Development and validation of the Hong Kong Early Childhood Development Scale – Short Form (HKECDS-SF)

Abstract

Background: Accurate assessment of early childhood development is essential for both clinical and research use. While there are well-used direct assessments of early childhood development in Hong Kong, they are often too long to be used in population-based large-scale studies. This study aims to develop and validate a brief direct assessment (‘Short Form’) based on the Hong Kong Early Childhood Development Scale (HKECDS). Methods: This is a cross-sectional study. Purposeful sampling of three kindergartens were recruited to represent a wide range of socioeconomic background. Trained research assistant conducted the HKECDS (283 sub-items). Children’s sociodemographic information was also collected. Iterative Rasch analysis was conducted to select items from the HKECDS to ensure good reliability and validity in the Short Form. Results: A total of 144 participants were recruited from three kindergartens. The original 283 subitems in the HKECDS have undergone the iterative Rasch model analysis. The fitting statistics have been stablished after 4 iterations of item selection, resulting in a total of 99 subitems (7 for personal and self-care, 12 for language, 12 for school readiness, 13 for cognitive, 12 for gross motor, 8 for fine motor, 13 for health and safety, 7 for moral, and 15 for society and environment), taking on average 20 minutes to complete. The Short Form has excellent criterion validity against the HKECDS long form (correlation 0.97). The Short Form has moderate-to-strong correlation (0.48–0.77) with children’s biological age and districtlevel socioeconomic status (Cohen’s d 0.25, p<0.0001), which are comparable to the HKECDS. The Short Form have slightly lower, but still excellent, person separation reliability (0.97) compared with the HKECDS (0.99). Conclusions: The Short Form of HKECDS is a brief instrument with excellent psychometric properties.