Skip to main content
NCBI home page
Philosophical Transactions of the Royal Society B: Biological Sciences logoLink to Philosophical Transactions of the Royal Society B: Biological Sciences
. 2021 May 3;376(1827):20200025. doi: 10.1098/rstb.2020.0025

Childcare support and child social development in Japan: investigating the mediating role of parental psychological condition and parenting style

Masahito Morita 1,✉, Atsuko Saito 2,✉, Mari Nozaki 3, Yasuo Ihara 1
PMCID: PMC8090826  PMID: 33938269

Abstract

In humans, support from partners and alloparents is crucial for successful child-rearing and optimal child development. However, the complex relationships among childcare support, children's outcomes and parental characteristics have not been fully examined. We investigate how three sources of partner and alloparental support—partner's childcare participation, support from children's grandparents and support from non-kin—can be associated with child social development. We hypothesize that the associations between childcare support from partners/alloparents and child social development are partly mediated by parental psychological condition and parenting style. To test this, we conducted path analyses on online survey data collected in 2016 from parents of 3- to 5-year-old children in Japan. We found no evidence that childcare support had direct positive effects on child social development. Rather, the benefit of childcare support was mediated by its effects on parental psychological condition and parenting style, which in turn improved children's outcomes. At the same time, we found some evidence that greater availability of childcare support was directly associated with more behavioural difficulties in children. Our findings reveal the complex pathways between childcare support, parental characteristics and children's outcomes in Japan, showing potential mechanisms behind parental and alloparental effects in industrialized populations.

This article is part of the theme issue ‘Multidisciplinary perspectives on social support and maternal–child health’.

Keywords: childcare participation, grandparent support, non-kin support, child quality, parent quality, parenting strategy

1. Introduction

(a). Background

In humans, children are often cared for by parents as well as other individuals, known as alloparents. Alloparents can include a broad range of individuals, such as grandparents, elder siblings, relatives and non-kin, who may provide childcare support (i.e. investment that will improve children's outcomes and parental characteristics). This communal way of raising children has been termed cooperative breeding ([1–3], see also [4]). Across non-industrialized, small-scale societies (i.e. hunter–gatherers, horticulturalists, and historical and contemporary nation states in pre-industrial socioeconomic contexts), the presence of kin, particularly grandmothers and older siblings, has been associated with greater child survival [5,6]. However, there is notable variation across populations, with kin presence also associated with null or negative effects on child survival [6], suggesting that kin support and its effects may be context-dependent and diverse.

While literature from non-industrialized societies focuses on child survival and mortality, in contemporary industrialized, high-income societies, researchers have paid increasing attention to the potential effects of alloparents on child cognitive, emotional and behavioural development [7–13]. This is partly because improved hygiene, nutrition states and medical services have raised the standard of children's physical health, making alloparental care less relevant for survival. Furthermore, in post-demographic transition, low-fertility-low-mortality societies, parental investment strategies prioritize quality rather than the quantity of children (i.e. prioritizing child capital over a number of children) [14–16]. In these settings, alloparenting may have more salient effects on child quality (over survival), including on social development which broadly captures children's cognitive and language ability, educational outcomes, emotional stability, and social behaviour and adjustment. These traits are known determinants of future outcomes including better health and greater income (e.g. [17–19]), indicating that it captures aspects of child quality/capital. Sadruddin et al.'s [12] systematic cross-cultural review has found that multigenerational care involving parents and grandparents is positively associated with child cognitive development and educational attainment in several cases, suggesting that kin support, particularly from grandparents, continues to be an important determinant of successful child-rearing in industrialized populations.

Understanding alloparental support in industrialized settings is complicated by the fact that parents also receive a variety of formal childcare support from professionals and institutionalized services [2], along with informal support from the wider community. While evolutionary theory predicts that, all being equal, kin are more likely to help (leading to much focus on kin as important caregivers), some studies highlighted the importance of non-kin support in non-industrialized populations (e.g. [20,21]), suggesting that support from non-kin may continue to be important in industrialized contexts. Indeed, Yamaguchi et al. [13] found that the use of childcare centres (such as nurseries) was associated with improved child developmental outcomes, such as better language development and reduced tendency of inattention, hyperactivity and aggression, among Japanese mothers especially with lower education levels.

When considering how parents and alloparents impact child quality, one must be mindful of the indirect effects: how childcare support influences parental condition (proxy of parent quality) and parenting style (proxy of parenting strategies), which then goes on to affect children's outcomes [2,22]. It can be expected that better parental condition gives people the ability to meet local cultural norms of parenting, and adjusting their parenting style (i.e. quality and quantity of parental investment) to the norms may be associated with successful child-rearing, higher child quality and higher fitness. While ‘good’ or ‘bad’ parenting style is subjective and depends on each sociocultural environment, the mediating role of parenting style should exist across different norms. Numerous studies have investigated associations between social support and parental psychological condition and/or parenting style in contemporary industrialized societies (e.g. [10,13,23–26]). For example, Yamaguchi et al. [13] found that parental subjective well-being and parenting quality (measured here as parenting style with greater warmth and fewer abusive attitudes) were higher with the use of childcare facilities among Japanese mothers especially with lower levels of education. Given the known effects of parental psychological condition and parenting style on children's outcomes [10,27–29], the indirect effects of childcare support may be significant determinants of child development and quality. For a related study, among Hungarian mothers, the negative effects of maternal postpartum depression on children's emotional warmth and invasive behaviour towards mothers were weakened with greater perceived social support [10].

(b). Hypothesis and scope

We hypothesize that the association between partner/alloparental childcare support and child social development is partly mediated by parental psychological condition and parenting style (figure 1; see also the electronic supplementary material for the directions of pathways). Specifically, we predict that childcare support is associated with better parental psychological conditions (lower parental stress and depression), parenting style adjusted to local cultural norms, and higher social development in children (more prosocial behaviour and fewer behavioural difficulties). For the current study, we focus on three sources of childcare support: partner's childcare participation; availability of support from children's grandparents (actual/potential practical/emotional/financial support from children's grandparents) and availability of support from non-kin including community networks and professional care supporters (actual/potential practical/emotional support from non-kin). As socio-cognitive ability dramatically develops in early childhood (e.g. executive function: [30]; theory of mind: [31]) along with increases in prosocial behaviour [32], we focus our analysis on parents of 3- to 5-year-old children.

Figure 1.

Figure 1.

Open in a new tab

Supposed associations between related factors.

We test these associations by conducting path analyses on online survey data from Japan, collected in 2016. Previous studies on childcare support and its effects on children's outcomes and parental characteristics have tended to focus on Western populations. Japanese household compositions are similar to those of the European Union and other high-income countries [33], with decreasing numbers of three-generation households and ‘nuclear families’ being the norm [34]. However, gender norms around childcare being in the female domain remain strong partly evidenced by notably lower childcare participation by fathers (e.g. [35,36]). In a recent national survey, 35.0% of respondents agreed that husbands should work in employment and wives should take care of the household, and of these respondents who agreed to the gender role, 55.2% thought that households managed by wives will be better for child development [37]. Unsurprisingly, many mothers in Japan struggle to manage household chores and childcare alone, even if they work outside the home (e.g. [38]). Consequently, the hypothesized pathways among childcare support, child social development, parental psychological condition and parenting style could vary between mothers and fathers.

