Project Overview This project analyzes social media usage patterns among students to understand their impact on mental health, sleep, academic performance, and addiction risk levels. Using Python data analysis and visualization techniques, the project converts raw survey data into meaningful insights and actionable recommendations.
The analysis focuses on identifying high-risk students, behavioral trends across age, gender, academic level, and country, and proposing detox strategies based on risk classification.
Dataset Description
Records: 705 students Features: 13 original columns (demographics, usage behavior, mental health indicators)
Tools & Libraries Used
Python Pandas – Data cleaning & transformation NumPy – Numerical operations Matplotlib – Data visualization Seaborn – Advanced visualizations
Project Workflow
1️⃣ Data Understanding & Cleaning
2️⃣ Exploratory Data Analysis (EDA) Descriptive statistics for key numerical features Distribution analysis by: Age Gender Academic Level Usage behavior comparison across demographics Country-wise addiction and usage trends
3️⃣ Feature Engineering
Risk Level Classification (Low / Medium / High) based on daily usage Performance Impact Flag based on academic performance Detox Strategy Recommendation based on: Usage hours Sleep duration Risk category
4️⃣ Data Visualization Visualizations created to support insights: Line charts for platform-wise usage Bar charts and pie charts for gender and age distribution Box plots for addiction score and usage patterns Scatter plots for mental health vs addiction Heatmap for correlation analysis Country-wise high-risk student analysis
Key Insights
Students with higher daily usage hours show: Lower sleep duration Higher addiction scores Poorer mental health High school and undergraduate students are at higher risk Females show slightly higher average daily usage High addiction scores strongly correlate with: Academic performance impact Increased social conflicts Most students fall into Medium Risk, followed by High Risk
Risk-Level Based Detox Strategy High Risk: Social media breaks Sleep improvement Gradual usage reduction
Medium Risk: Screen-time control Offline activity encouragement Awareness programs
Low Risk: Maintain healthy habits
Recommendations
Early identification of high-risk students Institutional awareness programs on social media addiction Parental and educator involvement Encourage offline social and physical activities Promote digital well-being education
Conclusion
This project demonstrates the ability to:
Clean and analyze real-world datasets Apply statistical and exploratory techniques Build meaningful visualizations Translate data insights into practical recommendations It showcases end-to-end data analytics skills suitable for Data Analyst / Business Analyst roles.