Time Spent on Social Media & Adolescent Mental Health Problems: A Meta-Analysis

Ferguson, C. J., Kaye, L. K., Branley-Bell, D., & Markey, P. (2024). There is no evidence that time spent on social media is correlated with adolescent mental health problems: Findings from a meta-analysis. Professional Psychology: Research and Practice.  https://doi.org/10.1037/pro0000589

Rationale

The study was undertaken to evaluate the current evidence base regarding the relationship between social media use and mental health outcomes in adolescents.

This topic has received significant attention from researchers, policymakers, and the public due to concerns about the potential negative impacts of social media on youth wellbeing (Twenge & Campbell, 2019).

However, previous research has produced mixed findings, with some studies suggesting harmful effects (e.g., Twenge et al., 2018) while others argue for more cautious interpretations of the data (e.g., Orben et al., 2019).

The authors note that the existing literature has several methodological limitations, including a focus on time spent on social media rather than specific behaviors, reliance on self-report measures, failure to distinguish between platforms, and inconsistent use of best research practices.

These issues have made it difficult to draw firm conclusions about the relationship between social media use and adolescent mental health.

Given the importance of this topic for informing policy and practice, the authors sought to conduct a comprehensive meta-analysis to:

1. Establish the overall effect size between social media use and mental health variables in adolescents
2. Evaluate the prevalence and impact of best research practices in the existing literature

This meta-analysis builds on previous work by Ferguson et al. (2022) examining screen time more broadly, by focusing specifically on social media use among adolescents aged 18 and younger.

The rationale for this focused approach was to address critiques that previous analyses were less focused on youth and social media use specifically.

Method

The authors conducted a meta-analysis following PRISMA guidelines. They searched the APA PsycInfo and Medline databases in September 2022.

The search terms used were:

(“Social Media” or “Facebook” or “Instagram” or “Twitter” or “snapchat” or “social networking” or “TikTok”) and (“depression” or “anxiety” or “loneliness” or “suicide” or “mental health” or “mental well” or “mental illness” or “mental well-being” or “psychological well-being”) and (“youth” or “teen” or “adoles*”)

The search was limited to studies from the past 10 years.

Studies were included if they:

• Included a measure of social media use or experimental comparison of social media with a control condition
• Had a sample only including participants between ages 12-18 (with a few exceptions slightly younger)
• Included sufficient information to calculate an effect size “r”

Studies were excluded if they:

• Focused on motivations for using social media or purposes of use rather than time spent
• Measured “problematic social media use” rather than general use

Statistical measures:

The authors used Jamovi to calculate a random-effects mean effect size, as well as measures of publication bias including funnel plot analysis, Egger’s regression, trim and fill, p-curve, and p-uniform. They used a restricted maximum-likelihood model with Fisher’s r-to-z transformation.

Effect sizes were calculated as standardized regression coefficients (betas) based on the most conservative value available in each study. An effect size of r = .10 was considered the minimum threshold for practical significance.

The authors also conducted moderator analyses examining factors such as biological sex, study type, data set type, best research practices, and citation bias.

Results

Overall effect size:

The mean effect of social media on mental health across all studies was β = .061, which is below the threshold for practical significance (r = .10).

This suggests the observed effects are indistinguishable from statistical noise.

Sex differences:

There were no statistically significant differences in effect sizes between boys (β = .044) and girls (β = .075). Both effects were below the threshold for practical significance.

Other moderator analyses:

• Best research practices were not a significant moderator of effect size
• Study year was not a significant moderator
• Age of participants was not a significant moderator (p = .053)
• Citation bias was not a significant moderator (p = .319)
• Study type (correlational vs. longitudinal) was not significant (p = .067)
• Use of self-report vs. time diaries was not a significant moderator (p = .430)
• Type of data set was a significant moderator (p = .043), with national data sets showing slightly larger effects (β = .067) compared to bespoke data sets (β = .041) and dissertations (β = .045)

Best research practices:

• Use of standardized and well-validated mental health measures was common (95% and 92% of studies respectively)
• Use of basic controls for gender, age, family environment, and Time 1 outcomes in longitudinal studies was moderately common (64%)
• Use of multiple respondents (19%), distractor questions (0% reported), preregistration (5%), and hypothesis guessing queries (1%) were rare

Publication bias:

Evidence generally suggested an absence of publication bias, though Egger’s regression (p = .021) and trim and fill (four missing studies) indicated some potential for bias.

Insight

The key finding of this meta-analysis is that there is no evidence for a clinically relevant link between time spent on social media and mental health issues in adolescents.

This conclusion challenges widespread claims about the potential harms of social media use on youth mental health.

The study is particularly informative because it synthesizes a large body of research (46 studies) and examines potential moderating factors.

The finding that effect sizes were consistently small, even when examining different subgroups (e.g., boys vs. girls, correlational vs. longitudinal studies), strengthens the overall conclusion.

These findings extend previous research by focusing specifically on social media use among adolescents, rather than screen time more broadly.