Extensive sociological studies on childcare support in Japan have outlined the importance of social support in the child-rearing system (e.g. [11,39–45]). For example, Matsuda [11] found that the size and composition of mothers' social networks were associated with child social development (e.g. self-control and language ability), both directly and indirectly, where the effects were mediated by maternal child-rearing anxiety. In another study, Matsuda [42] showed that a moderate level of diversity in childcare networks involving kin and non-kin was associated with better well-being among Japanese mothers, suggesting that childcare support from within and beyond the family remains important in contemporary Japan. In Japanese law and policy, however, childcare has traditionally been considered to be a private matter for each family (e.g. [46]), which indicate that support from the government and non-kin may have been still underdeveloped.

Previous studies, therefore, indicate that partners and alloparents are key determinants of child quality in Japan. However, researchers have not fully examined the hypothesized causal pathways between childcare support, children's outcomes and parental characteristics. Here, we build on previous evidence by examining how childcare support from partners and alloparents is directly and/or indirectly associated with child social development (a proxy of child quality), parental psychological condition (a proxy of parent quality) and parenting style (a proxy of parenting strategies in the environment). Overall, our study will provide further details on the mechanisms underpinning parental and alloparental effects in a non-Western, contemporary industrialized population.

2. Subjects and methods

(a). Data

We analysed data from an online questionnaire survey conducted by two of the authors (AS and MN) in Japan, in March 2016. The participants were registered as possible respondents at Macromill, Inc., a public marketing research company in Japan (https://monitor.macromill.com). Parents living with at least one child aged from 3 to 5 years old were invited to take part via notification from the company (such as through e-mail and the website) which included survey information. The survey had a broad purpose to exploratorily obtain basic information on childcare support and its effects on parents' and children's outcomes in Japan. For the present study, we selected available variables relating to childcare support, child social development, parental psychological condition and parenting style, and build on previous analyses using the same data; Saito & Nozaki [47] previously examined the association between maternal employment and their availability of childcare support, and Saito & Nozaki [48] explored the association between paternal childcare participation and maternal employment, and the correlation between maternal and paternal availability of childcare support.

The participants of the survey were 618 heterosexual couples (mothers and fathers) who had at least one child aged 2 to 6 years old, although our survey was advertised for parents with 3- to 5-year-old children. Each couple received a single online questionnaire composed of two sections, one for the mother and the other for the father, which had the same set of questions. Mothers and fathers were requested to answer separately. We explained to them the aim and contents of the survey and obtained their permission before they began to respond to the questions. This survey was approved by the ethical committees at the University of Tokyo (no. 306) and Musashino University (no. 29001). The dataset that we analysed in this paper and the questionnaire (in Japanese) are available in the electronic supplementary material (see Data accessibility).

(b). Variables

(i). Childcare support

The following three variables were designed by the authors with reference to previous social surveys in Japan [42,49–53]. Partner's childcare participation was derived based on the frequency of various caregiving activities, measured through information such as ‘How often do you give your child a bath?’ and ‘How often do you help your child to eat?’, with a possible minimum score of 9 and a maximum score of 36. We take these measures to capture partner's direct investment levels into the child, where higher scores indicate higher practical participation in childcare by the partner. Mothers and fathers provided information about their own participation by themselves respectively. Support from children's grandparents was measured via both actual and potential grandparental childcare involvement, through questions such as ‘How often does your child meet with his/her grandparents?’ and ‘Can you rely on your child's grandparents regarding childcare?’, with a possible minimum score of 0 and a maximum score of 88. We take these measures to capture grandparents' direct and indirect investment levels into the child and parent, where higher scores reflect greater availability of practical, emotional and financial childcare support from children's grandparents. Support from non-kin was measured via both actual and potential parental involvement in a broad range of childcare support from across the parental social networks, including formal and informal sources. Examples of questions are ‘How often do you use childcare centres?’ and ‘How many persons can you talk about childcare concerns with?’, with a possible minimum score of 0 and a maximum score of 32. We take these measures to capture non-kin's direct and indirect investment levels into the child and parent, where higher scores reflect greater availability of practical and emotional childcare support from parent's social networks and professional services. Further information on survey items are available in the electronic supplementary material.

(ii). Child social development

We used the Strengths and Difficulties Questionnaire (SDQ; [54], see [55] for the Japanese version) to measure child social development. The SDQ measures two components of child social behaviour and adjustment, where prosocial behaviour scores (PBS; possible minimum score 0 and a maximum score 10) capture children's prosocial tendencies, and total difficulties scores (TDS; possible minimum score 0 and a maximum score 40) capture children's emotional symptoms, conduct problems, hyperactivity/inattention and peer problems. Higher SDQ-PBS are assumed to indicate higher social development, and higher SDQ-TDS are assumed to indicate lower social development in children. The SDQ is based on parent reports and the correlations (Pearson's r) of the scores assessed by the mother and father within each couple were relatively high (0.68344 in SDQ-PBS and 0.77225 in the SDQ-TDS).

(iii). Parental psychological condition

We focused on measures of parenting stress and parental depression. We used the short form of the Parenting Stress Index based on 36 items (PSI; [56], see also [57] for the related Japanese version) to assess parenting stress. The PSI measures parental distress (PD), along with child difficulty and parent–child dysfunctional interaction, through the parent's self-assessment. The present study used the PSI-PD only (see the electronic supplementary material). With a possible minimum score of 12 and a maximum score of 60, we take higher scores to indicate a higher parenting stress. Parenting stress in the context of child-rearing will be partly associated with broader components of general psychological condition of parents. We used a form of the Beck Depression Inventory (BDI-II; [58], see [59,60] for the Japanese version) to assess parental depression and to analyse it as a continuous condition (see the electronic supplementary material). With a possible minimum score of 0 and a maximum score of 63, we take higher scores to indicate higher parental depression. The BDI-II is based on parent's self-assessment.

