Early childhood is a period in which environmental influences shape long-term developmental trajectories. A safe home environment, responsive caregiving, and opportunities for early learning constitute foundational supports for cognitive, socioemotion...
Early childhood is a period in which environmental influences shape long-term developmental trajectories. A safe home environment, responsive caregiving, and opportunities for early learning constitute foundational supports for cognitive, socioemotional, and physical development. Within households, norms that justify domestic violence increase the perceived acceptability of violence and may reinforce adverse parenting behaviors—such as corporal punishment and psychological aggression—that undermine early development. The Early Childhood Devlopment Index 2030 (ECDI 2030), developed by UNICEF, functions as a population-level composite indicator enabling policy monitoring and comparisons across countries and cohorts. In Nigeria, high levels of agreement with statements justifying wife-beating have been reported, indicating that the household normative environment remains a salient challenge. Prior studies have frequently relied on the earlier ECDI version or multi-country pooled analyses, leaving limited single-country evidence that leverages the updated ECDI 2030. In this context, estimating the association between women’s attitudes justifying domestic violence and children’s ECDI 2030 on-track status using nationally representative data provides relevant evidence for Nigeria’s child development and gender-equality policies and for tracking progress toward the SDGs.
The objective of this study is to estimate the association between mothers’ attitudes justifying domestic violence and ECDI 2030 on-track status among children aged 2–4 years in Nigeria, using data from the UNICEF-led 2021 Multiple Indicator Cluster Survey (MICS). Specifically, it (1) assesses the association between a binary indicator of justification (any endorsement vs. none; H1) and on-track status, and (2) examines a dose–response pattern by the justification score (0–6; H2), employing logistic regression with the complex survey design specified. Analyses adjust for child-, mother-, and household-level sociodemographic and socioeconomic covariates identified from prior literature, reporting adjusted odds ratios (aOR) with 95% confidence intervals. Given the cross-sectional observational design, interpretations are restricted to associations rather than causality.
A nationally representative cross-sectional analysis was conducted using Nigeria MICS 2021, specifying the complex survey design. The analytical unit was child–mother pairs linking co-resident mothers (15–49 years) to children aged 2–4 years via household ID and line number; after stepwise exclusions, the final sample comprised 6,469 observations. The outcome was ECDI 2030 on-track status derived from EC21–EC40; multi-category items (EC39–40) were dichotomized per the manual, and “don’t know” responses were recoded to “No” following guidance. The exposure was the mother’s attitude justifying domestic violence, used as (i) a binary indicator (any endorsement vs. none) and (ii) a summed score (0–6; DV1A–F). Covariates included child age/sex, mother’s age/highest education, household wealth quintile (1–5), and urban/rural residence. Descriptive statistics reported weighted frequencies/percentages with Rao–Scott χ² tests and design-adjusted t-tests. The main analysis employed design-based logistic regression estimating Model 1 (binary exposure) and Model 2 (continuous score), reporting adjusted odds ratios (aOR) with 95% confidence intervals. Robustness was assessed through a sensitivity analysis categorizing the score as 0/1–2/3–4/5–6, and multicollinearity was examined using variance inflation factors. To probe a proximal pathway, an auxiliary analysis using Nigeria DHS 2018 women’s data estimated the association between the justification score (0–5; v744a–e) and past-12-month IPV (reconstructed from d105a–j) with the DHS complex design, reporting aORs. All analyses were performed in R 4.3.3 with RStudio 2023.12.1 (Windows).
In Nigeria MICS 2021, a negative association was observed between mothers’ attitudes justifying domestic violence and children’s ECDI 2030 on-track status (ages 2–4). In Model 1, the adjusted odds of being on track were significantly lower among children whose mothers endorsed any justification (aOR < 1). In Model 2, each 1-point increase in the justification score (0–6) was associated with a decrease in the aOR, indicating a dose–response pattern (unadjusted OR = 0.857 vs. adjusted aOR = 0.935, reflecting attenuation after covariate adjustment). The household wealth index showed a consistently positive gradient, whereas the statistical significance of maternal education and urban/rural residence weakened after adjustment. Sensitivity analyses using the categorized score (0 / 1–2 / 3–4 / 5–6) reproduced the same direction of effects; the highest score group (5–6) showed marginal significance (p = 0.055) and the small sample size in this group suggested potentially unstable estimates. The study is based on a cross-sectional design; therefore, inferences are restricted to associations rather than causality.
Using nationally representative cross-sectional data from Nigeria, a consistent negative association was identified between mothers’ attitudes justifying domestic violence and children’s ECDI 2030 on-track status (ages 2–4). The direction of association was consistent across operationalizations of the justification measure (binary, continuous score, and categorized score), and a dose–response pattern was observed whereby higher justification scores were associated with lower adjusted odds of being on track. Specifically, children of mothers who justify domestic violence were significantly less likely to be on track for ECDI 2030, and the odds of being on track decreased with every 1-point increase in the justification score. Although this persistence after adjustment for socioeconomic covariates suggests a potential independent association of the normative indicator, causal interpretation is not warranted given the cross-sectional design. Synthesizing prior studies and relevant theoretical discussions, these findings may be explained by two pathways: first, attitudes justifying domestic violence may reflect patriarchal gender norms and limited decision-making power, thereby diminishing parental provision of developmental stimulation; second, they may function as a mechanism that deteriorates children's learning and safety environments by triggering resource loss, increased stress, and maternal gatekeeping in conjunction with actual domestic violence occurrence. The validity of this violence-mediated pathway was further strengthened by confirming the association between justification attitudes and actual domestic violence occurrence using data from the 2018 Nigeria Demographic and Health Surveys (NDHS), supported by USAID. Methodologically, leveraging recent nationally representative data and one-to-one mother–child linkage enabled estimation of the association between normative environments and early childhood development, and the consistency of findings across alternative variable definitions and model specifications strengthens the study’s academic contribution. The six-item justification index also captures a broader range of norms reported in the country context, extending the measurement scope.
Several limitations should be noted. Due to the cross-sectional design, the possibility of the opposite direction of the hypothesized pathway and unmeasured common causes cannot be excluded. Residual confounding is possible given unmeasured factors such as actual IPV victimization, intra-household decision-making power, and partner controlling behaviors. The composite, cutoff-based structure of ECDI 2030 may mask domain-level heterogeneity; reliance on self- or caregiver-report introduces risks of social desirability and recall bias; and complete-case analysis may be vulnerable to bias if missingness was nonrandom. In addition, the relatively high endorsement of the ‘sleep with another man’ item concentrates movement from score 0 to 1 and leaves the upper score range (5–6) sparse, which may yield unstable estimates for that group; furthermore, direct comparisons of prevalence or effect sizes with multicountry studies that use the standard five-item index are limited due to differences in indicator composition.