📊 Student Performance Analysis using xAPI-Edu-Data

📌 Project Overview

This project analyses factors influencing student performance using the xAPI-Edu-Data dataset, which contains 480 student records with 17 attributes (demographic, behavioural, and parental engagement).
We explored how absenteeism, parental involvement, and student engagement affect academic outcomes, and visualized findings using Tableau dashboards and storytelling features.

The study shows that:

• Higher parental engagement correlates with better academic outcomes.
• Absenteeism is strongly associated with lower student performance.
• Student engagement metrics (resource use, raised hands, discussions) are strong predictors of success.

This project highlights the value of data visualization for educators and policymakers by transforming raw data into actionable insights.

---

🎯 Objectives

• Examine proportions of student performance across different education levels.
• Identify the relationship between absenteeism and academic success.
• Analyse the impact of parental satisfaction and involvement.
• Compare engagement metrics across performance levels (Low, Medium, High).

---

📂 Dataset

• xAPI-Edu-Data.xlsx
  Contains demographic, behavioural, and engagement attributes with student performance labels (Low, Medium, High).

Attributes include:

• Demographics (Gender, Nationality, StageID, GradeID, SectionID, Topic, Semester)
• Parental factors (Relation, Satisfaction, Survey participation)
• Engagement metrics (Raised Hands, Resource Visits, Announcement Views, Discussions)
• Target variable: Class (L, M, H) → student performance level

---

🛠️ Methods & Tools

• Tool: Tableau (interactive dashboards, filtering, storytelling)
• Visualization idioms:
  • 100% Stacked Bar Chart → performance by stage, gender, nationality
  • Horizontal Bar Chart → absenteeism vs performance
  • Side-by-Side & Stacked Bar Charts → parental satisfaction & demographics
  • Side-by-Side Bar Chart → student engagement metrics

---

📊 Tasks & Contributions

Task 1 – Performance Levels Across Educational Stages

Visualized proportions of student performance (High, Medium, Low) by stage, gender, nationality. (100% stacked bar chart)

Task 2 – Absenteeism vs Performance

Analysed correlation between absence days (Under-7, Above-7) and performance categories. (Horizontal bar chart)

Task 3 – Parental Satisfaction & Involvement

Explored relationship between parental satisfaction and student performance across demographics. (Side-by-side & stacked bar charts)

✅ Task 4 – Student Engagement Metrics Across Performance Levels (Contribution by Alia Marliana)

Designed and implemented visualization of engagement metrics (Announcement Views, Raised Hands, Visited Resources, Discussions) across performance levels (Low, Medium, High).

Key Features:

• Side-by-Side Bar Chart in Tableau to compare engagement metrics across performance levels.
• Interactive filters for Topic and Section allow exploration of subgroup differences.
• Colour encoding for metrics to ensure clarity and comparability.
• Tooltips display exact average values for deeper analysis.

Findings:

• High-performing students consistently show higher Visited Resources and Raised Hands counts.
• Announcement Views and Discussions vary less across groups.
• Section-level filters revealed lower engagement in Section B.
• Topic-level filters showed Science consistently had the highest engagement.

Reflection:

• Side-by-side bar charts effectively compared engagement patterns.
• Limitation: filters allow subgroup exploration, but do not compare filters directly (e.g., Section A vs Section B side-by-side).
• Future work: add clustered column charts or scatterplots to allow more granular comparisons.