Examining STEM learning attitude and gender differences computational thinking among elementary school students

STEM learning attitudes and computational thinking (CT) are recognized as essential competencies for 21st-century education, yet empirical evidence linking these constructs among Indonesian elementary school students remains limited. This study aimed to validate a STEM Learning Attitude Scale and examine gender differences in CT skills among fifth-grade elementary school students in Indonesia. A quantitative survey design was employed, involving 266 students for attitude scale validation and 240 students for CT assessment. Data were collected using a 25-item Likert-scale questionnaire adapted from Sun et al. (2021) and an essay-based CT assessment adapted from El-hamamsy & Mondada (2022) and Richardo et al. (2023). Exploratory Factor Analysis (EFA) using Principal Component Analysis with Varimax rotation was conducted via IBM SPSS 29.0, and Confirmatory Factor Analysis (CFA) was performed using SmartPLS 3. Gender differences in CT were analyzed using an independent samples t-test, and a paired samples t-test was used to compare higher-order and lower-order CT skills. EFA yielded a seven-factor solution with 15 retained items explaining 53.151% of total variance (KMO = 0.834), and the scale demonstrated acceptable reliability (Cronbach's α = 0.889). Female students reported significantly more positive STEM attitudes in mathematics and technology and outperformed male students in overall CT skills. No significant difference was found between higher-order and lower-order CT. This study contributes a culturally validated STEM attitude instrument for the Indonesian elementary context and provides empirical evidence supporting gender-responsive STEM and CT learning strategies.