F1000Research (Jan 2024)
Validation of the Dominican system for measuring early childhood development [version 2; peer review: 2 approved]
Abstract
Background The purpose of the study was to determine the psychometric properties of the Dominican System for Measuring Early Childhood Development (SIMEDID, for its Spanish acronym), to adjust the sequence of item presentation, and to provide age-standardized norms for each item, to enable policy and program managers to make decisions based on specific and structured data. Methods After approval from an ethics committee, a total of 948 children from 0 to 60 months participated in this study. Participants were evaluated on four early childhood development domains (gross motor, fine motor, language development, and socio-emotional development). The data were collected from November 2021 to February 2022, either at early childhood care centers or at home, using mobile devices that guided the evaluators through the screener. Data were later synced to a global database. Psychometric properties were calculated using Cronbach’s alpha and split-half parallel reliability. For reorganizing item presentation and to obtain age-standardized norms, we conducted a logistic regression analysis for each item on dependent variable item success, and independent variable age. Results The instrument showed excellent reliability and additional evidence of validity. The item presentation order was rearranged according to the probability of item success progression. In addition, the study characterized the expected evolution of item success probability across participants’ age. Conclusions SIMEDID is a valid and reliable instrument for depicting childhood development in national evaluations. Its integration with electronic platforms for national monitoring represents a cost-effective, time-efficient screening tool adapted to the Dominican sociocultural context. This represents a promising tool to strengthen strategies that support early childhood development.