(iv). Parenting style

We focused on measures of parental care and child abuse. We used the Parental Bonding Inventory (PBI; [61]; see [62] for the Japanese version) to capture parental care. The PBI measures Care (warmth) and over-protection in terms of the parent–child bonding, as per the parent's self-assessment. The present study used the PBI-Care only (see the electronic supplementary material). With a possible minimum score of 0 and a maximum score of 30, we take higher scores to indicate greater parental care. We used a validated measure of potential risk for child abuse made in Japan [63]. This assesses risk for physical and psychological abuse that will be associated with poor parenting practices. The current measure includes questions such as ‘I do not spank my child’, ‘I often play with my child’ and ‘I do not obey my child when he/she complains' (we translated the original questions of [63] in Japanese). These items capture parental child-rearing attitudes that (i) do not rely on power, (ii) nurture children's high self-esteem, and (iii) teach children self-control; parents' violent behaviour, cold attitudes and obeying children's selfish behaviour are considered as potential risk factors for child abuse. Each item is assessed by the parents on a five-point scale. In our analysis, with a possible minimum score of 15 and a maximum score of 75, we take higher scores to indicate higher potential risk of child abuse.

(v). Control variables

We used information of household income (parents' combined personal income within a couple; 27 levels possible, from low to high; ordinal), parents’ education level (five levels possible, from low to high; ordinal), employment status (0 = none including homemaker; 1 = temporary; 2 = full-time; ordinal), age (years old; integer), children's age (numerical, days after birth divided by 365), sex (1 = male; 2 = female) and the number of siblings of the child in question (persons, up to five). These variables were included to control the effects of possible confounding factors for parental psychological condition and child social development (see the electronic supplementary material). The mothers and fathers provided information on the variables individually. The variable selection was performed before conducting the analysis.

(c). Statistical analyses

We conducted path analyses and generalized linear models to test the hypothesized causal pathways by using R version 3.5.1 [64] for macOS Sierra mainly. We excluded the couples who had at least one not applicable (N/A) items among the aforementioned variables. N/A items included missing or ‘unreliable’ data (e.g. the age difference between mothers and children/siblings was less than 14 or more than 49 years old; parents responded that there was a first and third child but no second child; and/or the provided sibling's birth date was a future date from the date of response). We also excluded the couples none of whose children were between 3- to 5-years old, because they were out of the subjects of the original survey. The number of N/A samples was 95 couples in total. Therefore, the final sample size for analyses was 523 couples. We analysed the mothers' and fathers’ data separately, except for the childcare participation of one's partner; when we analysed the mothers' data, fathers' (their partners’) childcare participation (this information was provided by fathers themselves) was included in the statistical models to investigate its effects on mothers' perceived outcomes. More details of the analyses including model selection procedures are provided in the electronic supplementary material.

3. Results

We show the descriptive statistics of participants used in the analyses (n = 523 couples) in table 1. The distributions of variables on childcare support, parental psychological condition, parenting style and child social development are shown in the electronic supplementary material, figure S1. The correlations (r) between the above variables are also shown in the electronic supplementary material, table S1. Note that the following analyses have estimated the strengths of possible effects between variables within the assumed causal framework.

Table 1.

Descriptive statistics of participants used in analyses: (a) mothers; (b) fathers and (c) children. (Possible range of measurements on childcare support, parental psychological condition, parenting style, and child social development: partner's childcare participation = 9–36, support from children's grandparents = 0–88, support from non-kin = 0–32, PSI-PD = 12–60, BDI-II = 0–63, PBI-Care = 0–30, abuse = 15–75, SDQ-PBS = 0–10, SDQ-TDS = 0–40.)

variable (a) mothers
 (b) fathers
range mean s.d. n range mean s.d. n
age 20–48 35.057 4.881 — 25–60 37.031 5.607 —
personal incomea 0–10 1.088 1.856 — 0–13 4.488 2.561 —
education level 1–5 3.140 0.867 — 1–5 3.321 1.011 —
employment status
 none including homemaker — — — 322 — — — 42
 temporary — — — 104 — — — 17
 full-time — — — 97 — — — 464
partner's childcare participation 9–36 20.474 7.353 — 10–36 31.260 5.392 —
support from children's grandparents 0–80 27.922 17.016 — 0–80 27.560 17.102 —
support from non-kin 0–31 11.891 6.589 — 0–32 7.168 6.501 —
stress (PSI-PD) 12–60 32.017 9.420 — 12–55 27.748 8.781 —
depression (BDI-II) 0–61 12.727 10.100 — 0–62 8.337 9.230 —
care (PBI-Care) 10–30 22.208 5.040 — 2–30 22.034 5.460 —
abuse 15–64 32.792 9.186 — 15–75 33.090 10.340 —
prosocial behaviour (SDQ-PBS) in children 0–10 6.166 2.496 — 0–10 5.967 2.607 —
total difficulties (SDQ-TDS) in children 0–33 11.033 6.083 — 0–34 11.415 6.157 —
variable (c) children
range mean s.d. n
age 3.025–5.997 4.531 0.878 —
sex
 boy — — — 250
 girl — — — 273
number of siblings of the subject child 0–5 1.090 0.754 —
Open in a new tab

ahousehold income (parents' combined personal income within a couple): range = 0–22, mean = 5.576, s.d. = 3.450.

Key results from our path analyses are summarized in figure 2 and table 2. For mothers, the direct path from partner's childcare participation to SDQ-TDS was statistically significant (p = 0.03358, standardized estimate = 0.08077, s.e. = 0.03092). However, against our prediction (i.e. childcare support will improve children's outcomes), the estimate was positive in value, meaning there was an association between higher childcare participation by fathers and greater behavioural difficulties in children. Greater availability of support from children's grandparents was related to lower potential risk of maternal abuse (p = 0.03066, standardized estimate = –0.09063, s.e. = 0.02264). Greater availability of support from non-kin was associated with lower levels of maternal stress and depression (lower PSI-PD and BDI-II scores; table 2). In turn, lower maternal stress and depression were related to higher social development in children (higher SDQ-PBS and lower SDQ-TDS; table 2). These associations were both direct and indirect via maternal parenting style (PBI-Care) for maternal stress, while indirect only (via PBI-Care and abuse) for maternal depression. Overall, there were no significant direct paths that would support our prediction between the availability of childcare support provided to mothers and child social development, whereas there were some indirect paths mediated by maternal psychological condition and parenting style.

Figure 2.

Figure 2.

Open in a new tab

Results of path analyses: (a) mothers and (b) fathers. Non-significant paths (ps ≥ 0.05), covariances or error terms were not shown. Standardized estimates of significant paths are shown, and bold arrows indicate positive coefficient values and normal arrows represent negative coefficient values. *p < 0.05; **p < 0.01 and ***p < 0.001.

Table 2.

