Investigation of the Cross-National Equivalence of a Measurement of Early Child Development
- First Online:
- Cite this article as:
- Duku, E., Janus, M. & Brinkman, S. Child Ind Res (2015) 8: 471. doi:10.1007/s12187-014-9249-3
- 248 Downloads
The Early Development Instrument (EDI) is a measure of children’s developmental health at school entry in five domains, developed in Canada. A shorter, 48-item version of the EDI covering the same domains, created based on Canadian population data was used. The present study examined the psychometric properties of the domains of the 48-item EDI and the invariance of the measurement models of the domains of the 48-item EDI by sex and across the two South Asian countries, the Philippines and Indonesia. Data for 946, 3 to 5-year-olds in the Philippines, and for 3244, four-year-olds in Indonesia were used. The Strengths and Difficulties Questionnaire (SDQ) subscales were used for purposes of testing concurrent validity. The internal consistencies of the domains of the EDI except for Physical Health and Wellbeing were acceptable. Concurrent validity of the EDI domains with the SDQ subscales was acceptable and associations were in the expected directions. Modifications to the measurement models for the domains were made, where necessary, to allow for cross-loading items across the domains and to improve fit indices without over-fitting. Multiple group analysis using categorical confirmatory factor analysis within the Structural Equation Modeling framework showed that the models for the social competence and emotional maturity domains fit very well across sex and across the two countries. The findings are discussed in the context of requirements and conditions for the validation in modification and application of measures in international contexts.