A Supervised Machine Learning Approach to Understanding Migration Decisions and Psychological Stress Among Youth.
Description
This dataset provides empirical insights into the migration decisions and associated psychological stress among Bangladeshi youth. It comprises survey responses from 2,614 individuals (aged 18–35), collected between January and April 2025 through a stratified convenience sampling method using Google Forms and face-to-face interviews. The dataset is critical for understanding the intersection of socio-economic aspirations and mental health in the context of youth migration. It captures diverse responses regarding migration interest, stress levels, coping mechanisms, and future plans from participants across various ages, occupations, and socio-cultural backgrounds in Bangladesh. Key Features & Variables: Demographics: Age group, Gender, Occupation. Migration Decisions: Decision status to migrate, intended duration, awareness of migration programs/scholarships. Influencing Factors: Role of social media, family pressure, peer influence, and mentors. Psychological Stress Indicators: Type and intensity of stress (e.g., visa anxiety, fear of rejection, family separation, emotional challenges). Coping Strategies: Methods used to manage stress (e.g., hobbies, physical activity, social support, online forums). Barriers to Migration: Financial constraints, lack of experience/qualifications, cultural resistance. Return Plans & Impact: Intention to return, perceived impact on family/community, trust in government support. Recommendation Likelihood: Willingness to recommend migration to peers (Scale 1–5). Files Included in this Repository: Cleaned_Youth_Migration_Data.csv: The primary dataset with standardized variable names (N=2,614). Data_Dictionary.csv: A codebook explaining each variable name and the corresponding original survey question. Survey_Questionnaire.pdf: The original blank questionnaire used for data collection. Readme.txt: Overview and usage instructions for the dataset. Potential Usage: This dataset is valuable for researchers, data scientists, policymakers, and mental health professionals. It supports: Machine Learning Models: Migration intent prediction (Logistic Regression, Random Forest), Stress level classification (SVM, Decision Tree), and Cluster analysis of motivations. Policy & Health: Designing support strategies for at-risk youth and understanding the psychological cost of migration.
Files
Steps to reproduce
This is a resubmission of the manuscript (Ref: DIB-D-25-02225). We have updated the Mendeley Data repository with the required Data Dictionary (Codebook), Questionnaire, and Readme files as requested in the previous decision letter.
Institutions
- Daffodil International University