Results of path analyses: (a) mothers and (b) fathers. (AIC, Akaike information criterion; GFI, goodness of fit index; AGFI, adjusted goodness of fit index; RMSEA, root mean square error of approximation; NFI, normed fit index; CFI, comparative fit index; SRMR, standardized root mean square residual.)

path (a) mothers
 (b) fathers
estimate standardized estimate s.e. z p estimate standardized estimate s.e. z p
partner → PSI-PD –0.05297 –0.04156 0.05416 –0.97807 0.32804 –0.10464 –0.06459 0.07129 –1.46784 0.14215
partner → BDI-II 0.00260 0.00193 0.05810 0.04476 0.96430 –0.13463 –0.08211 0.07119 –1.89111 0.05861
partner → PBI-Care 0.03401 0.04962 0.02858 1.19012 0.23400 0.09643 0.09524 0.04145 2.32635 0.02000
partner → abuse 0.02581 0.02066 0.05238 0.49271 0.62222 –0.19926 –0.10388 0.07804 –2.55333 0.01067
partner → SDQ-PBS 0.00596 0.01789 0.01308 0.45540 0.64882 0.00270 0.00566 0.01934 0.13948 0.88907
partner → SDQ-TDS 0.06572 0.08077 0.03092 2.12507 0.03358 –0.00844 –0.00744 0.04397 –0.19184 0.84787
grandparents → PSI-PD –0.03783 –0.06866 0.02336 –1.61929 0.10539 –0.00539 –0.01055 0.02231 –0.24146 0.80920
grandparents → BDI-II –0.03719 –0.06385 0.02506 –1.48397 0.13782 –0.02869 –0.05550 0.02228 –1.28784 0.19780
grandparents → PBI-Care 0.01638 0.05528 0.01235 1.32605 0.18482 0.01794 0.05619 0.01294 1.38607 0.16572
grandparents → abuse –0.04893 –0.09063 0.02264 –2.16147 0.03066 –0.03404 –0.05628 0.02436 –1.39702 0.16241
grandparents → SDQ-PBS –0.00150 –0.01045 0.00566 –0.26581 0.79039 0.00220 0.01464 0.00601 0.36593 0.71442
grandparents → SDQ-TDS –0.00462 –0.01313 0.01337 –0.34515 0.72998 –0.00911 –0.02547 0.01366 –0.66652 0.50508
non-kin → PSI-PD –0.32998 –0.23195 0.06044 –5.45933 4.77942 × 10−8 –0.00335 –0.00249 0.05913 –0.05670 0.95478
non-kin → BDI-II –0.25863 –0.17198 0.06484 –3.98900 6.63522 × 10−5 0.18209 0.13389 0.05905 3.08368 0.00204
non-kin → PBI-Care 0.01835 0.02399 0.03284 0.55892 0.57622 0.04808 0.05724 0.03461 1.38917 0.16478
non-kin → abuse 0.02443 0.01753 0.06018 0.40601 0.68473 0.01335 0.00839 0.06515 0.20485 0.83769
non-kin → SDQ-PBS 0.01359 0.03655 0.01499 0.90622 0.36482 –0.00346 –0.00875 0.01609 –0.21501 0.82976
non-kin → SDQ-TDS –0.00899 –0.00990 0.03544 –0.25374 0.79970 0.07847 0.08344 0.03657 2.14597 0.03188
PSI-PD → PBI-Care –0.08092 –0.15049 0.02594 –3.11938 0.00181 –0.18665 –0.29867 0.02728 –6.84094 7.86728 × 10−12
PSI-PD → abuse 0.07488 0.07641 0.04755 1.57490 0.11528 0.35494 0.29980 0.05137 6.91009 4.84339 × 10−12
PSI-PD → SDQ-PBS –0.02997 –0.11473 0.01194 –2.50974 0.01208 –0.02034 –0.06917 0.01335 –1.52423 0.12745
PSI-PD → SDQ-TDS 0.06620 0.10372 0.02823 2.34486 0.01903 0.08537 0.12199 0.03034 2.81394 0.00489
BDI-II → PBI-Care –0.09599 –0.18871 0.02418 –3.96926 7.20949 × 10−5 –0.06849 –0.11090 0.02732 –2.50661 0.01219
BDI-II → abuse 0.21691 0.23399 0.04432 4.89363 9.89937 × 10−7 0.14646 0.12520 0.05144 2.84743 0.00441
BDI-II → SDQ-PBS 0.01702 0.06888 0.01129 1.50743 0.13170 0.01827 0.06288 0.01277 1.43106 0.15241
BDI-II → SDQ-TDS 0.00061 0.00100 0.02670 0.02270 0.98189 0.06952 0.10053 0.02903 2.39498 0.01662
PBI-Care → SDQ-PBS 0.17721 0.36474 0.02553 6.94068 3.90222 × 10−12 0.15758 0.33487 0.02484 6.34380 2.24168 × 10−10
PBI-Care → SDQ-TDS –0.33333 –0.28083 0.06035 –5.52276 3.33710 × 10−8 –0.28310 –0.25281 0.05647 –5.01332 5.34979 × 10−7
abuse → SDQ-PBS –0.02296 –0.08610 0.01393 –1.64784 0.09939 –0.02141 –0.08618 0.01319 –1.62246 0.10471
abuse → SDQ-TDS 0.14925 0.22915 0.03293 4.53229 5.83468 × 10−6 0.09516 0.16099 0.03000 3.17256 0.00151
PSI-PD ↔ PSI-PD 82.47321 0.93854 5.10496 >16.15540 <1.03890 × 10−58 75.99815 0.99575 4.70417 >16.15540 <1.03890 × 10−58
BDI-II ↔ BDI-II 94.90061 0.96639 5.87420 >16.15540 <1.03890 × 10−58 75.78957 0.96958 4.69126 >16.15540 <1.03890 × 10−58
PBI-Care ↔ PBI-Care 22.90765 0.90149 1.41795 >16.15540 <1.03890 × 10−58 25.48667 0.85502 1.57759 >16.15540 <1.03890 × 10−58
abuse ↔ abuse 76.95124 0.91188 4.76316 >16.15540 <1.03890 × 10−58 90.32724 0.84438 5.59112 >16.15540 <1.03890 × 10−58
SDQ-PBS ↔ SDQ-PBS 4.76426 0.79429 0.29490 >16.15540 <1.03890 × 10−58 5.46995 0.82866 0.33858 >16.15540 <1.03890 × 10−58
SDQ-TDS ↔ SDQ-TDS 26.62280 0.74367 1.64791 >16.15540 <1.03890 × 10−58 28.26855 0.75627 1.74978 >16.15540 <1.03890 × 10−58
partner ↔ non-kin 3.05178 0.06298 2.12499 1.43614 0.15096 –4.25073 –0.12127 1.54542 –2.75053 0.00595
PSI-PD ↔ BDI-II 40.47934 0.43576 4.25827 9.50606 1.98016 × 10−21 28.09115 0.36369 3.54202 7.93082 2.17707 × 10−15
PBI-Care ↔ abuse –26.17975 –0.56535 2.16563 –12.08876 1.21108 × 10−33 –27.71597 –0.49082 2.42525 –11.42809 3.02679 × 10−30
SDQ-PBS ↔ SDQ-TDS –2.90102 –0.19797 0.50903 –5.69916 1.20400 × 10−8 –1.83494 –0.11682 0.55016 –3.33530 0.00085
χ2 = 0.2114, d.f. = 2, p = 0.8997, AIC = 86.2114 χ2 = 1.1386, d.f. = 2, p = 0.5659, AIC = 87.1386
GFI = 0.99991, AGFI = 0.99798, RMSEA = 0, NFI = 0.99975, CFI = 1, SRMR = 0.00323 GFI = 0.99952, AGFI = 0.98911, RMSEA = 0, NFI = 0.99846, CFI = 1, SRMR = 0.00743
Open in a new tab

