Kinship care linked to better early childhood development in Thailand

Grandparents, aunts, and uncles who read, play, and explore with young children may be among the most underappreciated forces in early childhood development, according to a new study from Thailand that finds kin involvement significantly boosts children’s physical, cognitive, and social-emotional outcomes—especially in poorer families. Drawing on data from more than 10,000 Thai children under the age of five, researchers at the National Institute of Development Administration have produced some of the strongest empirical evidence yet that extended family members do far more than fill childcare gaps: they measurably shape how young children grow.

The research, published in the International Journal of Child Care and Education Policy by Natthakorn Naknong and Piriya Pholphirul, analyzed data from the 2022 Multiple Indicator Cluster Survey (MICS-Thailand), a nationally representative survey conducted by Thailand’s National Statistical Office with support from UNICEF. The sample comprised 10,638 children aged zero to four years—5,508 boys and 5,130 girls—whose mothers or primary caregivers were interviewed face-to-face using computer-assisted personal interviewing methods. The dataset offered an unusually rich picture of household composition and daily caregiving practices in a country where extended families are deeply woven into the social fabric.

Thailand provided a compelling setting for the investigation. Roughly one-third of Thai households—33.6 percent in 2013, according to UNFPA Thailand—are extended families, a higher proportion than nuclear families in the country. This stands in sharp contrast to Western nations, where nuclear arrangements dominate. Globally, extended family households account for 38 percent of the world’s population, with prevalence reaching 45 percent in the Asia-Pacific region, 35 percent in Africa, 32 percent in Latin America, and 27 percent in the Middle East and North Africa—compared with just 26 percent in Europe and 11 percent in North America. Despite this demographic reality, the bulk of early childhood development research has been conducted in developed countries and centered on parental involvement, particularly the well-documented “motherhood effect,” leaving a critical gap in understanding how non-parental kin contribute to child development in the majority world.

The time pressures facing Thai parents made the question especially urgent. Thai workers average 2,185 hours of work annually, well above the OECD average of 1,742 hours and trailing only some of Southeast Asia’s most work-intensive economies, such as Cambodia at 2,456 hours and Myanmar at 2,438. Prior research by the same group has shown that married Thai women tend to participate more in the labor market and work longer hours than unmarried women, particularly if they are young, less educated, or have fewer dependents requiring care. In households where both parents work long hours, the active involvement of co-residing relatives becomes a critical supplement—or sometimes substitute—for direct parental engagement.

To quantify the developmental stakes, the researchers assessed children across three domains using caregiver-reported milestones. Physical development was measured by a child’s ability to walk on uneven or inclined surfaces without losing balance, jump with both feet leaving the ground simultaneously, and dress independently. Cognitive development required recognizing at least five letters in Thai or Roman script, counting ten objects accurately, and producing a vocabulary of at least ten words. Emotional and social development was judged by the absence of persistent emotional difficulties such as sadness or withdrawal, proficiency in socializing with peers, and the absence of overt physical aggression. A child was classified as typically developing in a domain only if all three criteria were met, and outcomes were coded dichotomously as typical or delayed.

The activity data came from questions asking whether the child had engaged in six specific activities with their mother, father, or co-residing relatives within the previous three days: reading books or looking at picture books, telling stories, singing songs, going outside the home, playing, and teaching the child to name animals or objects, count, or draw. Descriptive analysis showed that mothers were the most active participants across all activities, followed by fathers, with relatives showing participation rates comparable to fathers. The most common activities were taking children outside and playing, each involving roughly 48 to 49 percent of caregivers, followed by teaching basic skills at 46 to 48 percent. Notably, girls and boys experienced similar levels of engagement, but girls consistently outperformed boys in development: 35.19 percent versus 31.83 percent for physical development, 23.43 percent versus 20.95 percent for cognitive development, and 36 percent versus 31.81 percent for emotional and social development.

