were coded based on previous social domain research exploring children’s social and moral reasoning about gender roles (Brose et al., 2013; Sinno & Killen, 2009; 2011) such as equality, with additional codes adapted to the current topic of housework, such as family membership (For coding scheme see Table 1).

Table 1.

Coding Scheme of Children’s Justifications

Domain	Justification Category
Moral
	EQUITY: expectation of difference by taking into account time and each other’s situation, capacity, needs, it ends up balancing out: Example: Whoever is freer does more, whoever is busier does less. EQUALITY: expectation of sameness in condition, situation, turn taking/ sharing, amount, (strict equality). Example: They are both working, so they should both do the housework. PARTICIPATION: expectation that all individuals should do something (some housework), be active. Example: They all should participate. They are each doing something. PERSPECTIVISM: assertion that, in making a decision, the person accounts for and accommodates to the other person’s needs or wishes. Example: The other person needs to rest; needs help.
Conventional
	FAMILY MEMBERSHIP: expectation for involvement in housework based on the assumption that one has such a responsibility as a result of being part of the family. Example: As a member of the family, they should do housework. CONTRACTUAL: It is up to both of them to decide. Its up to each family to decide. If that works for that family. Example: It is okay, if it has been agreed upon. GENDERED COMPETENCE: assertion that a person can do something because of a special competence or responsibility associated with a specific role/gender. Example: She is the mother so she should do the childcare.
Personal
	CHOICE: Describes the situation up to the individual’s discretion or preference.
	Example: It is their choice. She is willing to do it.
Pragmatic
	FEASIBILITY: Describes the division as being a result of circumstances outside of individual control. Example: Because they are unable to do it, since the situation doesn’t allow it. MONEY : Describes income as a factor in deciding how the housework should be divided. Example: They make more money, so they don’t have to do as much.
Other
	RELATIONAL: Considers if the behavior is necessary to maintain the intimacy or harmony of a relationship. Example: They should do it because they love each other.

Coder reliability was established through independent coding of 20% of the interviews in each country by two trained coders per country who were fluent in the language. For Chinese interviews the coder reliability had an agreement of Cohen’s κ = .83 for domains and Cohen’s κ =.81 for justifications. For Korean interviews, coder reliability agreement was Cohen’s κ =.90 for domains and Cohen’s κ =.88 for justifications.

Results

Data Analytic Plan

To test Hypothesis 2, separate logistic regressions were conducted on judgments following each situation, with fairness judgments as the dependent variable, and gender (female=1, male=0), age group (9–11=1, 12–14 =2, 16–18=3), country (china=1, korea=0), and family structure (mother most=1, other=0) as predictor variables. To test possible differences in justification usage, this study followed statistical analysis procedures of children’s social and moral reasoning established in prior research (Posada & Wainryb, 2008; Sinno et al., 2017; Wainryb, Shaw, Laupa. & Smith, 2001). Since some participants employed multiple justifications types, justifications type usage was calculated as proportions, with justifications being coded as either 0= “no usage,” .5 “partial usage,” or 1= “full usage.” For each scenario a repeated-measures analysis of variance (ANOVA) was conducted with the justification type proportion usage as the dependent variable, and country (x2), family structure (x2), age (x3) and fairness judgment (2) as factors, with justification type as the repeated measure. Only justification types occurring at a proportion of .05 or higher were included in the analysis. Preliminary analysis revealed no significant gender differences in justification usage. As a result, gender was excluded from the final analysis. To analyze interactions, follow up Tukey Honest Significant Tests (HSD) and pairwise comparisons using independent sample t-tests with Bonferroni adjustments were conducted.

Evaluations of a Hypothetical Gender Unequal Division of Household Labor

Fairness Judgments.

Contrary to this study’s Hypothesis 1, the majority of children (81.20%) found the first hypothetical scenario depicting a gendered labor distribution in which the mother both worked as a teacher and did the majority of housework, unfair. Binomial probability tests determined this to be different from chance (assumed p=.5, observed p=.18). Also inconsistent with Hypothesis 2, across ages, genders, family structures and countries, participants did not significantly differ in their likelihood of considering the gendered division unfair or fair (Overall model was not significant, LR χ² (4, N=131)=6.06, p=.19, Nagelkerke R² =.04,

Justifications.

