Ensaios sobre a confiabilidade da escala de avaliação dos ambientes de aprendizagem dedicados à primeira instancia.

Abstract

The quality of early childhood education can be expressed through responsive interactions between adults and children and by learning environments, which in turn must be safe, stimulating, and holistic to foster child development (Raikes et al., 2023; Yoshikawa et al., 2013; Maia & Williams, 2005). Enhancing this quality is possible through monitoring (Yoshikawa et al., 2018), requiring the use of instruments for data collection. This work comprises two essays aimed at exploring the reliability of the “Escala de Avaliação dos Ambientes de Aprendizagens dedicados à Primeira Infância” (EAPI), which seeks to measure quality in early childhood education. The first essay seeks to analyze the reliability of the EAPI application in the continuous format of five days of application, through stability (over time), internal consistency (of the instrument), and equivalence (agreement among observers). The second uses a data panel to conduct an analysis of observer characteristics and their impact on the reliability of applicators' responses to the instrument. Previous knowledge of observers about child development and information regarding field data collection were used. It was found that the number of applications does not affect equivalence and stability; however, the number of observers influences reliability. In reapplications, it is ideal to diversify the observer to two or three different individuals, for better coefficients. Internal consistency was negatively affected by continuous applications but indicates stabilizing after four days. Finally, knowledge about child development, experience, and performance in certification contributed positively to reliability, while the weekly application load negatively impacted it.