Validation of an Emotional Model in Programming Learning in the University Context
Journal
Lecture Notes in Computer Science
Learning and Collaboration Technologies
ISSN
03029743
Date Issued
2026
Type
text::conference output::conference proceedings::conference paper
Abstract
Programming learning in higher education constitutes a complex academic experience, characterized by continuous cycles of problem solving, trial and error, feedback, and performance assessment. Beyond cognitive demands, affective and motivational factors are involved, influencing students’ persistence, engagement, and self-regulation. In this context, the aim of this study was to conduct the initial validation of an emotional model for programming learning in a university population through structural equations. The model was structured into five latent factors: academic emotions, motivation (intrinsic and extrinsic), emotional self-perception, the emotional climate of the educational environment, and emotional engagement. A quantitative, non-experimental, cross-sectional design was implemented with a sample of 159 university students enrolled in programming courses in Ecuador. Data were collected using a 100-item online questionnaire in English (20 per dimension), with a five-point Likert scale. Analyses were conducted in SPSS and AMOS using maximum likelihood estimation, assessing sampling adequacy, internal consistency, model fit, and convergent and discriminant validity. The results showed high sampling adequacy (KMO = 0.898) and strong internal consistency of the instrument (overall α = 0.988). The model demonstrated satisfactory overall fit (CMIN/DF = 2.371) and significant covariances among factors, highlighting the associations between climate and emotional engagement and between motivation and emotional engagement. Convergent validity was adequate; however, discriminant analysis suggested high proximity between some constructs, which may reflect conceptual overlap at this initial stage. The model provides robust psychometric evidence for characterizing emotional dimensions linked to programming learning. Future work should refine and reduce items, evaluate model stability across other samples, and strengthen discrimination among factors to optimize its application in real-world university teaching contexts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.