Linguistic Analysis Task (LAT): De-identified Data, Research Instruments and Analysis Code for a Digital Inductive Model Study in Pre-Service French Teacher Education

This dataset accompanies the article “From Concordance to Confidence: A Sequential Explanatory Evaluation of a Digital Inductive Model for Linguistic Thinking in Pre-Service French Teacher Education” (Mekheimer & Abdelaal), submitted to Humanities and Social Sciences Communications. Contents: (1) LAT_main_study_matched_N55.csv: pre- and post-test Linguistic Analysis Task (LAT) scores (pattern recognition 0–5, hypothesis formulation 0–5, metalinguistic reflection 0–10, total 0–20) for 55 pre-service French as a Foreign Language teachers who completed a ten-week corpus-supported inductive intervention at Beni-Suef University, Egypt (spring 2025); (2) LAT_validation_indicators_N300.csv: twelve analytic LAT indicators (PR1–PR4, HF1–HF4, MR1–MR4) for a separate validation sample of 300 pre-service language teachers, used for confirmatory factor analysis; (3) codebook.csv; (4) reproduce_statistics.py: a Python script that reproduces all statistics reported in the article (paired t tests, confidence intervals, Cohen’s d_z, ceiling counts, Cronbach’s alpha, CFA); (5) Supplementary_Information_S1-S4.docx: the LAT instrument and scoring rubric, reflective-journal prompts, focus-group guide, and data description. All records are de-identified with arbitrary study codes; no names, student numbers or demographic variables are included. Reflective-journal texts and focus-group transcripts are not deposited, to protect participant confidentiality. Ethical approval: Faculty of Education Research Ethics Board, Beni-Suef University (BSU-FoE-003-01-3-2025). License: Creative Commons Attribution 4.0 International (already selected) Copyright: © 2026 Mohamed Mekheimer and Fatma Abdelaal