Survey Dataset on Sleep Patterns, Health Effects, and Lifestyle Factors in Bangladesh
Description
Description: Proper sleep is essential for maintaining physical and mental well-being. Daily activities, stress, and lifestyle choices can significantly impact sleep quality. Inadequate or irregular sleep can lead to fatigue, reduced productivity, and long-term health issues. In countries like Bangladesh, where work demands, social habits, and environmental factors often disrupt sleep routines, understanding sleep behavior is crucial. This dataset was collected through a structured questionnaire distributed via Google Forms and in-person interviews across various regions of Bangladesh. It captures a wide range of sleep-related behaviors and health indicators from individuals of different ages, occupations, and living environments. Key Features of the Dataset: The dataset includes a mix of categorical, numerical, and multi-select textual data. Key variables include: Demographics: Age range, gender, occupation, weight, height Sleep schedule: Bedtime and wake-up times on weekdays and weekends Sleep quality: Average sleep duration, time taken to fall asleep Sleep disturbances: Breathing difficulties, restlessness, medical conditions Behavioral factors: Reasons for staying up late (e.g., social media, stress) Health impacts: Side effects of poor sleep (e.g., fatigue, lack of focus) Coping strategies: Methods used to manage sleep deprivation (e.g., caffeine, exercise) Environment: Self-rated comfort of the sleeping environment (scale of 1–5) Usage: This dataset offers valuable insights for researchers, data scientists, and public health professionals. Potential applications include: Calculating BMI from height and weight to explore correlations with sleep quality Building predictive models using machine learning algorithms such as: Sleep Quality Classification (Logistic Regression, Random Forest, XGBoost) Sleep Duration Prediction (Linear Regression, Random Forest) Sleep Behavior Clustering (K-Means, DBSCAN) Coping Strategy Recommendation (ML-kNN, Content-Based Filtering) Fatigue and Focus Drop Prediction (XGBoost, SVM) These models can help identify at-risk individuals, inform public health interventions, and support personalized wellness recommendations. Data Sources: Data was collected from 2,610 individuals across Bangladesh, including university students, professionals, and residents of both urban and rural areas. Responses were gathered through online forms and face-to-face interviews, with all entries standardized via Google Forms. Dataset Size: Total entries: 2,610 Number of features: 20 File format: CSV File size: ~750 KB
Files
Institutions
- Daffodil International University