Understanding the short-, medium- and long-term impacts of early childhood interventions across the globe.

• In Event: Evidence of the benefits of early childhood programs on child development in international contexts

Abstract

Introduction
Poverty compromises the development of hundreds of millions of children in the developing world, at great cost to individuals and countries (Black et al., 2016). This has spurred growing interest in public investments to improve opportunities for disadvantaged children in the early years (Berlinski and Schady, 2015; Britto et al., 2017; United Nations, 2016). While studies have found that early interventions can alter developmental trajectories (Almond, Curie & Duque, 2018, Engle et al., 2011), questions remain about which types of interventions may be stronger. This paper reviews the international evidence on the benefits of early childhood interventions in the short-, medium- and long-term, building on Nores & Barnett’s (2010).

Goals
1) Capture the variation in national contexts, timing and realized outcomes of early childhood interventions
2) Summarize studies’ outcomes and characteristics on a common scale in ways that may help design intervention policies and programs.
3) Understand the medium and long-term persistence of effects.
Sample
Studies had to fit the following inclusion criteria: (1) sample of children must have received a treatment between the ages of 0 and 5. (2) include outcome measures on child development and/or parents in early education programs, (3) focus on intervention outcomes with at least one group receiving an early childhood treatment (whether it is stimulation, nutrition, cash transfer, preschool, or similar), (4) use quantitative analysis and provide enough data to calculate effect sizes, (5) conducted outside of the United States and Canada and (6) published by 2018. We used various databases: Education Full Text, Educational Resources Information Center, Dissertation Abstracts, Google Scholar, ScienceDirect, Journal Storage Archive , ProQuest, Pubmed, Wiley, and The World Bank. Searches included the following outcomes: child cognition, aggression, child development, conduct disorders, conduct problems, executive function, health outcomes, hostility, non-compliance, nutrition, self-regulation, social behavior, socio-emotional. Searches also included various delivery formats as are cash transfer, cash transfers, childcare, children, home visitation, early care education, early childhood, early development, early education, international, interventions, lottery, nutritional, nutritional outcome, preschool, stimulation, violence prevention.
Methods
We analyze 1,721 outcomes in 114 experiments (reported in 151 studies) across the globe. Studies are coded according to: type of intervention (cash transfer, nutritional, educational or mixed), sample size, design & duration, country, age group, implementing organization, timing (short, medium or long-term), decade in which it was carried out, publication outlet (or unpublished), and methods. Cohen's D and Hedges’ g effect sizes are calculated for four types of outcomes: cognitive, behavioral, health, and schooling outcomes. We summarize average effects across study and population characteristics for the four domains. We follow this estimating mean effect sizes using Random Effects (McEwan, 2015; Ringquist, 2013). We check for publication bias (Light & Pillemer, 1984; Ringquist, 2013).
Results
Preliminary analyses (Table 1) show that regardless of context, children in different countries and across various decades show substantial cognitive, behavioral, health and schooling benefits from early childhood interventions. The effect sizes are persistent over medium- and long-term outcomes. Interventions that have an educational or stimulation component evidenced the largest overall effects.

Authors

• Milagros Nores, National Institute for Early Education Research
  Presenting Author

• Raquel Bernal, Universidad de los Andes
  Non-Presenting Author