From AI reliance to epistemic maturity: A longitudinal latent profile analysis of pre-service teachers in Vietnam

Dataset Description Title From AI reliance to epistemic maturity: A longitudinal latent profile analysis of pre-service teachers in Vietnam Description This dataset contains longitudinal survey data collected from pre-service teachers in Vietnam across four measurement waves within one academic year. The dataset captures how learners engage with AI-supported learning environments through the interaction of trust in AI, epistemic curiosity, need for cognition, metacognitive self-regulation, and peer feedback orientation. The structure supports advanced person-centered and longitudinal analyses, including latent profile analysis (LPA), latent transition analysis (LTA), and longitudinal structural equation modeling. The dataset enables examination of how configurations of learning engagement evolve over time and how these configurations relate to academic outcomes in AI-mediated contexts Data Structure The dataset is organized in wide format, with each row representing one participant and columns representing repeated measures across four time points: T1: Beginning of Semester 1 T2: End of Semester 1 T3: Beginning of Semester 2 T4: End of Semester 2 Variables 1. Control Variables id: Unique participant identifier gender: 0 = male, 1 = female year_of_study: Year 1, 2, or 3 frequency_of_ai_use: Frequency of AI usage gpa_start_year: GPA at the beginning of the academic year 2. Core Constructs (T1–T4, Likert 1–7) Trust in AI (TAI) Items: TAI1–TAI4 Reflects perceived reliability and reliance on AI systems Epistemic Curiosity (EPC) Items: EPC1–EPC4 Reflects motivation for knowledge exploration Need for Cognition (NFC) Items: NFC1–NFC4 Reflects preference for effortful thinking Metacognitive Self-Regulation (MSR) Items: MSR1–MSR4 Reflects planning, monitoring, and strategy adjustment Peer Feedback Orientation (PFO) Items: PFO1–PFO4 Reflects engagement with peer feedback processes 3. Derived Variables *_mean_T1 to *_mean_T4: Composite scores (mean of items) gpa_end_year: GPA at the end of academic year gpa_delta: Change in GPA across the year Missing Data Missing values are coded as 999 Recommended handling: Python: convert to NaN Mplus: MISSING = 999 Sample Characteristics Initial sample: approximately 588 participants Final sample: approximately 568 participants Population: pre-service teachers enrolled in teacher education programs in Vietnam Analytical Applications The dataset is suitable for: Latent Profile Analysis (LPA) Latent Transition Analysis (LTA) Longitudinal SEM (CLPM, RI-CLPM, LGCM) Machine learning classification and prediction Ethical Considerations Participation was voluntary Data were anonymized prior to analysis No personally identifiable information is included