For fathers, availability of support from non-kin had a significant direct effect on SDQ-TDS, but in the opposite direction to our prediction, with greater non-kin support associated with more behavioural difficulties in children (p = 0.03188, standardized estimate = 0.08344, s.e. = 0.03657). Greater availability of non-kin support was related to higher paternal depression (higher BDI-II; p = 0.00204, standardized estimate = 0.13389, s.e. = 0.05905), though the non-kin support was associated with lower depression among mothers. There were no significant direct paths that would support our prediction from childcare support to child social development as with mothers, while there were some indirect paths through paternal parenting style (e.g. mothers' greater participation in childcare was related to lower risk of paternal abuse, predicting greater behavioural difficulties in children). Among fathers, our analysis found no significant effects of grandparent support on both children's outcomes and parental characteristics; among mothers, however, greater availability of grandparent support predicted a lower risk of maternal abuse.

4. Discussion

We have estimated the effects of several childcare variables on both children and parents, and shown their possible relationships (figure 2). In line with our hypothesis, we found that positive associations between childcare support and child social development were mediated by parental psychological condition and parenting style. For example, among mothers, greater availability of support from non-kin was associated with lower levels of maternal stress and depression, which were related to parenting style with greater parental care and the lower potential risk of abuse, and ultimately predicting more prosocial behaviour and fewer behavioural difficulties in children. This suggests that the availability of partner and alloparental childcare support influences parent quality and parenting strategies, which in turn affects child quality in a specific sample in contemporary Japan. This means that the indirect positive associations between childcare support and child social development may be stronger, than the direct associations, in industrialized populations with nuclear family structures, where child-rearing practices tend to centre on parenting. This is also in line with previous studies finding positive effects of alloparents on children's outcomes [6,12] and reveals the potential mechanisms behind how partner and alloparental childcare support impacts child social development. In our analysis, greater parental care and lower risk of abuse were associated with higher social development in children. This result may indicate that these traits are components of a ‘good’ parenting style in a Japanese context.

Contrary to our prediction, we found evidence of direct effects between childcare support and increased behavioural difficulties in children. If this association is causal, this indicates that certain kinds of support may negatively impact children's outcomes. Indeed, some previous studies have found that non-maternal caregiving, such as the use of formal childcare and grandparenting, was associated with lower social development in children. However, the results are varied and inconsistent and may depend on the quality of care [7–9,12,13]. As child-rearing norms in Japan are still predominantly focused on mothers as caregivers (e.g. [38]), it might be that the quality of non-maternal childcare support does not adequately substitute maternal care in the environment (see also below), thereby leading to children's negative outcomes. Alternatively, our findings indicate reverse-causal associations, where more behavioural difficulties induce greater childcare support (see also [65,66]). The results may also reflect difficult environments for childcare in contemporary Japan, where parents in nuclear families tend to be overburdened with parenting responsibilities, and young children have few opportunities to interact with adults other than their parents or staff at childcare facilities (see also [67]). In such difficult situations, support may be associated with negative impacts on children's outcomes because help may be provided when families are under stress and greater support is needed (see also [20]).

Next, we note some differences in the pathways between mothers and fathers. As an example, childcare support from non-kin was associated with lower depression among mothers, whereas non-kin support predicted higher depression among fathers. This may reflect gender differences in the sociocultural conditions of childcare in Japan. As cultural norms and social structures tend to assume mothers to be principal caregivers, fathers may experience greater barriers and stress when accessing childcare support (e.g. diaper changing beds have often not been installed in toilets for men, see [68]), and this may be connected to lower childcare participation by fathers in childcare (see also [35,36]). Under this sociocultural context, providing childcare support may worsen paternal mental health and children's outcomes, which may partly explain why higher fathers' participation in childcare was associated with greater behavioural difficulties in the mother model (but fathers’ childcare participation did not take account of paternal psychological condition). Extensive investigations on fatherhood in Japan and the effects of fathers' characteristics on maternal condition and behaviour are required to fully understand the relationship.

Although our study has revealed the complex pathways between childcare support, parental psychological condition, parenting style and child social development, there are several limitations. First, maternal and paternal models were run separately, meaning we did not explore how broader maternal and paternal characteristics and experiences affect each other (e.g. how paternal condition and behaviour affect mothers’ perception of child social development, or how mothers' perceived childcare support influences paternal condition and behaviour). Second, as a cross-sectional survey, we do not know if the current associations would vary over time as children age (see also [69,70]). Third, while our findings support the hypothesized causal pathways, the associations identified in our models may represent reverse-causal mechanisms, or alternative causal paths may be assumed by other theoretical frameworks (e.g. [27,71]). Fourth, we need to keep in mind that family dynamics can vary even within countries depending on the local environment. In Japan, nuclear family households are more common in urban compared to rural areas [44,72], which may alter the associations among variables. Finally, our aggregated measures of childcare support are relatively blunt, lacking information on the specific natures of support (i.e. what kind of contribution was actually made by whom and to whom?). Factors, such as the quality of childcare support, or conflict among potential caregivers (e.g. [73]) were not carefully considered within this study. While our study was designed to test broad associations between childcare support and both children's outcomes and parental characteristics, future studies should examine both cooperation and conflict within the family, different alloparental effects by maternal/paternal grandmothers/grandfathers (e.g. [74]), and different impacts of practical/emotional/financial support on children and parents (e.g. [23,26]).

Despite these limitations, our study has provided insight into the complex mechanisms behind a partner and alloparental childcare support and child social development in contemporary Japan, a non-Western industrialized context. The results indicate that alloparenting remains important but can be mediated by its effects on parental psychological condition and parenting style. We hope that our findings will enhance cross-cultural understanding of alloparenting and social support.

Acknowledgements

We would like to thank Mr Yudai Tokumasu and other members of the Evolutionary Anthropology Laboratory at the University of Tokyo for their insightful comments. Two anonymous reviewers and Drs Emily H. Emmott, Abigail E. Page and Sarah Myers gave us valuable and constructive suggestions to improve our paper. Dr Emily H. Emmott also provided us with immense help in editing our manuscript.

Contributor Information

Masahito Morita, Email: mmorita.human@gmail.com.

Atsuko Saito, Email: atsaito@sophia.ac.jp.

Ethics

Our questionnaire survey was approved by the ethical committees at the University of Tokyo (no. 306) and Musashino University (no. 29001).

Data accessibility

The dataset that we analysed in this paper and the questionnaire (in Japanese) are available on the journal's website. The English translation of our original questions about childcare support and some R codes for statistical analyses are also provided in the electronic supplementary material.

