Libro de Actas del III Congreso Latinoamericano y del Caribe e Investigación en Educación Superior- LatinSoTL- 2025

249 engaged in clinical case analysis, Engineering students tackled complex technical problems, and Veterinary Medicine students participated in diagnostic simulations. Following the intervention, students completed the SPQ and CEQ questionnaires. The SPQ measured two dimensions: Deep Learning (AP) and Surface Learning (AS), while the CEQ assessed four dimensions: Good Teaching (BD), Workload (CT), Clear Goals and Standards (MC), and Appropriate Assessment (EA). The instruments were administered via Microsoft Forms, and informed consent was obtained from all participants. Ethical approval for the study was granted by the Scientific Ethics Committee of Universidad San Sebastián. Data analysis was conducted using Microsoft Excel, Jamovi (version 2.3.28), and STATA (version 14). The analytical process included: • Data Cleaning : Verification of missing values and outliers. • Reliability Analysis : Internal consistency was assessed using Cronbach’s alpha and McDonald’s omega for each dimension. • Exploratory Factor Analysis (EFA) : Conducted using minimum residual extraction and varimax rotation to identify underlying factor structures. • Suitability Tests : Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin (KMO) index were used to confirm the adequacy of the data for factor analysis. • Descriptive Statistics : Means, standard deviations, skewness, and kurtosis were calculated for each dimension. • Inferential Statistics : One-way ANOVA was performed to compare perceptions across academic levels (undergraduate vs. postgraduate) and across different programs. The EFA revealed six factors explaining 53% of the total variance, with the first factor accounting for 17.6%. Bartlett’s test yielded a chi-square value of 2090 (p < .001), and the KMO index was 0.806, indicating good sampling adequacy. Reliability coefficients were high across all dimensions (α > 0.70), with Good Teaching (BD) showing the highest internal consistency (α = 0.85). Confirmatory Factor Analysis (CFA) further validated the theoretical structures of the SPQ and CEQ, with fit indices indicating acceptable model fit (SPQ: CFI = 0.92, RMSEA = 0.06; CEQ: CFI = 0.90, RMSEA = 0.07). Results The instruments demonstrated high internal consistency, with Cronbach’s alpha of 0.898 and McDonald’s omega of 0.907. The mean response score was 3.80, with a standard deviation of 0.457, indicating homogeneity in student perceptions. EFA revealed six factors explaining 53% of the total variance, with the first factor accounting for 17.6%. Bartlett’s test yielded a chi-square value of 2090 (p < .001), and the KMO index was 0.806, confirming the adequacy of the data for factor analysis. Descriptive statistics showed that Deep Learning (AP) and Good Teaching (BD) received the highest mean scores (AP: 3.72; BD: 4.21), suggesting a strong preference for active, student-centered learning. Appropriate Assessment (EA) exhibited the highest variability (SD = 1.03), indicating diverse perceptions of evaluation methods. Correlational analysis revealed significant positive relationships between Deep Learning and Good Teaching (r = 0.32, p < 0.01), and between Deep Learning and