The Impact of Social Media Usage on Mental Health Among Teens and Young Adults

Social media has become an integral part of daily life, especially for teens and young adults. Platforms like Instagram, TikTok, and Snapchat are commonly used for self-expression, socializing, and seeking validation. However, the increasing prevalence of social media usage has raised concerns about its potential impact on mental health. Studies have shown a correlation between heavy social media use and mental health issues such as anxiety, depression, and stress. Understanding the relationship between social media use and mental health is crucial for informing interventions and setting healthier usage guidelines.

Research Question and Hypothesis

The research question is: How does the duration/frequency of social media usage affect the mental health of teens and young adults?

The null hypothesis is: There is no significant relationship between the duration of social media usage and mental health outcomes in teens and young adults.

The alternative hypothesis is: Increased social media usage is positively correlated with higher levels of depression and anxiety among teens and young adults.

Literature Review

Nagata et al. (2025) found that early adolescents who frequently use social media report more depressive symptoms, using longitudinal data to link screen time with worsening mental health. Murarka et al. (2025) used machine learning to detect mental illness through social media behavior, confirming strong digital footprints of distress. Alambo et al. (2020) analyzed Reddit posts during COVID-19 and saw rising anxiety and substance-related discourse among young users. Thukral et al. (2020) revealed stress patterns in student social media posts, highlighting isolation and academic pressure. Minamitani (2024) emphasized growing public health concerns about social media addiction and youth well-being. While these studies affirm associations, most are observational or cross-sectional. There’s a lack of experimental research directly manipulating social media exposure to test causal effects. This study fills that gap by examining how altered usage levels affect anxiety and depression symptoms in real time among individuals aged 13–25.

Study Design

This study will use a within-subjects experimental design, manipulating social media usage duration across three time-controlled conditions: low (30 minutes/day), moderate (1.5 hours/day), and high (3+ hours/day). Each participant will experience all three usage levels over a three-week period, with one week dedicated to each level, counterbalanced to control for order effects. Mental health assessments will be conducted at the end of each week. This design is chosen to reduce inter-individual variability and increase statistical power. By comparing each participant’s mental health across usage conditions, we can assess whether changes in social media exposure lead to measurable psychological effects. The design is feasible at CUA using existing university resources, such as campus Wi-Fi access controls and digital logging tools, and avoids reliance on self-selection or pre-existing usage habits, ensuring more robust conclusions about causality between media exposure and mental health.

Participants

The study will recruit 30 participants aged 13 to 25, representing diverse genders, ethnicities, and academic backgrounds. Participants must not have a current diagnosed mental health disorder or be on psychiatric medication to avoid confounding effects. Recruitment will take place via flyers, school counselor referrals, and social media ads targeting the CUA community and surrounding schools. Parental consent and minor assent will be obtained for those under 18. Participants will receive a small gift card as an incentive. Ethical approval will be sought from the CUA IRB, and all participants will be informed of their rights and the study's confidentiality protocols.

Materials and Measures

Independent Variable: Social media usage duration will be manipulated into three levels (low, moderate, high), with daily usage time controlled using app timer tools and scheduled access on devices provided by the study team.

Dependent Variables: Mental health will be assessed using:

• PHQ-9: A 9-item depression scale (Nagata et al., 2025) with scores ranging from 0 to 27.
• GAD-7: A 7-item anxiety scale with scores from 0–21, widely used in adolescent populations.

Participants will also complete a brief demographic and media use history survey at baseline. Validity and reliability for PHQ-9 and GAD-7 are well-established in this population.

Control Variables: Age, academic workload, sleep duration, and screen time for non-social apps will be tracked and considered in the analysis to isolate the effect of social media usage specifically.

Procedure

Participants will be randomly assigned to one of six condition sequences to avoid order bias in this within-subjects design. Each week, they will follow the designated social media usage limit, enforced via app monitoring (e.g., Screen Time for iOS). At the end of each week, they will complete PHQ-9 and GAD-7. Week 1 may be low-use, Week 2 moderate, and Week 3 high-use or another order, depending on random assignment. Research assistants will check compliance daily and troubleshoot access issues. At the study’s end, participants will be debriefed and offered resources, including mental health counseling referrals. Data will be coded anonymously. The total study period is four weeks, including onboarding and exit interviews. The procedure is realistic in scope and can be implemented within CUA using lab scheduling, secure survey platforms, and minimal tech infrastructure.

Analysis and Expected Results

A repeated-measures ANOVA will be used to compare PHQ-9 and GAD-7 scores across the three usage levels. It is expected that both anxiety and depression scores will increase as social media usage increases. We anticipate significant differences between high- and low-use conditions, supporting the hypothesis.

Discussion

If results support the hypothesis, this study can guide schools, counselors, and families in recommending safe screen-time limits. It could also influence app developers to create healthier engagement strategies. Limitations include reliance on self-report and the short intervention period, which may not reflect chronic exposure or broader life stressors.