Study setting and data

We conducted an analysis of secondary data derived from the 2019 Multiple Indicator Cluster Survey (MICS) in Bangladesh [15]. This comprehensive survey took place from January to June 2019. It is worth noting that the MICS is a household survey developed by UNICEF, conducted in various countries, with the aim of providing statistically rigorous data on a wide range of indicators concerning the situation of children and women. In Bangladesh, 2019 MICS survey was carried out by the Bangladesh Bureau of Statistics in collaboration with the UNICEF and other partners. Covering all 64 districts of Bangladesh, the survey utilized a two-stage stratified cluster sampling design. In the first stage, 1,300 primary sampling units (PSUs) were selected. These PSUs were defined as either rural or urban areas with an approximate population of 200 to 300 households. The second stage involved the systematic random sampling of 28 households from each PSU, resulting in a total sample size of 36,400 households. Within these households, 24,686 eligible children were identified, and data were collected from 23,099 of these children with a response rate of 93.60%.

Analytical sample

We have included children aged 3–4 years old from whom the ECDI-related questions were asked, following the guidelines provided by UNICEF [15, 16]. Out of the total 23,099 children included in the main survey, there were 8,833 children aged 3–4 years old, with 4,513 children aged 3 years and 4,320 children aged 4 years. Therefore, they were included in the analysis.

Outcome variable

The study focused on the Early Childhood Development Index (ECDI), a composite measure encompassing a ten-item module across four domains: literacy-numeracy (3 items), physical (2 items), social-emotional (3 items), and learning (2 items) [16, 17]. Children were deemed on track in the literacy-numeracy domain if they achieved milestones like letter identification, reading simple words, and number recognition. For the physical domain, on-track status depended on fine motor skills and limited illness hindrance. Positive interaction with peers and non-aggressive behavior were indicators in the social-emotional domain. In the learning domain, following directions and independent task completion were considered on track. Domain indexes were calculated based on specific criteria recommended by UNICEF, with at least two positive responses needed in literacy-numeracy and social-emotional domains, two positive replies in the physical domain, and at least one positive reply in the learning domain for a positive classification [16, 17]. The ECDI was then derived as the percentage of children meeting developmental milestones in at least three of these domains, distinguishing between those who were developmentally on track and those who were not. This comprehensive index is widely utilized by organizations, including UNICEF, to monitor progress towards early childhood development goals [16, 17].

Explanatory variable

The primary explanatory variable examined was the parental migration status. The relevant data pertaining to parental migration status were collected during the survey by posing specific questions to eligible respondents, typically the caretakers of the children. The questions included inquiries such as "Where does the natural father of (name of the selected children) reside?" and "Where does the natural mother of (name of the selected children) reside?" If an affirmative response was recorded, additional follow-up questions were asked to determine the duration (in months) that the natural father or mother had lived away from home. Based on these responses, variables were generated to indicate father migration and mother migration, with a positive response indicating the absence of the respective parent for at least 6 months. Additionally, based on these two other migration variables were created: one representing any type of parental migration (either father or mother migrated, versus none), and the other representing both parents’ migration (both father and mother migrated, versus none). These classifications were based on previous studies conducted in Bangladesh [4].

Covariates

Numerous covariates were carefully selected to adjust potential confounding effects on the association between parental migration and the early childhood development of left-behind children in Bangladesh. The selection of these covariates was based on prior research findings [4, 6, 8–10, 14]. The covariates taken into account consist of the child’s age, sex, birth order, and health status; household socioeconomic status (SES); maternal education and employment status; as well as the presence of other adult caregivers in the household.

Statistical analysis

Descriptive statistics were utilized to provide a summary of the study population’s characteristics, including age, sex, socioeconomic status, and other pertinent variables. To examine the relationship between parental migration status (father migration, mother migration, both parents’ migration) and ECDI of left-behind children, multilevel logistic regression analysis was conducted, controlling for covariates. The decision to employ a multilevel logistic regression model was based on the hierarchical structure of the MICS data, where individuals were nested within households, and households were nested within clusters. This nesting structure resulted in a linkage between individual responses, rendering simple logistic regression inadequate for producing accurate results. In contrast, the multilevel logistic regression model was deemed the recommended approach for generating more robust outcomes [18]. The findings were reported as odds ratios (OR) with 95% confidence intervals (95% CI). A p-value of less than 0.05 was considered statistically significant. All analyses were performed using the statistical package STATA version 15.1.

