How Social Media, Sleep, and Screen Time Affect Teen Mental Health

By shrijeetverma13 · June 14, 2026

This analysis examines what actually influences stress, anxiety, and depression in teenagers—and some findings may surprise you. We tested common…

The analysis produced two visualizations — a heatmap and a bar chart — showing average addiction levels across different age and gender combinations. The heatmap displays how addiction levels vary across every age-gender pairing, while the bar chart highlights the top 10 segments with the highest average addiction scores on a 1–10 scale.

The analysis explored which factors most strongly associate with lower academic performance scores. While the correlation computation encountered a processing issue, three data tables were successfully generated that capture the relevant factor relationships in your dataset.

There is a weak positive correlation (r = 0.175) between daily social media usage and depression among teens. This means that teens who spend more time on social media tend to have a slightly higher likelihood of being labeled as depressed, though the relationship is not very strong. Two bar charts were generated to visualize this relationship — one showing average daily social media hours by depression status, and another showing depression rates across different usage buckets (0–2 hrs, 2–4 hrs, 4–6 hrs, 6–8 hrs).

The analysis produced two visualizations and several data tables comparing mental health metrics — stress level, anxiety level, addiction level, and depression rate — across the three platform usage categories. A grouped bar chart shows the average scores for each metric side by side per platform, making it easy to spot which platform usage group tends to have higher or lower mental health burdens. A box plot further reveals the spread and distribution of scores within each group, highlighting variability beyond just averages.

The analysis examined how sleep hours and pre-sleep screen time relate to stress and anxiety levels. Three visualizations and several data tables were generated to explore these relationships. Interestingly, the data shows very weak (near-zero) correlations across all four relationships measured — meaning that in this dataset, sleep duration and pre-sleep screen time have negligible linear associations with stress and anxiety levels.

Based on the data, higher physical activity does NOT meaningfully reduce stress, anxiety, or depression among teens. The correlations between physical activity and all three mental health metrics are essentially zero, suggesting no strong relationship exists in this dataset. Two bar charts and supporting data tables were generated to visualize these findings.

The grouped bar chart and data tables break down average stress, anxiety, and addiction levels across all combinations of social interaction level (low, medium, high) and gender. The visualization clearly shows how these three mental health indicators vary by group, making it easy to spot patterns at a glance.

The IQR outlier detection analysis was completed for daily social media hours, screen time before sleep, and addiction level. The results show that no outliers were detected in any of the three columns using the 1.5x IQR rule. Box plots were generated showing the distribution of all three variables side by side, and a comparison chart was produced showing average depression and stress levels across groups.

The analysis compared social media platforms across five key metrics — daily social media hours, addiction level, stress level, anxiety level, and sleep hours — and produced a radar/spider chart, a grouped bar chart, and multiple comparison tables. These visuals show each platform's normalized profile side by side, making it easy to spot which platforms are associated with higher risk (more usage, higher stress/anxiety/addiction, less sleep) versus healthier patterns. The radar chart highlights each platform's shape across all five dimensions, while the bar chart shows the raw average values per platform.

The correlation analysis was successfully computed for all 10 numeric columns in the dataset. Two data tables were generated showing the correlation results. The analysis examined relationships between age, daily social media hours, sleep hours, screen time before sleep, academic performance, physical activity, stress level, anxiety level, addiction level, and depression label.