Measuring Early Childhood Development Among 4-6 year Olds: The Identification of Psychometrically Robust Items Across Diverse Contexts

• In Event: Validity for What? Challenges and Opportunities in Direct Assessment of Young Children Internationally and at Scale

Abstract

It is critical for governments, development partners, and parents to have useful information on early childhood development (ECD) outcomes, as without this information, programs and policies to support children and families are less likely to be as effective, efficient, and equitable as they could be. Valid and reliable measures of ECD outcomes and quality of early learning environments that can be implemented at scale across countries are critical both for monitoring purposes and to understand whether programs and policies are having their intended impacts on children and their families around the world.

There have been various global initiatives and measures developed to promote large-scale measurement of children’s development and quality of education in the preschool years. However, while there has been this proliferation of ECD outcome measurement efforts and data, it has been difficult to get an accurate view of how child development varies across the globe because measures used do not always capture the same constructs in the same way.

In order to link across tools and measurement efforts, there needs to be a common set of items embedded in each tool or effort. Identifying these items would allow for increased efficiencies and synergies across data collection efforts, for example through the provision of common guidance around data collection, management, and analysis, thus helping to ensure that national statistical offices and researchers are collecting comparable and relevant data with high-value information on child development in the preschool years. The <>’s Scaling Up Measurement in Early Childhood team has been working on the identification a core set of caregiver report items and a core set of child direct assessment items that measure key developmental domains for children aged 4-6, with adequate psychometric properties across multiple contexts.

The team compiled and harmonized ECD data collected from twelve low- and middle-income countries around the world using the same base caregiver report and child direct assessment tools. This process resulted in the data analysis of 16,015 caregiver-report questionnaires and 24,533 child direct assessments. The data analysis involved the use of classical test theory, item response theory, and factor analysis psychometric frameworks to obtain validity and reliability evidence for the interpretation and use of these ECD measurement tools. Results showed that a relatively small set of items in the child direct assessment and caregiver report tools yield robust psychometric properties for children aged 4 to 6 across countries.

This work represents a starting point for (i) linking across different early childhood measurement tools for children aged 4-6; (ii) increasing quality across measurement efforts; and (iii) facilitating the scale up of early childhood measurement. It is hoped that these efforts would help to advance understanding of universal and context-specific factors that underlie child development and thus help policymakers make decisions that ensure children receive quality early childhood care and education they need in order to reach their full potential.

Authors

• Diego Armando Luna Bazaldua, World Bank - District of Columbia
  Presenting Author

• Adelle Pushparatnam, World Bank - District of Columbia
  Non-Presenting Author