This project focuses on analyzing student behavior, engagement levels, and feedback data to uncover insights and suggest improvements for an online learning platform.
The analysis is based on three CSV files:
- students.csv: Contains student demographics and enrolment details.
- course_activity.csv: Logs time spent and completion rates for each student-course interaction.
- feedback.csv: Includes student ratings and written feedback per course.
- Python
- Pandas, NumPy
- Matplotlib, Seaborn
- Jupyter Notebook
- Handled missing and inconsistent values
- Converted date columns to datetime format
- Removed duplicates and standardized column types
Key questions explored:
- Average course completion rate
- Courses with highest and lowest engagement
- Engagement trends by age group and location
- Course-wise feedback ratings
- Correlation between completion rate and satisfaction
- Bar charts for average time spent per course
- Scatter plots to analyze engagement across age groups
- Line plots showing monthly activity and completion trends
- Course-wise rating comparisons
- Some age groups showed lower completion despite high login time.
- Location-based differences in course engagement were evident.
- Courses with higher ratings generally had better completion rates.
- A few courses had high drop-off after 30% completion — signaling content gaps.
- Engagement time strongly correlated with overall satisfaction.
- Optimize low-performing courses with high drop rates
- Personalize engagement strategies by age and region
- Use feedback sentiment analysis to improve course structure