The core of the study’s analytical power lay in its econometric modeling. The researchers employed a binary probit regression, estimating the probability that a child exhibited typical development in each domain as a function of the cumulative distribution function of the standard normal distribution applied to a linear combination of predictors. The model included both family activity variables and a comprehensive set of controls: child gender and age, parental education, residential location, region, household size, the number of young children and school-age children in the home, and household economic status. Economic status was captured through a wealth index constructed via principal component analysis using household assets, dwelling characteristics, and access to basic services, then divided into five quintiles.

The marginal effects from the probit models told a striking story. When parental activity variables were included in the model, kin involvement remained statistically significant and even strengthened. Children who read or looked at picture books with relatives showed a 2.94 percent higher probability of typical physical development, 3.91 percent higher for cognitive development, and 5.96 percent higher for emotional and social development compared with children without such involvement. Outdoor activities with kin raised the probability of typical physical and emotional-social development by 3.01 and 3.02 percent, respectively, while playing with kin increased the probability of typical emotional and social development by 4.2 percent. Crucially, the researchers argue that endogeneity concerns are mitigated by the fact that kin involvement in Asian extended families is typically established early in a child’s life as part of an entrenched caregiving pattern.

The parental findings provided a useful contrast. Maternal activities chiefly influenced cognitive and emotional-social development—maternal reading raised the probability of typical cognitive development by 3.81 percent and emotional-social development by 13.20 percent, while maternal play raised the latter by 9.54 percent. Paternal activities, by contrast, were most strongly tied to physical development: fathers taking children outdoors increased the probability of typical physical development by 9.88 percent, and paternal singing by 4.67 percent. Kin involvement stood apart as the only category of adult engagement associated with significant benefits across all three developmental domains simultaneously.

Perhaps the most policy-relevant finding emerged when the researchers stratified results by economic status. Kin involvement was concentrated in poorer households: relatives in the poorest two quintiles showed engagement rates in child-centered activities reaching 32.23 percent, compared with markedly lower rates in the wealthiest households, where parents can afford paid domestic help or formal childcare. In the stratified regressions, kin engagement in reading and outdoor activities significantly increased the probability of typical physical and cognitive development among children from impoverished families. In other words, kinship networks appear to function as an informal developmental safety net precisely where formal services and parental time are scarcest.

The results align closely with Bronfenbrenner’s ecological systems theory, which frames child development as the product of interacting environmental layers, from the immediate microsystem of family relationships outward to broader cultural and economic macrosystems. In this framework, co-residing kin activate and enrich the microsystem, expanding what the authors call the “developmentally instigative” capacity of the household—ensuring consistent, stimulating engagement even when parental availability is constrained. The stronger effects observed in low-to-middle-income families suggest kinship networks also serve as a vital mesosystem link, channeling extended family resources toward the child’s immediate needs and fostering resilience.

The study is not without limitations. The developmental assessments—counting, letter recognition, basic motor tasks—may not fully capture cognitive skills relevant to contemporary childhood, and the activity measures recorded only whether an interaction occurred in the previous three days, not its frequency or quality. The cross-sectional design precludes causal inference over time, and the absence of data on kin occupations, kin gender, and detailed family structure raises the possibility of omitted variable bias.

Even so, the findings carry clear policy implications for Thailand and other developing countries. The authors recommend that the Ministry of Social Development and Human Security, the Ministry of Public Health, and the Ministry of Education launch public awareness campaigns—leveraging community health volunteers and culturally tailored media—to elevate the perceived value of kin engagement in early childhood. They further propose distributing Thai-language picture books, educational play kits, and illustrated caregiver guides through community libraries and health centers, potentially in partnership with NGOs and private publishers. Finally, they advocate designing accessible, intergenerational play spaces using low-cost local materials, citing models such as the “Play for All” initiative led by UNICEF Thailand and the MSDHS, which transformed underutilized community areas into inclusive playgrounds through participatory design. In a world where nearly four in ten people live in extended family households, the message is simple: grandparents and other relatives are not merely childcare backups—they are active architects of how the next generation learns to walk, talk, and thrive.