Repeated-measures ANOVA revealed a significant main effect for justification usage, F(3, 520)=173.34, p<.001, η²=.50. Contrary to expectations (H1), a Tukey HSD post hoc test revealed that following children’s fairness judgments of the gendered hypothetical scenario, the moral justification of equality was significantly more likely to be used than any other justification. To explore possible interactions, repeated-measures ANOVAs were conducted on justifications with country (x2), family structure (x2), age (x3) and fairness judgment (x2) as possible factors interacting with justification types. As expected, the repeated-measures ANOVAs revealed a significant interaction between justification type and age group, F(6,520)=2.40, p<.02, η²=.02. As can be seen in Table 2, consistent with prior research on developmental differences within the moral domain (Nucci et al., 2017), post hoc pairwise comparisons revealed that the youngest participants (9–11), were significantly more likely to employ equality as a justification overall than the other two age groups. Contrary to expectations (H2), no country or family structure differences were found.

Table 2.

Gendered Labor Division Scenario Justifications by Fairness Judgment in Proportions by Country and Age

Grouping	Justifications
		Participation	Equality	Equity	Family Membership
Country
China
	Not Fair	.01(.06)	.78(.31)	.02(.11)	.08(.20)
	Fair	.27(.43)	.05(.16)	.33(.42)	.05(.16)
Korea
	Not Fair	.00(0)	.94(.20)	.00(0)	.01(.06)
	Fair	.46(.49)	.06(.24)	.27(.44)	.00(0)
Age
9–11	Not Fair	.01(.07)	.90(.22)	.00(0)	.01(.07)
	Fair	.59(.54)	.00(0)	.19(.44)	.00(0)
12–14
	Not Fair	.00(0)	.85(.26)	.01(.09)	.05(15)
	Fair	.37(.47)	.08(.28)	.33(.43)	.04(.14)
16–18
	Not Fair	.00(0)	.81(.34)	.03(.12)	.09(.22)
	Fair	.31(.45)	.06(.17)	.31(.45)	.00(0)
Evaluation
	Not Fair	.00(.04)	.86(.27)	.01(.08)	.05(.16)
	Fair	.39(.47)	.06(.17)	.29(.42)	.02(.09)
Overall		.07(.25)	.71(.41)	.06(.22)	.04(.14)

Note. Only justifications with overall usage of .05 or higher presented.

Fairness judgment & justification interaction.

There was a significant interaction between fairness judgment (2) × (4) justification type, F(3, 524)= 135.07, p<.001, η²=.43. Such an interaction, although rarely explored in social domain empirical studies, is not entirely unexpected (Mulvey & Killen, 2016). Post hoc pairwise comparison revealed that, contrary to expectations (H1), moral reasoning was employed both for judging the gendered division fair and unfair. The analysis found that participation (t=8.45, p<.001) and equity (t=6.11, p<.001) were significantly more likely to be employed as justifications by children who found the gendered division of labor scenario fair. On the other hand, the justification of equality was significantly more likely to be employed by participants who judged the gendered division unfair (t=17.29, p<.001).

As a result of the significance of fairness judgment on justification usage, a repeated-measures ANOVA was conducted separately examining possible interactions of age, family structure, and country on justification type, when the story was judged fair=1, and when the story was judged unfair=0.

Repeated-measures ANOVA found a significant country X justification type interaction, F(3, 424)=10.30, p<.001, η²=.06, for children who evaluated the story as unfair. Post hoc pairwise comparisons revealed that, contrary to this study’s Hypothesis 2, equality, a moral justification, was significantly more likely to be used by Korean children (t=5.05, p<.001). On the other hand, family membership, a conventional justification, was significantly more likely to be employed by Chinese children (t=2.27, p<.02). Therefore, of the children who found the gendered division unfair (81.20%), Chinese children were more likely to reason that a family member had the responsibility to contribute, while Korean children were more likely to argue that there should be an expectation of equality between spouses.

A significant age X justification interaction was also found following judgments that the division was unfair, F(6, 420)=2.14, p<.04, η²=.02. Consistent with previous developmental findings, post hoc pairwise comparisons revealed that 9–11-year-olds were significantly more likely to employ the moral justification of equality than 16–18 year olds (t=2.26, p<.02). However, contrary to Hypothesis 2, 16–18-year-olds were significantly more likely to employ a conventional justification, family membership, than 9–11 year olds (t=2.51, p<.01).

