Comparison of Framework and Native Programming Effectiveness in Website Development Among Informatics Engineering Students

Abstract

This study investigates student perceptions regarding the comparative effectiveness of frameworks versus native programming approaches in website development education. The research aims to examine how informatics engineering students perceive the educational value, practical utility, and career relevance of both approaches. Using a quantitative research design, survey data was collected from 18 informatics engineering students at a university in Indonesia. The instrument measured perceptions across ten Likert-scale items (five for native coding and five for frameworks) and included open-ended questions about preferences and educational recommendations. Data analysis was performed using SPSS version 26, including validity testing, reliability analysis, descriptive statistics, and Pearson correlation analysis. Results revealed excellent internal consistency for the native coding perception scale (Cronbach's α = 0.837) and acceptable consistency for the framework scale (α = 0.676). Students strongly valued native coding for building programming foundations (M = 4.06) while appreciating frameworks for documentation quality (M = 4.44) and development efficiency (M = 4.22). Although 77.8% preferred frameworks for practical development, 72.2% recommended beginning with native coding to establish conceptual foundations before progressing to frameworks. A significant negative correlation was found between documentation quality and creativity limitations (r = -0.611). These findings support implementing a sequential learning approach that explicitly connects framework features to their native implementations, balancing theoretical understanding with industry relevance. While the small sample size limits generalizability beyond this context, the findings provide valuable insights for curriculum development in similar Indonesian higher education settings and suggest directions for future research with broader samples.