Demographic characteristics of the respondents

The demographic characteristics of the respondents are presented in Table 1. The study included an almost equal number of children aged 3 year and 4 year, as well as an equal distribution of males and females. Approximately 18% of the total children participated in early childhood education. It was found that around 45% of the children’s mothers had attained a secondary level of education, while 4% of the children’s mothers reported functional difficulties. Roughly one-fourth of the children were from households classified as the poorest. Moreover, approximately four-fifths of the children lived in rural areas, and Chattogram division was reported as the region of residence by a similar proportion of the children.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of the respondents, N = 8,833.

https://doi.org/10.1371/journal.pone.0287828.t001

Distribution of early childhood development and parental migration situation in Bangladesh

Table 2 presents the distribution of domains of ECDI and parental migration. The findings revealed that 71% of the total children were found to be developmentally on track in terms of ECDI. In the independent domains of ECDI, approximately 73% and 78% of the total children demonstrated developmental progress in the literacy-numeracy and physical domains, respectively. Around 34% of the total children exhibited developmental progress in the socio-emotional domain, while only 9% were observed to be on track in the learning domain of ECDI. In terms of parental migration, it was found that approximately 9% of the total children had fathers who migrated abroad, while 3.3% had fathers who migrated within the country. The migration of mothers alone accounted for a mere 0.3% of the total children. Additionally, the migration of both parents was reported for only 2% of the total children.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Descriptive statistics exposure and outcome variables.

https://doi.org/10.1371/journal.pone.0287828.t002

Association between parental migration and early childhood development and its independent domains

The associations between parental migration and early childhood development and its independent domains are presented in Table 3. Children with fathers who migrated abroad had significantly decreased odds of being developmentally on track in the literacy-numeracy domain (OR: 0.84, 95% CI: 0.73–0.96), physical domain (OR: 0.88, 95% CI: 0.76–0.99), learning domain (OR: 0.80, 95% CI: 0.70–0.91), and overall ECDI (OR: 0.74, 95% CI: 0.54–0.93). Furthermore, when both parents migrated, a significant association was found with decreased odds of the learning domain (OR: 0.86, 95% CI: 0.55–1.35) and overall ECDI (OR: 0.63, 95% CI: 0.48–0.97). Notably, any type of parental migration exhibited significant associations with decreased odds of being developmentally on track in the literacy-numeracy domain (OR: 0.90, 95% CI: 0.80–0.99), learning domain (OR: 0.83, 95% CI: 0.71–0.98), and overall ECDI (OR: 0.85, 95% CI: 0.77–0.95).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Effects of parental migration on early childhood development in Bangladesh adjusting for covariates.

https://doi.org/10.1371/journal.pone.0287828.t003

Among the adjusted covariates, it was found that poor early ECDI outcomes were associated with several factors (S1–S4 Tables). Specifically, children in Barishal division, those who did not attend early childhood education, and children whose mothers had functional difficulties or belonged to the poorest wealth quintile were found to have lower likelihoods of positive early childhood development. These factors were identified as significant contributors to the observed disparities in ECDI outcomes.

Discussion

The primary objectives of this study were to examine the impact of parental migration on early childhood development in Bangladesh, specifically focusing on the ECDI and its individual domains. The findings of this study provide compelling evidence of the significant negative effects of parental migration on the developmental outcomes of children in the country. The results indicate that children with fathers who migrated abroad faced notable disadvantages, experiencing a 26% decrease in the likelihood of being developmentally on track overall. This reduction becomes even more pronounced, reaching 37%, when both parents migrate. Furthermore, children whose parents had migrated, irrespective of the type of migration, exhibited a significant 15% decrease in their ECDI scores. These findings underscore the wide-ranging consequences of parental absence due to migration on a child’s overall development, emphasizing the critical importance of the early years as these effects can have lifelong implications.

