Kang Muslim
As policymakers worldwide grapple with the pressing challenge of adolescent digital safety—ranging from legislative social media bans to age-gating mandates—a new study published in JAMA Pediatrics offers a nuanced look at the reality of youth internet use. The research reveals that while negative encounters are a pervasive part of modern teenage life, the specific types of harm young people face vary significantly depending on the digital platforms they frequent. Rather than viewing the internet as a monolithic threat, the findings suggest that the architectural design, user demographics, and content moderation strategies of individual apps play a vital role in shaping adolescent exposure to online hazards.
The nationwide study, spearheaded by an interdisciplinary team of researchers from Stanford Medicine, the University of Wisconsin-Madison, the University of Southern California, and New York University, sought to map out the frequency and distribution of these negative experiences. By evaluating user reports across 20 distinct communication platforms, the research provides a granular look at the digital landscape that parents, educators, and legislators are striving to regulate.
Background and Context of Youth Digital Safety
The debate over the psychological well-being of adolescents in the digital age has intensified dramatically over the past decade. Concerns regarding cyberbullying, exposure to explicit materials, and the addictive design features of modern feeds have sparked international policy debates. Previous academic inquiries have frequently documented specific facets of these online risks. For instance, a 2022 survey highlighted the alarming prevalence of online sexual abuse and grooming among minors, while a subsequent 2026 study revealed that nearly half of adolescents who share sexually explicit imagery experience nonconsensual distribution or extortion.
Despite the growing body of literature, existing research has largely suffered from a lack of granularity. Most studies tend to examine isolated categories of negative behavior, such as general harassment or body dissatisfaction, without tracking how these issues manifest across specific applications. Furthermore, little effort has been made to compare high-frequency platforms—such as short-form video apps like TikTok and image-centric networks like Instagram—against traditional messaging applications or long-form video platforms like YouTube.
To bridge this data gap, the research team, led by Angela Y. Lee—then a postdoctoral fellow at Stanford Medicine’s Center for Digital Health and now an assistant professor at the University of Wisconsin-Madison—set out to establish a clear, consistent baseline for measuring youth online experiences. The initiative was born out of a shared cross-disciplinary goal to move beyond broad generalizations and provide actionable metrics for policymakers and technologists alike.
Chronology and Methodology of the Study
The empirical investigation was conducted between February and December 2025. The researchers deployed a nationally representative, probability-based survey targeting 472 adolescents residing in the United States, aged 13 to 17. By utilizing a probability-based sample, the team ensured that the demographic representation closely mirrored the broader adolescent population of the country.
Participants were asked to complete detailed questionnaires tracking their interactions and exposures across 20 popular online communication and social media platforms over a preceding four-week window. The surveyed platforms included widely used networks such as Facebook, Instagram, Snapchat, TikTok, and YouTube.
During the assessment, teenagers reported whether they had encountered specific adverse events either offline or on specific digital platforms. The evaluated indicators included direct harassment, witnessing the harassment of others, receiving unwanted sexual imagery, exposure to violent content, viewing self-harm encouragement, and suffering sleep disruption due to device usage. To maximize accuracy, the research team applied statistical weighting techniques to the collected data, adjusting for potential demographic skews and ensuring alignment with national benchmarks.
Detailed Findings: Online Versus Offline Realities
The resulting data underscores a sobering reality: negative encounters are remarkably common among contemporary teenagers. Across both physical and digital environments, nearly half of the surveyed adolescents (44.7%) witnessed another individual being harassed within the four-week observation period, while 29.8% reported experiencing direct harassment themselves.
However, a comparative analysis of offline versus online environments reveals a striking divergence in the nature of these harms. Direct harassment occurred at surprisingly comparable rates in both spheres, with 18.5% of teens reporting offline interpersonal conflict and 16.4% encountering direct harassment online. This finding suggests that bullying and interpersonal hostility remain fundamental social challenges that transcend technological mediums, occurring wherever teenagers congregate.
Conversely, digital platforms emerged as exponentially more common vectors for exposure to graphic, explicit, or psychologically distressing content. The disparities are stark:
- Violent Imagery: Adolescents reported encountering violent visual content online at a rate of 10.7%, compared to just 1.6% in offline settings.
