Academic Performance & Gaming Behavior Analysis: Data-Driven Insights from Student Lifestyle Patterns

By abhishek.verma75000 · May 21, 2026

Understanding how lifestyle habits influence academic success is becoming increasingly important in today’s digital world. This analysis examined the…

The analysis produced two visualizations comparing student grades across different gaming genres. A bar chart shows the average grades per genre, and a box plot reveals the full grade distribution for each genre, making it easy to spot which genres are linked to higher or lower academic performance.

Among study hours, sleep hours, and attendance, study hours is by far the strongest predictor of high grades. It shows a strong correlation of r=0.733 with grades and a standardized regression coefficient of β=0.730, meaning it explains the most variance in grades even when accounting for the other factors. Sleep hours comes in second (r=0.250, β=0.241), while attendance has the weakest influence (r=0.131, β=0.144). Two bar charts were generated showing both the correlation rankings and the standardized regression coefficients visually.

The analysis examined grades and gaming behavior across 9 age-gender combinations (16-18, 19-21, 22-24 × Male, Female, Other). Three bar charts and a scatter plot were generated showing how these metrics vary across groups, along with detailed summary tables.

The analysis reveals how addiction scores relate to academic performance and wellbeing. Three visualizations were generated: a grouped bar chart comparing average grades and attendance across addiction levels, a stacked bar chart showing stress level distributions by addiction group, and a scatter plot of addiction score vs. grades colored by stress level.

The analysis reveals a clear pattern: students with higher stress levels actually study more and achieve better grades on average. Three charts and summary tables were generated to visualize these differences across Low, Medium, and High stress groups.

There is a meaningful negative relationship between gaming hours and academic grades, with an overall Pearson correlation of -0.551. This means students who game more tend to earn lower grades. Three visualizations were generated: a scatter plot showing gaming hours vs. grades colored by stress level, a bar chart comparing average grades across gaming hour segments, and a grouped bar chart showing correlations across different student segments.

The analysis successfully produced two bar charts and supporting data tables breaking down key metrics by gaming genre. The first grouped bar chart compares average gaming hours, study hours, sleep hours, addiction score, and grades across all genres side by side. The second chart shows how many students belong to each genre. Data tables are also available with the full numeric breakdown per genre.

The Pearson correlation analysis was successfully completed for all 9 numeric columns. Two visualizations were generated: a full correlation heatmap showing relationships between all variables, and a horizontal bar chart highlighting how each variable correlates specifically with grades. The heatmap uses a Red-Blue color scale where blue indicates strong positive correlations and red indicates strong negative correlations, with coefficients annotated directly on each cell. The bar chart clearly shows which variables are positively (green) and negatively (red) associated with grades.

The analysis successfully produced visualizations and data tables breaking down the distribution of sleep hours and study hours across all students. Two charts were generated: an overlapping histogram with KDE curves showing the density distributions of both metrics side by side, and a grouped bar chart comparing average grades across Low, Medium, and High bins for each metric. Additional data tables provide the underlying statistics and binned grade averages.

The IQR outlier analysis (1.5x rule) was applied to addiction score, reaction time ms, and gaming hours. The results show that none of the three variables contain any outliers — all data points fall within the calculated IQR bounds. Box plots were generated showing the distribution of each variable, and a grade comparison table was also produced. Since no outliers were detected, the grade comparison between outlier and non-outlier groups only reflects the full dataset average of 66.18 for all three metrics.