This repository contains the results of an Exploratory Data Analysis (EDA) on student academic performance. The analysis aims to identify key factors influencing student grades and engagement, providing data-driven insights that can be utilized by educational institutions.
The EDA process involved cleaning, transforming, and visualizing a student performance dataset to uncover patterns and relationships between various attributes (such as parental education, lunch type, and test preparation) and student scores in different subjects.
-
Data Import & Cleaning: Demonstrates how to load raw data and handle missing or inconsistent values.
-
Descriptive Statistics: Provides summary statistics for key variables.
-
Data Visualization: Includes various plots (e.g., bar charts, box plots, scatter plots) to illustrate relationships between factors and academic performance.
-
Key Factor Identification: Highlights the most impactful factors affecting student scores, such as:
-
Correlation between parental education level and student performance.
-
Influence of lunch type on academic outcomes.
Actionable Insights: Derives recommendations for schools to improve grades and engagement based on the analytical findings.
- Test Preparation Course: Students who completed the test preparation course showed significantly higher mean scores across all subjects (Math: 69.83, Reading: 73.98, Writing: 74.53) compared to those who did not (Math: 64.92, Reading: 67.31, Writing: 65.31). This indicates a positive impact of such courses.
- Parental Level of Education: There is a clear increasing trend in mean scores across all subjects as the parental level of education increases, from 'high school' to 'master's degree'. This suggests that parental education is a strong predictor of student performance.
- Lunch Type: Students who received 'standard' lunch had notably higher mean scores in all subjects (Math: 70.24, Reading: 71.84, Writing: 71.02) compared to those with 'free/reduced' lunch (Math: 60.08, Reading: 65.76, Writing: 64.22).
- Promoting Test Preparation Programs: Given the significant positive impact, schools could actively promote and encourage student participation in test preparation courses, and potentially offer more accessible options for all students.
- Engaging Parents: Recognizing the influence of parental education, schools could implement programs that encourage parental involvement and provide resources to support parents in fostering a conducive learning environment at home. This could include workshops or informational sessions for parents on how to best support their children's academic journey.
- Addressing Socioeconomic Factors: The correlation between 'lunch type' and scores suggests that socioeconomic factors might play a role in student performance. Schools could investigate and address the underlying reasons for this disparity, potentially through targeted support programs or initiatives for students from lower socioeconomic backgrounds. By focusing on these areas, schools can use data-driven strategies to implement targeted interventions and support systems aimed at improving student academic performance and engagement.