Abstract
Early childhood education increasingly emphasizes child agency and play-based learning, yet the systematic use of longitudinal patterns in children's self-selected activities as sources of observational information remains comparatively underexplored, particularly within culturally situated Islamic early childhood education. This study examined the initial implementation of the Prasmanan Activity Learning (PAL) Framework, a structured child-choice-based longitudinal observation framework developed in an Islamic preschool in Yogyakarta, Indonesia, within an ongoing Educational Design Research program. Naturalistic longitudinal records from 11 children, whose recorded baseline ages ranged from 2 years 11 months to 5 years 9 months, were examined within a six-month implementation window. PAL organizes the learning environment into four interconnected activity areas: Sensory and Emotional Regulation; Motoric and Body Movement; Fine Motor and Constructive Play; and Imaginative, Social and Meaning Play. Children could select, repeat, combine, and transition between activities while educators documented activity choice, duration, affective expression, social participation, independence, transitions, and salient behavioral events. A qualitative longitudinal multiple-case analysis used within-case pattern identification followed by cross-case comparison. Findings revealed identifiable but dynamic individual longitudinal engagement patterns. Repeated activity did not necessarily indicate static preference; similar materials frequently acquired increasing complexity through manipulation, construction, representation, symbolic transformation, narrative development, and social coordination. Duration, recurrence, transitions, initiation source, social participation, and contextual affect provided observational information beyond area-selection frequency alone. The findings provide initial observational support for the proposition that meaningful child choice within a responsive prepared environment can make individual patterns of engagement visible over time. These findings do not establish causal or psychometric validity and should not be interpreted as empirical verification of the theological premises underlying PAL. PAL is therefore proposed as an emerging culturally situated longitudinal educational observation framework rather than a validated developmental assessment instrument.