In terms of the children who judged the gendered division labor fair, there were no significant interactions, likely as a result of a small sample size (n=25). Together, these results contradict Hypothesis 1: the majority of children did not consider the hypothetical gendered division fair, and employed mostly moral justifications. Furthermore, contrary to Hypothesis 2, differences in participant characteristics (e.g. age group) did not result in differences in fairness judgments, and in terms of justifications, Korean and younger children were more likely to employ moral reasoning than their Chinese and older adolescent counterparts.

Evaluations of Own Family Household Labor Division

Family’s Current Division.

Consistent with previous adult survey reports (OECD, 2008/2014), most participants (68.7%) reported their mother as primarily responsible for the family’s household labor. Almost fifteen percent (14.5%) of participants reported that their father did most of the household labor, while 10.7% reported that their grandmother did most of the household labor, and the remaining 6.11% of participants reported that their family’s division was evenly divided. The logistic regression model on family division (mother=1, other=0) with gender, age group, and country as predictors, was significant, LR χ² (3, N=131)=10.95, p<.01, Nagelkerke R² =.06. In line with prior research (OECD, 2008/2014), the effect for country was significant, OR .32, 95% CI [.14, .70], Z=−2.83, p<.005. The great majority of Korean children (80%) and a little over half of Chinese participants (57%) reported that their mothers did the majority of the housework. Therefore, as expected, Korean participants were significantly more likely to report that their mothers did most of the family’s household labor. No other significant effects were found.

Children’s survey reports were consistent with these findings. Korean mothers were reported as averaging significantly more time on housework than Chinese mothers (t=8.43, p<.001). Korean mothers averaged 138.67 minutes (SD=54.94) of housework a day, while Chinese mothers were reported as averaging 67.58 minutes (SD=40.58) a day. Fathers in both countries did significantly less housework on average than mothers (t=9.70, p<.001), with Chinese fathers averaging 39.23 minutes (SD=29.46), and Korean fathers averaging 46.10 minutes (SD=39.78) a day. Reports did not significantly differ by participant age or gender.

Own Family Fairness Judgment.

Contrary to this study’s Hypothesis 1, and, unlike the fairness judgments of hypothetical scenarios, participants were evenly divided in their fairness judgments of their own family’s household labor division. Almost fifty percent (47.73%) of participants judged their family’s division unfair, while the other 50 percent of participants judged their family’s division fair. A small minority of participants (2.27%, n=3), were unwilling to evaluate their family’s division as fair nor unfair, but believed the division to be reasonable.

An initial multinomial regression including the three fairness judgment types (fair= 1 base, unfair=0, reasonable=2), revealed significant differences in fairness judgments between those who judged the division fair and unfair, but not between them and those who judged the division reasonable. As a result, participants who judged the division as reasonable (n=3), were dropped from further analysis, and a binomial logistic regression on fairness judgments as a dichotomous variable (fair=1, unfair=0), on individual characteristics (i.e., gender, age, family division, and country) was conducted. The overall model was significant (LR χ² (4, N= 128) = 19.95, p =.000, Nagelkerke R²=.11) Results revealed significant effects for country, Odds Ratio (OR)=2.67, 95% CI[1.24, 5.77], Z=2.51, p<.01, and family structure, Odds Ratio (OR)=.39, 95% CI[.17, .92), Z−2.13, p<.03. Unexpectedly (H2), Chinese children (country=1) were significantly more likely to report their family’s division as fair (fair=1), whereas Korean children were significantly more likely to report their family’s division as unfair (See Table 3). Furthermore, also contrary to expectations (H2), children from traditional families (structure=1), were found to be less likely to judge their family’s division fair (f=0)

Table 3.

Children’s Fairness Judgment of Own Family’s Household Labor Division (Percentages)

Factors	Fair	Not Fair	Reasonable
Age Group
9–11	60.0(29)	40.0(19)	00(0)
12–14	50.0(21)	48.0(20)	2.0(1)
16–18	38.0(16)	57.0(24)	5.0(2)
Country
China	61.5 (40)	33.8(22)	4.0(3)
Korea	38.8(26)	61.2(41)	00(0)
Gender
Female	44.0(29)	53.0(35)	3.0(2)
Male	56.0(37)	42.0(28)	2.0(1)
Most
Mother	41.0(37)	56.6(51)	2.0(2)
Father	68.4(13)	31.6(19)	00(0)
Grandmother	57.3(8)	35.7(5)	7.0(1)
Equal	87.5(7)	12.5(1)	00(0)
Overall	50.0(66)	47.7(63)	2.27(3)

Note. Frequency in parenthesis. “Reasonable” was excluded from statistical analysis as a result of low number of participants.

