Survey Data on BSCS, DOSPERT, PSSS, SCL-90, and Family APGAR in Chinese Undergraduates

This dataset contains the fully preprocessed and reverse‑coded survey data used in “Interplay of Internal and External Factors in Adolescent Risk‑Taking: Insights from Latent Profile and Network Analyses,” submitted to Journal of Health Psychology. It includes responses from 1,159 Chinese first‑year undergraduates (M_age = 18.2, 62.3% female) collected between May and October 2023 at a university in Guangzhou.In addition to the main dataset, this repository includes supplementary figures referenced in the manuscript. All materials are provided to ensure transparency and reproducibility for peer reviewers and future researchers.ContentsDemographics:Participant IDAge, sex, only‑child status, urban/rural originHousehold monthly income bracketSubjective social and school status (MacArthur Scale scores)Psychosocial Measures:Brief Self‑Control Scale (BSCS): 7 items (reverse‑coded; higher = poorer self‑control)Perceived Social Support Scale (PSSS): 12 items (family, friends, significant others)Family APGAR: 5 items measuring family functionInterpersonal Sensitivity (SCL‑90 subscale): 9 items (reverse‑coded; higher = greater sensitivity)DOSPERT Risk‑Taking (RT) Subscale: 30 items across five domains (Ethical, Financial, Health/Safety, Recreational, Social)Data Structure & Processing:All scales scored according to original manuals; two reverse‑coding steps applied (BSCS, SCL‑90 interpersonal sensitivity)Cases with aberrantly fast or patterned responding were excluded (final N = 1,159)No missing data remain; raw item scores and composite subscale scores are providedVariables are labeled consistently for latent profile analysis (LPA) and subsequent network analysisUsage NotesIntended for replication of latent profile analyses (identifying four interpersonal profiles: SI, IH, IL, SC) and network models (bridge node detection)Each row represents one participant; columns include both item‑level and aggregated subscale scoresRecommended software: R (packages tidyverse, poLCA or mclust, qgraph/bootnet)