The observed negative correlation between parental migration and poor ECDI scores are comparable to the other study findings in China and other countries [10, 12, 14, 19–21]. This relationship can be attributed to several contributing factors. Firstly, children of migrant parents are at an increased risk of facing challenges related to malnutrition, compromised physical health, and unintentional injuries compared to children whose parents do not migrate [21]. This disparity arises from the lack of parental involvement and support in fulfilling childcare responsibilities, resulting in inadequate supervision and care for the child’s well-being [1]. Moreover, the absence of parents due to migration disrupts the crucial parent-child relationship during the formative years, which plays a fundamental role in fostering healthy child development [21]. Parental involvement, support, and nurturing are vital for creating a nurturing environment that promotes cognitive, social, and emotional growth [1, 4, 22]. When parents are absent for extended periods, children may experience a deficiency in emotional bonding and have limited opportunities for interactive learning experiences, thus hindering their overall development [1, 3, 4].

Furthermore, the emotional strain and stress associated with parental migration can have profound detrimental effects on children’s overall well-being [3]. The separation from a parent can trigger feelings of anxiety, sadness, and insecurity, significantly impacting their emotional and cognitive functioning [12, 21]. These emotional disruptions may manifest as challenges in regulating emotions, forming relationships, and adapting to new environments, ultimately compromising socio-emotional development [21]. Another influential factor contributing to the negative association between parental migration and ECDI scores is the economic strain faced by families with migrating parents [9]. The financial pressures and the need for remittances to support the family back home can divert attention and resources away from investing in the child’s early development [14]. Limited financial means may result in reduced access to quality early childhood education, healthcare services, and enriching learning materials, all of which are essential for promoting optimal cognitive and physical development [23, 24]. Furthermore, the absence of parents can disrupt the stability of the home environment, leading to alterations in daily routines, changes in caregiver arrangements, and shifts in family dynamics [25, 26]. These disruptions introduce stressors and inconsistencies that negatively impact a child’s sense of security and stability, potentially hindering their learning progress and overall development [1, 9].

These findings depict a deeply concerning reality in which parents, driven by unwavering hope to secure a brighter future for their children, unknowingly subject them to adverse consequences. In Bangladesh, there is an increasing trend of parents opting for migration, underscoring the urgent need for comprehensive and targeted interventions. It is crucial for policymakers, educators, and practitioners to prioritize the development and implementation of support systems specifically tailored to address the unique needs of children with migrating parents. Effective initiatives should include the establishment of community-based programs that serve as a lifeline, providing emotional support, counselling services, and educational resources [1]. The government’s current initiative to ensure early childhood education, which has been found to have a negative association with poor ECDI outcomes in this study, should be strengthened by addressing the existing challenges [27]. These challenges include improving poor infrastructure and addressing the shortage of skilled teaching professionals [28]. Additionally, it is crucial to prioritize the needs of children from poor households, as they are more likely to have migrating parents and caregivers who may have limited education and knowledge about early childhood care [29]. This study also revealed a correlation between poor ECDI outcomes and children from disadvantaged households, aligning with existing data from LMICs [3, 4]. Furthermore, it is important to focus on children whose parents have disabilities, as this study identified a link between poor ECDI outcomes and such cases, particularly when one parent is disable and the other is migrated [30, 31]. Taking these factors into consideration and implementing targeted interventions will be vital in addressing the specific challenges faced by these vulnerable children and ensuring their optimal early childhood development. These programs should aim to equip these vulnerable children with the necessary tools to navigate the emotional and developmental challenges that accompany parental migration [4]. By ensuring equal access to enriching learning environments, fostering nurturing social interactions, and maintaining consistent routines, children of migrating parents will get the crucial support they desperately require for wholesome development, even in the painful absence of their migrating parents [11].

This study has several strengths, as well as a few limitations. One of the key strengths is that it is the first study in Bangladesh to report on the impact of parental migration status on ECDI. Advanced statistical techniques were employed, and nationally representative survey data were analyzed, taking into account a range of confounding factors that were selected based on previous literature in low- and middle-income countries. Additionally, the categorization of ECDI followed the recommended guidelines provided by UNICEF. However, a major limitation of this study stems from the analysis of cross-sectional survey data, which means that the reported findings are correlational in nature and not causal. Furthermore, the data recorded in this study relied on self-reported responses with limited options for validating the data. This introduces the possibility of recall bias, although any such bias is likely to be random. Moreover, there are additional factors, such as the quality of early childhood education and dietary patterns, that may impact ECDI and were not considered in the model due to their unavailability in the survey data. Despite these limitations, this study serves to raise awareness among policymakers about the ongoing issue of parental migration and its impact on ECDI. It can inform evidence-based policymaking and the development of programs aimed at reducing the negative effects of parental migration on early childhood development.