DISC Personality Assessment Dataset for Machine Learning Research

📊 Dataset Overview: DISC-Inspired Personality Questionnaire (ML Study)This repository contains the anonymised dataset for a machine learning–driven exploration of the DISC personality framework. The data was collected as part of a research project examining whether computational methods can enhance the classification accuracy, efficiency, and interpretability of traditional DISC-based assessments. 📌 About the DatasetThe dataset consists of responses to 40 personality items adapted from validated psychological inventories available via the International Personality Item Pool (IPIP). Items were drawn from four behavioural dimensions that loosely map onto the DISC model: Assertiveness (CPI/IPIP) Social Confidence (JPI-R/IPIP) Adventurousness (NEO/IPIP) Dominance (CPI/IPIP) Each participant rated 40 statements on a 5-point Likert scale: 1 = Strongly disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly agree All statements were phrased in the first person (e.g., "I express myself easily", "I like to visit new places"). 🧩 Item StructureThe questionnaire includes 10 items per DISC-related dimension: AS (Assertiveness): AS1–AS10 SC (Social Confidence): SC1–SC10 AD (Adventurousness): AD1–AD10 DO (Dominance): DO1–DO10 Items cover behaviours such as leadership, sociability, novelty-seeking, and assertiveness. 👥 Demographic InformationParticipants were asked to provide: Gender (selected from: Male, Female, Other, or Prefer not to say) Age (free-text entry; entries below age 18 were excluded) While no formal demographic quotas were enforced, the final sample reflects a reasonably balanced distribution across age groups and genders, supporting the generalisability of findings across diverse professional and life-stage backgrounds. To preserve anonymity, no personally identifiable information (PII) was collected. Only participants who confirmed the accuracy and research suitability of their responses were retained in the final dataset. 🔍 Purpose of the DatasetThis dataset supports analyses related to: Supervised classification (e.g., Logistic Regression, XGBoost) Feature selection using RFE Unsupervised clustering (e.g., K-Means, PCA, Silhouette analysis) Personality trait profiling and questionnaire reduction It serves as the foundation for a research manuscript submitted to Nature Humanities and Social Sciences Communications. 📂 File IncludedDISC.csv – cleaned and anonymised dataset (1001 rows × 47 columns) 📜 CitationIf using this dataset, please cite the corresponding research paper (link forthcoming upon publication).