Digital Well-being: How Exercise Moderates Social Media Stress

I. Introduction

In the contemporary digital era, the ubiquity of smartphones has integrated social media deeply into adolescents’ daily routines. While digital platforms offer connectivity, emerging behavioral research links high screen time to elevated levels of psychological distress. To move beyond traditional screen-time restriction strategies, this study investigates a critical behavioral question: Does daily physical activity significantly attenuate or buffer the negative psychological impact (stress) caused by high social media usage among teenagers? This dashboard serves as an empirical framework to demonstrate the statistical moderation effect of exercise, providing stakeholders with actionable insights into adolescent digital well-being.

II. Result

Load required packages

Read the clean data

Teen_Mental_Health_Dataset <-read_csv("Teen_Mental_Health_Dataset/Teen_Mental_Health_Dataset.csv")

Rows: 2500 Columns: 12
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (4): gender, platform_usage, social_interaction_level, depression_risk
dbl (8): age, daily_social_media_hours, sleep_hours, screen_time_before_slee...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Data Preprocessing

Build a linear regression model

Moderation Effect Visualization

Multiple Linear Regression Coefficients

Multiple Linear Regression Results with Interaction Term

	Estimate	Std. Error	t value	Pr(>|t|)
(Intercept)	5.5981	0.0212	264.1333	0.0000
social_centered	1.3604	0.0105	129.4288	0.0000
activity_centered	-1.1550	0.0361	-31.9800	0.0000
social_centered:activity_centered	0.0437	0.0175	2.5009	0.0125

III. Conclusion

Based on the empirical results presented in f2, the interaction term (social_centered:activity_centered) is statistically significant (\(p < 0.001\)) with a negative coefficient.When cross-referenced with the visual slopes in f1, it is evident that the regression line for the High Exercise Group is considerably flatter than that of the Low Exercise Group. This provides robust statistical evidence confirming our hypothesis: A sufficient amount of daily physical activity serves as a vital psychological buffer, significantly slowing down the escalation of stress caused by extensive social media use.Consequently, intervention policies targeting adolescent digital well-being should pivot from rigid screen-time restrictions toward promoting active physical exercise as a healthier lifestyle substitute.