Authors' contributions

M.M., A.S., M.N. and Y.I. developed the theoretical framework. A.S. and M.N. designed the original questionnaire survey and collected the data. M.M., A.S. and M.N. prepared the dataset for statistical analyses. M.M. and M.N. carried out the statistical analyses with crucial advice from Y.I. M.M., A.S., N.M. and Y.I. contributed to the interpretation of the results. M.M. drafted the manuscript. M.M., A.S., M.N. and Y.I. revised the manuscript. All authors gave final approval for publication and agree to be held accountable for the work performed therein.

Competing interests

We declare that we have no competing interests.

Funding

This study was financially supported by MEXT/JSPS KAKENHI (Grant-in-Aid for Scientific Research) grant nos. JP25118003 and JP18K02461 (to A.S.), Incentive Allowance for Dissemination of Individual Research Results of Sophia University (to A.S.) and MEXT KAKENHI (Grant-in-Aid for Scientific Research) grant no. JP17H06381 (to Y.I.).

References

  • 1.Bentley G, Mace R. 2009. Substitute parents: biological and social perspectives on alloparenting in human societies. New York, NY: Berghahn Books. [Google Scholar]
  • 2.Emmott EH, Page AE. 2019. Alloparenting. In Encyclopedia of evolutionary psychological science (eds Shackelford TK, Weekes-Shackelford VA). Cham, Switzerland: Springer. [Google Scholar]
  • 3.Kramer KL. 2010. Cooperative breeding and its significance to the demographic success of humans. Annu. Rev. Anthropol. 39, 417-436. ( 10.1146/annurev.anthro.012809.105054) [DOI] [Google Scholar]
  • 4.Bogin B, Bragg J, Kuzawa C. 2014. Humans are not cooperative breeders but practice biocultural reproduction. Ann. Hum. Biol. 41, 368-380. ( 10.3109/03014460.2014.923938) [DOI] [PubMed] [Google Scholar]
  • 5.Sear R, Mace R. 2008. Who keeps children alive? A review of the effects of kin on child survival. Evol. Hum. Behav. 29, 1-18. ( 10.1016/j.evolhumbehav.2007.10.001) [DOI] [Google Scholar]
  • 6.Sear R, Coall D. 2011. How much does family matter? Cooperative breeding and the demographic transition. Popul. Dev. Rev. 37, 81-112. ( 10.1111/j.1728-4457.2011.00379.x) [DOI] [PubMed] [Google Scholar]
  • 7.Anderson LR, Sheppard P, Monden CWS. 2018. Grandparent effects on educational outcomes: a systematic review. Sociol. Sci. 5, 114-142. ( 10.15195/v5.a6) [DOI] [Google Scholar]
  • 8.Del Boca D, Piazzalunga D, Pronzato C. 2017. The role of grandparenting in early childcare and child outcomes. Rev. Econ. Househ. 16, 477-512. ( 10.1007/s11150-017-9379-8) [DOI] [Google Scholar]
  • 9.Hansen K, Hawkes D. 2009. Early childcare and child development. J. Soc. Pol. 38, 211-239. ( 10.1017/S004727940800281X) [DOI] [Google Scholar]
  • 10.Lefkovics E, Rigó J Jr, Kovács I, Talabér J, Szita B, Kecskeméti A, Szabó L, Somogyvári Z, Baji I. 2018. Effect of maternal depression and anxiety on mother's perception of child and the protective role of social support. J. Reprod. Infant Psychol. 36, 434-448. ( 10.1080/02646838.2018.1475726) [DOI] [PubMed] [Google Scholar]
  • 11.Matsuda S. 2008. What supports child-rearing? The strength of moderate networks [Nani ga ikuji o sasaerunoka: chuyo na nettowaku no tsuyosa]. Tokyo, Japan: Keiso Shobo; (in Japanese). [Google Scholar]
  • 12.Sadruddin AFA, Ponguta LA, Zonderman AL, Wiley KS, Grimshaw A, Panter-Brick C. 2019. How do grandparents influence child health and development? a systematic review. Soc. Sci. Med. 239, 112476. ( 10.1016/j.socscimed.2019.112476) [DOI] [PubMed] [Google Scholar]
  • 13.Yamaguchi S, Asai Y, Kambayashi R. 2018. How does early childcare enrollment affect children, parents, and their interactions? Labour Econ. 55, 56-71. ( 10.1016/j.labeco.2018.08.006) [DOI] [Google Scholar]
  • 14.Goodman A, Koupil I, Lawson DW. 2012. Low fertility increases descendant socioeconomic position but reduces long-term fitness in a modern post-industrial society. Proc. R. Soc. B 279, 4342-4351. ( 10.1098/rspb.2012.1415) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Kaplan HS, Lancaster JB. 2003. An evolutionary and ecological analysis of human fertility, mating patterns, and parental investment. In Offspring: human fertility behavior in biodemographic perspective (eds Wachter KW, Bulatao RA), pp. 170-223. Washington, DC: The National Academies Press. [PubMed] [Google Scholar]
  • 16.Lawson DW, Borgerhoff Mulder M. 2016. The offspring quantity–quality trade-off and human fertility variation. Phil. Trans. R. Soc. B 371, 20150145. ( 10.1098/rstb.2015.0145) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Feinstein L, Bynner J. 2004. The importance of cognitive development in middle childhood for adulthood socioeconomic status, mental health, and problem behavior. Child Dev. 75, 1329-1339. ( 10.1111/j.1467-8624.2004.00743.x) [DOI] [PubMed] [Google Scholar]
  • 18.Jones DE, Greenberg M, Crowley M. 2015. Early social-emotional functioning and public health: the relationship between kindergarten social competence and future wellness. Am. J. Public Health 105, 2283-2290. ( 10.2105/AJPH.2015.302630) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Moffitt TE, et al. 2010. A gradient of childhood self-control predicts health, wealth, and public safety. Proc. Natl Acad. Sci. USA 108, 2693-2698. ( 10.1073/pnas.1010076108) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Page AE, et al. 2019. Testing adaptive hypotheses of alloparenting in Agta foragers. Nat. Hum. Behav. 3, 1154-1163. ( 10.1038/s41562-019-0679-2) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Page AE, Emmott EH, Dyble M, Smith D, Chaudhary N, Viguier S, Migliano AB. 2021. Children are important too: juvenile playgroups and maternal childcare in a foraging population, the Agta. Phil. Trans. R. Soc. B 376, 20200026. ( 10.1098/rstb.2020.0026) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Belsky J. 1984. The determinants of parenting: a process model. Child Dev. 55, 83-96. ( 10.2307/1129836) [DOI] [PubMed] [Google Scholar]
