╔══════════════════════════════════════════════════════════════════════════╗ ║ PEER REVIEW SUPPORTING DOCUMENTATION PACKAGE ║ ║ "QR Code-Mediated Instruction Enhances Writing Performance and ║ ║ Learner Engagement in Advanced EFL Learners" ║ ╚══════════════════════════════════════════════════════════════════════════╝ CONTENTS OF THIS PACKAGE ────────────────────────────────────────────────────────────────────────── raw_data.csv Raw data as collected: individual participant scores from both raters, OQPT scores, sub-scores on all four IELTS writing dimensions (pre & post), and demographic variables. cleaned_data.csv Analysis-ready dataset. Contains all variables from raw_data.csv plus derived variables: - writing_pre / writing_post (consensus scores, mean of two raters) - gain_score (posttest − pretest) - improved (binary: 1 if gain > 0) - group_binary (0=Control, 1=Experimental) - sub-score gains (ta_gain, cc_gain, lr_gain, gr_gain) - outlier flags (z > 3 threshold) Raw rater columns are excluded (see raw_data.csv). qualitative_themes.csv Frequency data and operational definitions for all seven themes identified in the thematic analysis of semi-structured interviews (n=20). Includes representative quotes, category labels, valence, and coding rules used by both coders. codebook.csv Complete variable dictionary covering all three datasets. Includes variable name, data type, value range, description, and transformation notes for every column. statistical_analysis.py Fully reproducible Python script that loads cleaned_data.csv and raw_data.csv and replicates all analyses reported in the manuscript: 0. Descriptive statistics 1. Assumption testing (Shapiro-Wilk, Levene, outlier screening) 2. Baseline group equivalence (OQPT & pretest) 3. Within-group paired t-tests 4. Between-group independent t-test (posttest) 5. ANCOVA (posttest ~ group + pretest) 6. Gain score analysis 7. Multiple regression (gain ~ group + pretest) 8. Non-parametric robustness checks (Wilcoxon, Mann-Whitney U) 9. Subcomponent analysis (4 IELTS dimensions) 10. Inter-rater reliability (ICC) README.txt This file. ────────────────────────────────────────────────────────────────────────── HOW TO REPLICATE THE ANALYSIS ────────────────────────────────────────────────────────────────────────── Requirements: Python 3.8+, pandas, numpy, scipy Install dependencies (if needed): pip install pandas numpy scipy Place all files in the same directory, then run: python3 statistical_analysis.py All results will be printed to the console and match the values reported in the manuscript tables and text. ────────────────────────────────────────────────────────────────────────── VARIABLE NAMING CONVENTIONS ────────────────────────────────────────────────────────────────────────── Prefix Meaning ───── ─────────────────────────────────────────────────────────────── ta_ Task Achievement (IELTS dimension 1; 0–9 scale) cc_ Coherence and Cohesion (IELTS dimension 2; 0–9 scale) lr_ Lexical Resource (IELTS dimension 3; 0–9 scale) gr_ Grammatical Range and Accuracy (IELTS dimension 4; 0–9 scale) _pre Pretest (baseline) measurement _post Posttest (post-intervention) measurement _gain Derived: posttest − pretest _r1 Rater 1 score (raw data only) _r2 Rater 2 score (raw data only) Composite writing score (writing_pre / writing_post): - Mean of Rater 1 and Rater 2 composite scores - Composite = (ta + cc + lr + gr) rescaled to 0–20 [sum of 4 × 0–9 sub-scores = max 36; divided by 36 × 20] - Reported on 0–20 scale throughout the manuscript ────────────────────────────────────────────────────────────────────────── PARTICIPANT CODING ────────────────────────────────────────────────────────────────────────── C01–C30 Control group participants E01–E30 Experimental group participants ────────────────────────────────────────────────────────────────────────── QUALITATIVE DATA NOTE ────────────────────────────────────────────────────────────────────────── Semi-structured interviews were conducted in Persian and subsequently translated into English by the first author (bilingual, IELTS 8.5). Back-translation was carried out by an independent bilingual colleague to verify accuracy. Representative quotes in qualitative_themes.csv are the finalised English translations. Interview transcripts are not included in this package to protect participant confidentiality in accordance with the ethics approval (IR.ACECR.REC.1402.818). Full anonymised transcripts are available to editors and reviewers upon request through the journal's secure data-sharing mechanism. ────────────────────────────────────────────────────────────────────────── ETHICS AND DATA AVAILABILITY ────────────────────────────────────────────────────────────────────────── Ethics approval: IR.ACECR.REC.1402.818 All participants provided written informed consent. Data are anonymised; no personally identifiable information is included. Datasets are available from the corresponding author on reasonable request. ──────────────────────────────────────────────────────────────────────────