Justifications.

Repeated-measures ANOVAs revealed significant differences in overall justification type usage, F(2, 384)=55.98, p<.001, η²=.22 (See Table 4 for justifications). Tukey HSD post hoc tests revealed that the moral justifications of equality and equity were significantly more likely to be employed than participation (i.e. the only other justification occurring more than .05 overall). A significant interaction between country and justification type was also found, F(4, 382)=37.16, p<.001, η²=.28. Contrary to expectations (H2), post hoc pairwise comparisons revealed that Korean children were more likely to employ equality as a justification (t=4.72, p<.001), while Chinese children were significantly more likely to employ participation (i.e., they are participating, doing something and amount doesn’t matter; t=2.34, p<.02), as a justification.

Table 4.

Justifications for Fairness Judgments of Own Household by Age, Country, and Family Structure

Justifications
Grouping	Participation	Equality	Equity
Age
9–11	.07(.25)	.48(.47)	.20(.38)
12–14	.07(.25)	.60(.46)	.18(.36)
16–18	.11(.30)	.56(.47)	.12(.32)
Country
China	.16(.35)	.40(.44)	.21(.37)
Korea	.01(.12)	.70(.44)	.14(.34)
Family Structure
Traditional	.06(.23)	.57(.47)	.18(.37)
Non-Traditional	.14(.33)	.49(.45)	.15(.32)
Overall	.09(.27)	.55(.47)	.17(.36)

Note. Only justifications that occurred 5% or higher are presented in table.

Fairness judgment & justification interaction.

As found with children’s reasoning regarding the gendered division of labor scenario, there was a significant interaction between children’s fairness judgments (fair/not fair) and justification types, F(4, 620)=52.06, p<.001, η²=.25 (See Table 5). Post hoc pairwise comparisons revealed a more complex relationship between moral justifications and fairness judgments than first hypothesized (H1). Participants who judged their family’s division fair were significantly more likely to employ participation, a moral justification, than those who judged it unfair (t=2.66, p<.001). Equity, also a moral justification, was also more likely to be employed by participants who judged their division as fair (t=6.22, p>.001) than participants who judged their division unfair. Moreover, participants who judged their division as unfair, were significantly more likely to employ equality, a moral justification, than those who judged their division as fair (t=12.75, p<.001).

Table 5.

Own Household Division Justifications by Fairness Judgments in Proportions by Country and Overall

Grouping	Justifications
	Participation	Equality	Equity	Perspectivism	Contractual
Country
China
Fair	.22(.40)	.21(.39)	.31(.43)	.05(.15)	00(0)
Not Fair	.06(.23)	.76(.33)	.02(.10)	.09(.19)	00 (0)
Korea
Fair	.03(.19)	.29(.43)	.33(.47)	.06(.47)	.16(37)
Not Fair	00(0)	.94(.21)	.02(.15)	00(0)	.02(.15)
Overall
Fair	.15(.34)	.24(.40)	.32(.44)	.05(.18)	.06(.24)
Not Fair	.02(.13)	.88(.27)	.02(.13)	.03(.12)	.01(.12)

Note. Standard deviation in parenthesis.

As a result of the significant interaction between fairness judgment and justification type, a repeated-measures ANOVA was run on justifications as a function of country, age, and family structure by fairness judgment. The analysis revealed a significant country and justification type interaction for those evaluating their family’s division as unfair, F(4,303)=4.15, p<.001, η²=.05. Unexpectedly (H2), however, pairwise comparison tests revealed that equality was significantly more likely to be employed as a justification by Korean participants than Chinese participants (t=2.37, p<.01). On the other hand, perspectivism (i.e., they should help) was significantly more likely to be employed by Chinese participants (t=2.92, p<.001).

Overall, the above results contradicted this study’s first two hypothesis. The majority of children did not judge either a hypothetical gendered division or their own family’s gendered division fair, and relied on mostly moral rather than conventional justifications. Furthermore, Korean children and children from traditional families, were more likely to consider their family’s division as unfair. No significant differences in fairness judgments by age or gender were found.

