RITA-T in toddlers with developmental delay: analysis code and verification log
Description
Complete analysis code for a two-centre cross-sectional study of 97 toddlers aged 18–36 months (50 with global developmental delay in whom autism was clinically excluded, 47 typically developing). The R script reproduces every number reported in the article, runs 38 automated data checks before analysis, and writes a verification log. Participant-level data are not deposited in keeping with the approved ethics protocol; de-identified data are available from the corresponding author on reasonable request. Also included: inter-rater agreement script and its de-identified input, graphical abstract code, codebook and corrections log.
Files
Steps to reproduce
1. Install R 4.3 or later with the packages readxl, dplyr, sandwich and lmtest. All other methods are implemented in base R; psych, pROC and BayesFactor are optional and used only for cross-checks. 2. Place rita_pipeline.R in a working directory together with the source data file rita-veri_v2.xlsx (not deposited; available from the corresponding author under a data use agreement) and, for the by-centre analyses, merkez.csv (two columns: sequence number, centre A/B). 3. In R, set that directory as the working directory and run: source("rita_pipeline.R", encoding = "UTF-8") The script first runs 38 automated checks on the data file (dimensions, column positions, item-block widths, value ranges, column-sum fingerprint, recomputation of every derived scale score). If any check fails it stops and names the check. It then writes dogrulama_log.txt (every statistic reported in the article, in section order), the result tables as CSV files, roc_egrileri.pdf (Figure S1) and bayraklar_gerilik.csv. Runtime is about one minute. Bootstrap procedures use fixed seeds. Without merkez.csv, Section 12 and the centre-dependent lines of Section 13 are skipped and the log says so. 4. Inter-rater agreement (Supplementary Table S6): source("irr_analizi.R") in a directory containing irr_video_puanlari.csv (included). Writes irr_sonuclar.csv and prints the full report. 5. Graphical abstract: python3 graphical_abstract.py in a directory containing bayraklar_gerilik.csv (included; also regenerated by step 3). Writes graphical_abstract.svg and stops if the counts do not match the article (50 / 10 / 16 / 2 / 24). Verified to give identical output on Windows 11 (R 4.6.1) and Linux (R 4.3.3), under C, C.UTF-8 and en_US.UTF-8 locales.