  • 23.Emmott EH, Mace R. 2015. Practical support from fathers and grandmothers is associated with lower levels of breastfeeding in the UK Millennium Cohort Study. PLoS ONE 10, e0133547. ( 10.1371/journal.pone.0133547) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Emmott EH, Page AE, Myers S. 2020. Typologies of postnatal support and breastfeeding at two months in the UK. Soc. Sci. Med. 246, 112791. ( 10.1016/j.socscimed.2020.112791) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chang Y-S, Li KMC, Li KYC, Beake S, Lok KYW, Bick D. 2021. Relatively speaking? Partners' and family members' views and experiences of supporting breastfeeding: a systematic review of qualitative evidence. Phil. Trans. R. Soc. B 376, 20200033. ( 10.1098/rstb.2020.0033) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Myers S, Page AE, Emmott EH. 2021. The differential role of practical and emotional support in infant feeding experience in the UK. Phil. Trans. R. Soc. B 376, 20200034. ( 10.1098/rstb.2020.0034) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Deater-Deckard K. 1998. Parenting stress and child adjustment: some old hypotheses and new questions. Clin. Psychol. 5, 314-332. ( 10.1111/j.1468-2850.1998.tb00152.x) [DOI] [Google Scholar]
  • 28.Kawai E, et al. 2017. Maternal postpartum depressive symptoms predict delay in non-verbal communication in 14-month-old infants. Infant Behav. Dev. 46, 33-45. ( 10.1016/j.infbeh.2016.11.006) [DOI] [PubMed] [Google Scholar]
  • 29.Oyetunji A, Ghandra P. 2020. Postpartum stress and infant outcome: a review of current literature. Psychiatry Res. 284, 112769. ( 10.1016/j.psychres.2020.112769) [DOI] [PubMed] [Google Scholar]
  • 30.Garon N, Bryson SE, Smith IM. 2008. Executive function in preschoolers: a review using an integrative framework. Psychol. Bull. 134, 31-60. ( 10.1037/0033-2909.134.1.31) [DOI] [PubMed] [Google Scholar]
  • 31.Wellman HM, Cross D, Watson J. 2001. Meta-analysis of theory-of-mind development: the truth about false belief. Child Dev. 72, 655-684. ( 10.1111/1467-8624.00304) [DOI] [PubMed] [Google Scholar]
  • 32.Eisenberg N, Mussen PH. 1989. The roots of prosocial behavior in children. Cambridge, UK: Cambridge University Press. [Google Scholar]
  • 33.United Nations, Department of Economic and Social Affairs, Population Division. 2019. Database on household size and composition. See https://population.un.org/Household/#/countries/840 (last accessed on 24th November 2020).
  • 34.Ministry of Health, Labour and Welfare. 2017. Summary report of comprehensive survey of living conditions 2016. See https://www.mhlw.go.jp/english/database/db-hss/dl/report_gaikyo_2016.pdf (last accessed on 8th December 2020).
  • 35.Morrone MH, Matsuyama Y. 2010. Japan's parental leave policy: has it affected gender ideology and child care norms in Japan? Childhood Edu. 86, 371-375. ( 10.1080/00094056.2010.10523172) [DOI] [Google Scholar]
  • 36.Taga F. 2016. EU countries’ implications for promoting fathers' participation in parenting in Japan. Jpn. J. Fam. Sociol. 28, 207-213. ( 10.4234/jjoffamilysociology.28.207) [DOI] [Google Scholar]
  • 37.Cabinet Office, Government of Japan. 2019. Public opinion survey on a gender-equal society [Danjo kyodo sankaku shakai ni kansuru yoron chosa]. See https://survey.gov-online.go.jp/r01/r01-danjo/2-2.html (last accessed on 24th November 2020) (in Japanese).
  • 38.Fujita Y. 2017. One-operation child-rearing: I want you to understand that I cannot rest every day [One-ope ikuji wakatte hoshii yasumenai nichijo]. Tokyo, Japan: Mainichi Shinbun Publishing Inc. (In Japanese) [Google Scholar]
  • 39.Hoshi A. 2012. Effects of support networks on the well-being of child-rearing women: analysis of NERJ08. Quart. Soc. Secur. Res. 48, 279-289. [Google Scholar]
  • 40.Ishii-Kuntz M. 2013. Work environment and Japanese fathers' involvement in child care. J. Fam. Issues 34, 252-271. ( 10.1177/0192513X12462363) [DOI] [Google Scholar]
  • 41.Maeda N. 2004. The structural effects of child-rearing women's personal networks on their social support and psychological strain. Jpn. J. Fam. Sociol. 16, 21-31. ( 10.4234/jjoffamilysociology.16.21) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 42.Matsuda S. 2001. Childcare networks and the well-being of mothers. Jpn. Sociol. Rev. 52, 33-49. ( 10.4057/jsr.52.33) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 43.Ochiai E. 1989. The modern family and feminism [Kindai kazoku to feminizumu]. Tokyo, Japan: Keiso Shobo; (in Japanese). [Google Scholar]
  • 44.Shinada T. 2010. Living environment and parent-child life [Kyoju kankyo to oyako seikatsu]. In Shifting bases for child-rearing: current situations and difficulties in a low-birthrate society [Yuragu kosodate kiban: shoshika shakai no genjo to konnan] (eds Matsuda S, Shiomi K, Shinada T, Suemori K), pp. 61-87. Tokyo, Japan: Keiso Shobo. (In Japanese) [Google Scholar]
  • 45.Tsuya NO, Bumpass LL. 2004. Marriage, work, and family life in comparative perspective: Japan, South Korea, and the United States. Honolulu, Hawaii: University of Hawai'i Press. [Google Scholar]
  • 46.Takeuchi M. 2003. Danish social policy supports child and family. St. Andrew‘s Univ. Sociol. Rev. 37, 35-84. (In Japanese with an English abstract) [Google Scholar]
  • 47.Saito A, Nozaki M. 2017. Social support in childrearing for mothers of 3-5-year-old children. Bull. Musashino Univ. Fac. Educ. 1, 11-19. (In Japanese) [Google Scholar]
  • 48.Saito A, Nozaki M. 2018. Childrearing and social support in fathers of 3-5 years old children. Bull. Musashino Univ. Fac. Educ. 4, 9-17. (In Japanese) [Google Scholar]
  • 49.Benesse Institute for the Development of Next Generation. 2007. The first basic survey of pregnancy, childbirth, and childcare [Dai ikkai ninshin shussan kosodate kihon chosa (odan chosa)]. See https://berd.benesse.jp/jisedai/research/detail1.php?id=3319 (last accessed on 12th November 2020) (in Japanese).
  • 50.Cabinet Office, Government of Japan. 2002. White paper on the national lifestyle, fiscal year 2001 [Heisei 13 nendo kokumin seikatsu hakusho: kazoku no kurashi to kozo kaikaku]. See https://warp.da.ndl.go.jp/info:ndljp/pid/9990748/www5.cao.go.jp/seikatsu/whitepaper/wp-pl/wp-pl01/index.html (last accessed on 12th November 2020) (in Japanese).