Evaluations of A Strictly Equal Division Between Parents

In support of Hypothesis 3, the great majority of children (98.5%) judged the second hypothetical scenario in which both parents share an equal division of household labor as fair. As a result of high participant agreement, no significant differences based on participant characteristics were found. Participants’ sweeping support of an equal division of household labor provides further evidence that individuals endorse an equal division of household labor (Gere & Helwig, 2012; Thomson, 2007). As expected, participants who judged an equal division as fair all employed equality as their justification (i.e., they both work therefore they should both do the same amount of housework).

Fairness Judgments Across Situations

Children’s judgments for a gendered hypothetical scenario (situation==1) and their own family’s division (situation==0) significantly differed from each other, LR χ² (1, N = 262) = 31.08, p =.00, Nagelkerke R²=.09 (See Table 6). Despite the majority of children reporting having a gendered division of labor in their own homes (68.7%), children were significantly less likely to find a gendered hypothetical scenario fair (fair==1), than their own family’s division, OR=.22, CI[.12, .38], Z=−5.33, p<.001. At the same time, children’s fairness judgments also significantly between their own family’s division (story==0) and a perfectly equal hypothetical division(story==1), LR χ² (1, N = 262) = 88.31, p =.00, Nagelkerke R²=.29. Children were significantly more likely to find the story of perfect equality fair than their own family’s division, OR=41.36, CI[12.51–136.71], Z=6.10, p<.001.

Table 6.

Children’s Fairness Judgments Across Situations (Percentages)

Situation	Evaluation
	Not Fair	Fair	Reasonable
Traditional Situation (1st)	.81(108)	.19(25)	00(0)
Perfect Equality (2nd)	.01(2)	.98(131)	00(0)
Own Family	.48(63)	.50(66)	.02(3)

Note. Frequency in parenthesis.

Discussion

This study’s findings provide important new insights into children’s social and moral judgments and reasoning regarding the gendered division of household labor. While the gendered nature of the division of household labor has been well-established (Lachance-Grzela & Bouchard, 2010; Oshio et al., 2013) this study was the first of its kind to directly investigate children’s reasoning regarding the gendered division of housework within the family. In light of this study’s exploratory nature, it is unsurprising that the many of the hypotheses were refuted. The novelty of some of this study’s findings provided important methodological implications for future research.

Contrary to this study’s first hypothesis, the majority of children were not supportive of a hypothetical gendered division of household labor. While previously, children have been found to be supportive of hypothetical scenarios in which mothers took on the dual-burden of childcare and work (Sinno & Killen, 2011), or were primarily responsible for the home (Brose et al., 2013; Schuette & Killen, 2009), it seems that at least in the case of a division of household labor, children find a gendered division of labor unfair, mainly based on expectations of equality (Gere & Helwig, 2012; Park et al., 2012; Thomson, 2007). This suggests that children’s reasoning about gender roles differs based on the type of activity they are asked to evaluate. Although both housework and childrearing are gendered activities, they differ in the clarity of the consequences for the actor. While housework is a task related activity in which differences in laboring can be clearly perceived, more global gender roles such as “homemaker” versus “breadwinner,” (e.g. Sinno & Killen, 2009), do not provide as clear outcomes regarding potential differences in labor or value. Therefore, in the case of housework, as a result of the clear inequality in outcomes, children may be less concerned about whether the family’s behavior is gender norm consistent, and prioritize concerns for equality.

Furthermore, Korean children and children from traditional families (i.e. where the mother did the majority of the housework) were more likely to judge their family’s division unfair. This finding contradicts the assumption that children’s “traditional” family structure (or community context) makes them more accepting or habituated to inequality directly (Sinno & Killen, 2011; Sinno et al., 2017). Instead, children’s observation of inequality at home appears to increase their likelihood of finding their family’s division unfair. However, this finding also raises the question about future behavior: if children in such communities and families are more likely to find the situation unfair, why do such practices continue?

In support of previous developmental findings (Rizzo & Killen, 2016; Schmidt, Svetlova, Johe, & Tomasello, 2016), the youngest participants (9–11) were significantly more likely than older participants to employ equality as a justification following a judgment of unfairness. On the other hand, older participants (16–18) were significantly more likely to argue that the gendered division of labor in the hypothetical scenario was unfair because both spouses were part of the same family and had agreed to create a family together. In other words, it seemed that with age, participants were more likely to rely on a conventional justification (family membership), than appealing to the moral principle of equality. It may be that as a result of a greater understanding of the role of convention in society (Midgette, Noh, Lee & Nucci, 2016; Turiel, 1977), older adolescents are more likely to rely on convention as a justification for how families should be organized, rather than strict principles of equality. Furthermore, this finding suggests that conventional reasoning can also be employed to critique inequality within the domestic sphere.

