Narrative coding is a pedagogical approach that combines storytelling with classroom programming to enhance computational thinking (CT) through meaningful projects for children. In young learners, integrating storytelling with tangible interfaces improves access to digital coding. The Erasmus+ STORYCODE project (storycode.eu) focuses on stimulating narrative coding through tangible programming and storytelling, examining its impact on CT development within STEAM education in kindergarten and primary education. This study explores the effects of the STORYCODE narrative coding Lab on CT development through an experimental, mixed-method, and cross-sectional design, involving 140 Spanish students from 2nd year of kindergarten to the 3rd year of primary school. Before the experiment began, all students were assessed on their visuo-perceptual reasoning and CT skills. Additionally, teachers evaluated each student's social-relational, CT, and narrative skills. Subsequently, participants were randomly assigned to either an experimental group, which took part in the STORYCODE narrative coding Lab over five weeks, or a control group, which followed the regular academic curriculum. The narrative coding Lab comprised 10 collaborative sessions, designed to progressively integrate unplugged coding activities, a tangible coding tool for early childhood education (i-code), and storytelling-based programming. Within the experimental group, each class was further divided into collaborative mini-groups of 3-5 students, following a standardized protocol. After completing the intervention, all students were reassessed on their CT skills. Additionally, children in the experimental group provided feedback on their satisfaction with the Lab. The study hypothesis was that students who participated in the Lab would achieve higher post-test CT scores compared to their pre-test scores and the post-test scores of the control group. Results showed improved CT scores in the 1st year of primary education (P1), with significant differences favoring the experimental group. In contrast, in the 2nd year (P2), the control group showed greater gains. Pre-test CT scores were the strongest predictors of improvement across all grades, while the experimental condition had a significant effect only in P1. Moreover, only in P1, visuo-perceptual reasoning scores correlated with both CT performance and teacher-rated students’ learning and computational competencies, suggesting an emerging alignment between students’ reasoning, CT task performance, and teacher perception of students’ skills. When considering all grades together, moderate to strong correlations were also observed between students’ visuo-perceptual reasoning and teacher ratings, indicating a broader association across the sample. Overall, students' satisfaction with the Lab was high, reflecting a largely positive experience. These findings suggest that narrative coding may effectively support the development of CT, particularly in P1. While the strongest improvements were observed in P1, positive trends in other age groups and gains in some control groups point to both intervention effects, contextual and developmental factors. The consistent role of baseline CT levels and the association with visuo-perceptual reasoning underscore the importance of cognitive readiness in CT acquisition. These insights may inform future research and educational designs aimed at integrating CT in early learning through engaging, developmentally appropriate approaches.

Keywords: Narrative Coding, Computational Thinking, Storytelling, Tangible Programming, Early Childhood Education, STEAM Education, Educational Technology.