Early childhood teachers are increasingly asked to assess and support children's scientific thinking, yet little is known about what teachers actually attend to when they describe a child's thinking in their own words. This study examined how teachers' spontaneous narrative notes about preschool children correspond to their structured ratings of scientific thinking, with particular attention to hypothesis and experimentation skills. A convergent mixed-methods design was used. Teachers from 277 early childhood institutions in Indonesia rated 295 children aged 4–6 years on a 24-item instrument covering attention, working memory, problem solving, observation, and hypothesis and experimentation, and 225 child-specific open-ended notes were analysed with directed content analysis; the two strands were integrated through cross-tabulation, correlation, and regression. Hypothesis and experimentation was the lowest-rated dimension (M = 3.31), yet it was almost absent from teachers' narratives: no note mentioned prediction, hypothesising, or revising an idea, and only 11.1% referred to any scientific reasoning process. Notes were dominated by global evaluations (43.6%) and attention or behavioural-regulation concerns (22.7%). For children in the lowest quartile of rated hypothesis and experimentation, teachers wrote about attention or regulation problems in 30.6% of notes but about reasoning in only 4.8%; in the highest quartile, reasoning was mentioned in 23.9%. Attention concerns did not predict hypothesis scores once rated attention was controlled, whereas mentions of reasoning added unique information beyond all four other rated dimensions. The findings suggest that children's emerging scientific reasoning is largely invisible in teachers' everyday descriptions, and that difficulties in this domain tend to be described as problems of focus, although the cross-sectional, correlational design cannot establish whether this reflects genuine attentional constraints, teachers' perception, or both. Professional development should therefore train teachers to notice and document children's predictions, explanations, and revisions explicitly.
Copyrights © 2026