- Unwanted Sexual Content: Unwanted sexual images were viewed by 13.4% of respondents online, contrasted with a mere 2.0% reporting offline exposure.
- Self-Harm Material: Content encouraging self-injury was encountered by 9.7% of teenagers online, while 5.2% reported offline exposure.
- Sleep Disruption: Over half of all participants (51.5%) indicated that their technology use directly interfered with their sleep patterns.
Platform-Specific Variations and Disparities
Perhaps the most critical contribution of the study is its demonstration that online harms are not distributed evenly across the digital ecosystem. Instead, specific platforms are disproportionately associated with distinct types of negative experiences.
For example, exposure to violent imagery varied dramatically depending on the app. While 9.1% of Instagram users reported seeing graphic violence, only 1.9% of YouTube users noted similar encounters.
Psychological impacts tied to social comparison also showed clear platform-specific clustering. Feelings of inadequacy—the sense that one’s life is inferior to the lives portrayed in other people’s posts—were most frequently reported on Instagram (16.6%) and TikTok (13.4%). In contrast, such feelings were significantly lower among users of text messaging (5.6%) and YouTube (4.1%).
Similarly, technological interference with circadian rhythms was concentrated primarily within short-form video environments. A substantial 42.2% of TikTok users and 27.8% of Instagram users attributed their sleep disruption directly to those applications. Meanwhile, only 10.2% of respondents pointed to text messaging as the source of their sleep deprivation.
Interestingly, the study found no statistically significant correlation between the incidence of online harms and traditional socioeconomic variables, such as household income, parental education levels, or the implementation of parental control software on smartphones. This suggests that technical restrictions or socioeconomic status alone do not insulate teenagers from digital risks.
Methodological Limitations and Analytical Scope
While the findings offer valuable granular insights, the authors emphasize several important methodological limitations that should temper broad interpretations.
First, the investigation utilized a relatively modest sample size of 472 adolescents. As a pilot study designed to demonstrate a novel measurement framework, the authors caution against viewing the percentages as absolute, definitive metrics for every individual platform. Smaller sample sizes inherently limit precision when evaluating rare events or less mainstream applications.
Second, the data relies entirely on retrospective self-reporting over a four-week recall period. This methodology is inherently vulnerable to recall bias, wherein participants may vividly remember highly emotionally charged incidents while failing to register milder, routine daily interactions. Furthermore, stigmatized occurrences—such as sending or receiving sexually explicit materials—may suffer from underreporting due to social stigma, privacy concerns, or embarrassment.
Third, attribution bias poses a challenge in multi-app environments. Because modern teenagers frequently switch between numerous applications simultaneously, accurately pinpointing the exact platform where a specific distressing video was viewed or a hurtful message was received can be difficult. Users may reflexively attribute negative experiences to the apps where they spend the majority of their time.
Finally, the cross-sectional nature of the survey prevents researchers from drawing definitive causal conclusions. Lead author Angela Y. Lee noted that the data cannot definitively establish whether a specific platform’s design causes increased exposure to harms, or if the findings simply reflect self-selection biases—such as certain demographics being more naturally inclined to use specific apps that inherently feature higher rates of social comparison.
Broader Implications and Future Directions
The implications of the JAMA Pediatrics study extend far beyond academic circles, offering a vital empirical foundation for technology developers, advocacy groups, and lawmakers. By establishing a replicable methodology for tracking platform-specific harms, the research provides a framework that can be used to evaluate whether future regulatory interventions or product safety updates actually succeed in mitigating adolescent exposure to risk.
As the digital ecosystem continues to evolve, researchers are already looking ahead to the next frontier of youth technology consumption. Lee revealed that her team is currently preparing a larger, expanded study involving American adolescents and their parents. This upcoming phase aims to broaden the scope of platform tracking to include emerging technologies such as artificial intelligence chatbots and AI companion apps, which are rapidly integrating into the daily lives of minors.
Ultimately, the study underscores that addressing youth digital safety requires moving past broad ideological debates and embracing precise, empirical monitoring. By understanding where, when, and how specific platforms expose teenagers to harm, stakeholders can better tailor interventions designed to safeguard the psychological well-being of the next generation.
