2025-10-29
Students Performance Analysis
Exploring the factors influencing students’ academic achievement using the UCI dataset.
Objectives include:
To understand how social, demographic, and academic factors such as study time, parental education, and absences affect final student grades.
The findings aim to help educators and policymakers design data-driven interventions for improving student success.
Education plays a vital role in shaping future opportunities, yet many students struggle to achieve their full potential. By analyzing the Student Performance Dataset from Portuguese schools, we can uncover patterns that explain why some students perform better than others. Why this analysis matters:
It helps identify key predictors of academic success.
Provides actionable insights for teachers and administrators.
Demonstrates how data analytics can be used for social good and educational improvement.
Higher study time leads to higher grade. The grade is highest when study time is 3 or 4 hours. Very slight difference is observed between grade of those who study 3 hours and those who study 4 hours. The grade is lowest for those who study only 1 hour and second lowest for those who study 2 hours.
Those parents who have no education, their children has highest average grade compared to all other students. Apart from them, there is a positive relationship observed between grade and parental education. Higher parental education leads to higher average grade.
The scatter plot attached below shows that there is a negative relationship between age and final grades. Younger students get better final grade compared to older students. As the students grow older, they might loose interest in studies leading them to score lower.
Older students tend to score slightly lower (r = –0.16), while higher parental education (medu, fedu) improves performance (r = 0.22, 0.15). Study time has a mild positive effect (r ≈ 0.10), but past failures show the strongest negative impact (r = –0.36).
Male students have a slightly higher average grade (10.9) compared to females (10.0). This small difference suggests that gender has minimal impact on overall academic performance.
The regression results show that failures have a strong and significant negative impact on grades (p < 0.001), meaning each additional failure lowers performance notably. Going out also negatively affects grades (p = 0.03), while other factors such as study time, absences, and alcohol use show no significant influence on final grades.
## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 11.94667456 1.09227189 10.9374549 1.916105e-24 ## studytime 0.23882046 0.26887745 0.8882131 3.749779e-01 ## failures -2.13656381 0.29747295 -7.1823802 3.539308e-12 ## absences 0.03211808 0.02722751 1.1796185 2.388766e-01 ## goout -0.45760245 0.21438420 -2.1344971 3.342993e-02 ## dalc -0.04934429 0.31801977 -0.1551611 8.767752e-01 ## walc 0.20392918 0.23785173 0.8573794 3.917659e-01 ## health -0.12898977 0.15631156 -0.8252094 4.097611e-01
The analysis reveals that students who study longer generally achieve higher grades, with performance peaking among those studying 3–4 hours daily. A clear positive link is observed between parental education and student achievement, though an anomaly shows slightly higher grades for children of uneducated parents. Younger students tend to perform better, as grades decline modestly with age (r = –0.16). Gender differences are minimal, with males averaging 10.9 and females 10.0. Failures strongly and significantly reduce grades (p < 0.001), and frequent social outings also harm performance (p = 0.03). Overall, academic success is most influenced by study consistency, fewer failures, and supportive educational backgrounds.