  • 51.Kashiwagi K, Wakamatsu M. 1994. ‘Becoming a parent’ and personality development: a lifespan developmental view. Jpn. J. Dev. Psychol. 5, 72-83. ( 10.11201/jjdp.5.72) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 52.Motoyasu K, Yaegashi M. 2003. A study on the relation between mother's anxiety about bringing up children and the participation of the father in household matters and child rearing. Kawasaki Med. Welf. J. 13, 1-13. (In Japanese with an English abstract) [Google Scholar]
  • 53.Ogata K. 1995. The fathers' care of pre-school children and their social adaptability: a comparison of single income families and double income families. Jap. J. Educ. Psychol. 43, 335-342. ( 10.5926/jjep1953.43.3_335) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 54.Goodman R, Scott S. 1997. The Strength and Difficulties Questionnaire: a research note. J. Child Psychol. Psychiatry, 38, 581-586. ( 10.1111/j.1469-7610.1997.tb01545.x) [DOI] [PubMed] [Google Scholar]
  • 55.Matsuishi T, et al. 2008. Scale properties of the Japanese version of the Strengths and Difficulties Questionnaire (SDQ): a study of infant and school children in community samples. Brain Dev-Jpn. 30, 410-415. ( 10.1016/j.braindev.2007.12.003) [DOI] [PubMed] [Google Scholar]
  • 56.Abidin RR. 1995. Parenting Stress Index professional manual, 3rd edn. Odessa, FL: Psychological Assessment Resources. [Google Scholar]
  • 57.Kanematsu Y, Araki A, Narama M, Shirahata N, Maru M, Arayashiki R. 2006. PSI: Parenting Stress Index manual (second edition) [PSI tebiki (nitei ban)]. Tokyo, Japan: Koyo Mondai Kenkyukai. (In Japanese) [Google Scholar]
  • 58.Beck AT, Steer RA, Brown GK. 1996. Manual for the Beck Depression Inventory-II. San Antonio, Hawaii: Psychological Corporation. [Google Scholar]
  • 59.Kojima M, Furukawa T. 2003. Manual for the Beck Depression Inventory [BDI-II tebiki], 2nd edn. Tokyo, Japan: Nihon Bunka Kagakusha. (In Japanese) [Google Scholar]
  • 60.Kojima M, Furukawa TA, Takahashi H, Kawai M, Nagaya T, Tokudome S. 2002. Cross-cultural validation of the Beck Depression Inventory-II in Japan. Psychiatry Res. 110, 291-299. ( 10.1016/S0165-1781(02)00106-3) [DOI] [PubMed] [Google Scholar]
  • 61.Parker G. 1979. Parental characteristics in relation to depressive disorders. Br. J. Psychiatry 134, 138-147. ( 10.1192/bjp.134.2.138) [DOI] [PubMed] [Google Scholar]
  • 62.Sugawara M, Yagishita A, Takuma N, Koizumi T, Sechiyama H, Sugawara K, Kitamura T. 2002. Marital relations and depression in school-age children: links with family functioning and parental attitudes toward child rearing. Jpn. J. Educ. Psychol. 50, 129-140. ( 10.5926/jjep1953.50.2_129) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 63.Hanada H, Konishi M. 2003. Reliability and validity of a screening questionnaire for child abuse potential risk by mother's parenting. J. Health Sci. Hiroshima Univ. 3, 55-62. ( 10.15027/16428) (In Japanese with an English abstract) [DOI] [Google Scholar]
  • 64.R Core Team. 2018. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. See https://www.R-project.org/. [Google Scholar]
  • 65.Blair C, et al. 2014. Two approaches to estimating the effect of parenting on the development of executive function in early childhood. Dev. Psychol. 50, 554-565. ( 10.1037/a0033647) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Pearl AM, French BF, Dumas JE, Moreland AD, Prinz R. 2014. Bidirectional effects of parenting quality and child externalizing behavior in predominantly single parent, under-resourced African American families. J. Child Fam. Stud. 23, 177-188. ( 10.1007/s10826-012-9692-z) [DOI] [Google Scholar]
  • 67.Ministry of Education, Culture, Sports, Science and Technology. 2006. Survey on the educational power of communities [Chiiki no kyoiku-ryoku ni kansuru jittai chosa]. See https://www.mext.go.jp/b_menu/shingi/chukyo/chukyo2/003/siryou/__icsFiles/afieldfile/2014/06/20/1265399_001.pdf (last accessed on 24th November 2020) (in Japanese).
  • 68.Ministry of Land, Infrastructure, Transport and Tourism. 2012. Survey on the actual conditions on the use of multifunctional toilets for families with children [Kodomo-dure ni taisuru takino toire nadono riyo jittai anketo chosa]. See https://www.mlit.go.jp/common/000209819.pdf (last accessed on 24th November 2020) (in Japanese).
  • 69.Spry EA, et al. 2021. Preventing postnatal depression: a causal mediation analysis of a 20-year preconception cohort. Phil. Trans. R. Soc. B 376, 20200028. ( 10.1098/rstb.2020.0028) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Tanskanen AO, Danielsbacka M. 2018. Multigenerational effects on children's cognitive and socioemotional outcomes: a within-child investigation. Child Dev. 89, 1856-1870. ( 10.1111/cdev.12968) [DOI] [PubMed] [Google Scholar]
  • 71.Huang C-Y, Yen H-C, Tseng M-H, Tung L-C, Chen Y-D, Chen K-L. 2014. Impacts of autistic behaviors, emotional and behavioral problems on parenting stress in caregivers of children with autism. J. Autism Dev. Disord. 44, 1383-1390. ( 10.1007/s10803-013-2000-y) [DOI] [PubMed] [Google Scholar]
  • 72.Benesse Institute for the Development of Next Generation. 2009. The third basic survey of childcare life [Dai sankai kosodate seikatsu kihon chosa (yoji ban)]. See https://berd.benesse.jp/shotouchutou/research/detail1.php?id=3284 (last accessed on 24th November 2020) (in Japanese).
  • 73.Shenk MK, et al. 2021. Social support, nutrition and health among women in rural Bangladesh: complex tradeoffs in allocare, kin proximity and support network size. Phil. Trans. R. Soc. B 376, 20200027. ( 10.1098/rstb.2020.0027) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Tanskanen AO, Jokela M, Danielsbacka M, Rotkirch A. 2014. Grandparental effects on fertility vary by lineage in the United Kingdom. Hum. Nat. 25, 269-284. ( 10.1007/s12110-014-9200-9) [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The dataset that we analysed in this paper and the questionnaire (in Japanese) are available on the journal's website. The English translation of our original questions about childcare support and some R codes for statistical analyses are also provided in the electronic supplementary material.

Articles from Philosophical Transactions of the Royal Society B: Biological Sciences are provided here courtesy of The Royal Society

RESOURCES