Moreover, as hypothesized (H3), in a scenario where all factors were held strictly equal, across nationality, gender, and age, participants preferred and found an equal division of labor fair. The near universality of children’s judgments of the fairness of an equal division, and their following justification that this situation was fair as a result of being equal, suggests that children were reading such a scenario as a straightforward moral issue (Nucci et al., 2017). This lends further support to findings that suggest that children do not believe that inequality should be maintained based on gender differences alone (Noh & Midgette, 2018). However, children’s judgment that equality is fair did not directly translate into them finding their household’s inequality unfair. This indicates the complex and apparent paradoxical nature of children’s fairness judgments. In the next section, it is argued that fairness judgments regarding inequality involves children’s consideration of not only the question of whether “is there inequality?” but also“how much inequality is there?”

How Much Inequality is Too Much?

As expected, the majority of Chinese and Korean children reported that their mothers spent on average more time doing housework (Oshio et al., 2013). Furthermore, only 6.1% of participants reported their family as having an equal division of household labor. Perceiving inequality, then, was not in itself sufficient cause to condemn it. Rather, children were evenly split on their fairness judgments regarding their household’s division of labor. Unexpectedly, however, Korean children, who were more likely to report that their mothers did more housework, and a greater disproportionate share of housework, were also found to be more likely to find their family’s division unfair. Based on these findings, it is suggested that rather than inequality itself, the proportion of inequality may be the most important factor in informing children’s evaluations of the fairness of gendered labor distribution (Coltrane, 2000).

Additionally, children’s justifications further suggest that the proportion of inequality is an important factor in moral reasoning. Investigation of children’s justifications following their fairness judgments revealed that rather than relying on gender stereotypes or social norms (Brose, Conry-Murray, & Turiel, 2013; Sinno & Killen, 2011), children relied on moral justifications for their acceptance of gender inequality in their homes. Children who judged their family’s division fair were significantly more likely to employ equity (i.e. the busier spouse does a little less) and participation (i.e. everyone does something), whereas children who judged the division unfair employed equality (i.e. one person is doing more than the other) as a justification. Moreover, Korean children, who perceived a more proportionate inequality, were more likely to employ equality, while Chinese children were more likely to employ participation. Therefore, while a small proportion of inequality may be justified, a greater disproportionate divide in household labor may be perceived as too unequal, and therefore unfair.

This study found that moral reasoning was employed in support of inequality, rather than only employed to be critical of it. This implies that moral reasoning is not in itself sufficient to support children’s developing critique of society’s social inequalities. This occurred because children did not connect their acceptance of their family’s “smaller” inequality (i.e. my mother does a little more, because she has a little more free time), with larger structural inequalities (i.e. women face employment discrimination). Chinese children, for example, who reported their mothers as doing a less disproportionate unequal division of labor, were more likely to find their family’s division fair. However, when the hours that Chinese women work on paid labor and household labor are combined, they actually work an additional 267.7 hours more a year than Chinese men, and 249.1 more hours than Korean women (OECD, 2008/ 2014). Therefore, Chinese children’s acceptance of their family’s “smaller” inequalities, actually supports a societal practice in which women must labor more overall, To judge the fairness of the gendered division of household labor in the home, then, children also need to understand how gender inequality is connected to societal organization. Therefore, for children to challenge current societal inequalities, they must also understand how individual behavior connects to the larger society in which they live.

This study found children to be less critical of their own family’s inequality than inequality presented in a hypothetical situation. This suggests possible qualitative differences between how children judge gender inequality in a hypothetical scenario versus how they judge the issue in their own lives. However, much of social domain theoretical research has mainly employed hypothetical scenarios to evaluate children’s fairness and acceptability judgments regarding gendered practices (Conry-Murray et al., 2015; Killen, Lee-Kim, McGlothin & Stangor, 2002; Schuette & Killen, 2009; Sinno & Killen, 2011; Turiel, 2008). Future research should endeavor to investigate not only children’s evaluations regarding hypothetical scenarios, but also children’s evaluations of their own actual experiences. It may be the case that children are less willing to be critical of social practices that they experience than they are of the hypothetical situations researchers present to them.

Currently, it is not standard practice for social domain researchers to analyze the interaction between fairness judgment and justification (Mulvey & Killen, 2016). This study’s findings suggest the value and importance in analyzing the relationship between children’s fairness judgments and their justifications, as well as how they may interact with other factors such as country and age. For example, this study found that Korean children were more likely than Chinese children to employ equality following their judgments that the division was fair. Therefore it is likely that previous research children’s moral reasoning that has not investigated the possible interaction between fairness judgments and justification may be masking or missing other forms of interactions. Future research into children’s social and moral reasoning should consider investigating the interaction between individuals’ judgments and their justification usage.

These findings have several implications. The fact that previous research with American children has found them to be more accepting of gender unequal roles (Sinno & Killen, 2009; 2011), may be as a result of their experiencing less disproportionate inequality (OECD, 2017), rather than because they may be more accepting of gendered roles (i.e. where disproportionate inequality may occur) in general. Furthermore, rather than asking why inequality may be considered fair, this study’s findings suggest that a better question may be to ask “how much inequality?” is considered fair. Comparison between a gendered hypothetical scenario with great inequality, a hypothetical perfectly equal division, and children’s own family division, suggests that while equality is considered fair and extreme inequality tends to be considered unfair, certain proportions of inequality are considered fair. Future research should investigate the limits and ranges of proportionate inequality that children would accept as fair, along with the justifications provided in support of unequal distributions. Moreover, children’s acceptance of inequality may be as a result of a disconnect between their family’s behavior and larger structural inequalities. Future research should investigate how understanding structural inequality may improve children’s moral critique of inequality at the individual and family level.

Limitations.

This study had a few limitations. This study had a limited sample size, and therefore was limited in being able to detect small effects. Future research should consider replicating this study with a larger sample size. In addition, this study was limited to children in professional white-collar two-parent families in urban centers in China and South Korea. However, it is possible that children in contexts with disparate proportions of inequality may be more or less sensitive to their family’s division. Future research should investigate families in different cultural contexts and with different formations and consider possible class and rural-urban differences in moral judgments and reasoning. Moreover, the study did not establish coder reliability across countries, and therefore it is possible that certain concepts do not translate across cultures. Future research should investigate in further detail how children in each culture comes to define concepts such as fairness, equality, and equity. Furthermore, while this present study investigated differences in judgments and justifications across hypothetical and actual experiences, only one scenario described a gendered division of labor. Future research should investigate consistency and variance in fairness judgments and justifications across multiple hypothetical scenarios and multiple situations that children face in their daily lives. There is a need to further investigate how children reason about their own lives as this may provide greater insight into how children understand and come to challenge or accept social inequalities.

Conclusion.

In conclusion, this study set out to investigate how children reason about the gendered division of household labor. Interestingly, as found with previous adult cross-cultural studies(Braun, Lewin-Epstein, Stier, & Baumgärtner, 2008; Greenstein, 2009), Chinese and Korean children were evenly divided regarding whether their family division was considered fair. Future research should investigate the relationship between children’s and their parents’ judgment and reasoning about the gendered division of labor, as well as the relationship between parenting practices, parental division, and larger cultural context and children’s social and moral reasoning.

Appendix

Scenario 1)

A husband and wife both work as teachers. The wife is an elementary school teacher and the husband is a middle school teacher. They have an 8 year old son. The husband sometimes has to stay up long hours grading homework. Everyday after they both come home from work the husband goes to his office to grade the homework and the wife makes dinner. After dinner, the wife cleans the dishes and the husband plays with their son. After this he prepares to go to bed. The husband, tired from teaching and grading homework decides that he will do housework on the weekends, and unless he remembers to do it, he will only do it if his wife asks for help.

Scenario 2)

A husband and wife both work in the same company. They work everyday from 8am-5pm. They have a 5 year-old daughter. The husband and wife decided that they would share the housework by each person doing the housework every other day. On Mondays the wife drops off and picks up their daughter from school and the husband goes grocery shopping and cooks dinner. While the husband puts their daughter to bed the wife cleans the dishes and clears the table. On Tuesdays the husband drops off and picks up their daughter from school and the wife makes dinner and puts their daughter to bed. On Wednesday they switch again. During the weekends they eat out so no one has to cook dinner and they take turns taking out the trash.