They also highlight methodological issues in the existing literature, such as the rarity of preregistration and controls for demand characteristics.

The authors argue that the disconnect between their findings and public discourse about social media harms reflects a pattern of moral panic surrounding new technologies.

This insight suggests a need for more nuanced and evidence-based approaches to understanding and addressing potential impacts of social media on youth.

For future research, the authors recommend:

1. Focusing on specific behaviors and content on social media rather than just time spent
2. Improving methodological rigor through practices like preregistration and controlling for demand characteristics
3. Examining potential positive outcomes of social media use, not just negative ones
4. Investigating how individual differences may influence the impacts of social media use

Strengths

The study had many methodological strengths including:

1. Comprehensive search strategy covering multiple databases
2. Clear inclusion and exclusion criteria
3. Use of standardized effect sizes (betas) to allow comparison across studies
4. Examination of multiple potential moderating factors
5. Assessment of publication bias using multiple methods
6. Evaluation of best research practices in included studies
7. Preregistration of the meta-analysis plan
8. Transparent reporting of methods and results, with data made publicly available

Limitations

1. The meta-analysis was limited to studies published in the past 10 years, potentially missing earlier research.
2. Only two databases (APA PsycInfo and Medline) were searched, which may have missed relevant studies from other fields.
3. The analysis relied heavily on self-report measures of social media use and mental health outcomes, which have known limitations.
4. Most included studies used cross-sectional designs, limiting causal inferences.
5. The focus on time spent on social media may not capture more nuanced aspects of use that could impact mental health.
6. The age range was restricted to adolescents, so findings may not generalize to younger children or adults.
7. There was limited variance in best research practices scores, which may have affected the ability to detect moderating effects.
8. The study did not distinguish between different social media platforms, which may have different impacts.

These limitations suggest caution in generalizing the findings and highlight the need for more rigorous and nuanced research in this area.

Implications

The results of this meta-analysis have significant implications for public discourse, policy, and clinical practice related to adolescent social media use and mental health:

1. Public messaging: The findings challenge widespread claims about social media harms to youth mental health. Professional organizations and policymakers may need to reassess and potentially moderate their public statements on this issue.
2. Policy approaches: The lack of evidence for clinically relevant effects suggests that policy approaches focused on restricting social media access by age are not well-supported. Instead, policies promoting digital literacy and education about healthy social media use may be more appropriate.
3. Clinical practice: Mental health professionals working with adolescents should be cautious about attributing mental health issues primarily to social media use. A more holistic assessment of potential contributing factors is warranted.
4. Research priorities: The study highlights the need for more methodologically rigorous research examining specific aspects of social media use, rather than just time spent. This could inform more targeted interventions if needed.
5. Parental guidance: Parents may need to reevaluate concerns about social media use, focusing more on helping adolescents develop healthy online behaviors rather than strictly limiting access.
6. Media literacy: The study underscores the importance of critical thinking about media effects and the potential for moral panics around new technologies. This has implications for education and public understanding of science.
7. Individual differences: The lack of strong overall effects suggests a need to consider how social media impacts may vary based on individual factors, informing more personalized approaches to addressing potential issues.

These implications highlight the need for a more nuanced and evidence-based approach to understanding and addressing the role of social media in adolescent mental health.

References

Primary reference

Ferguson, C. J., Kaye, L. K., Branley-Bell, D., & Markey, P. (2024). There is no evidence that time spent on social media is correlated with adolescent mental health problems: Findings from a meta-analysis. Professional Psychology: Research and Practice.  https://doi.org/10.1037/pro0000589

Other references

Ferguson, C. J., Kaye, L. K., Branley-Bell, D., Markey, P., Ivory, J. D., Klisinan, D., Elson, M., Smyth, M., Hogg, J. L., McDonnell, D., Nichols, D., Siddiqui, S., Gregerson, M., & Wilson, J. (2022). Like this meta-analysis: Screen media and mental health. Professional Psychology: Research and Practice, 53(2), 205–214.

Orben, A., Dienlin, T., & Przybylski, A. K. (2019). Social media’s enduring effect on adolescent life satisfaction. Proceedings of the National Academy of Sciences of the United States of America, 116(21), 10226–10228.

Twenge, J. M., & Campbell, W. K. (2019). Media use is linked to lower psychological well-being: Evidence from three datasets. Psychiatric Quarterly, 90(2), 311–331.

Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2018). Increases in depressive symptoms, suicide-related outcomes, and suicide rates among U.S. adolescents after 2010 and links to increased new media screen time. Clinical Psychological Science, 6(1), 3–17.

Keep Learning

Here are some Socratic questions for a college class to discuss this paper:

1. How might societal expectations and media narratives influence research questions and interpretations in the field of social media and mental health?
2. What are the potential consequences of overestimating or underestimating the effects of social media on adolescent mental health?
3. How can researchers balance the need for standardized measures with capturing the nuanced and evolving nature of social media use?
4. What ethical considerations arise when studying adolescent social media use and mental health?
5. How might cultural differences impact the relationship between social media use and mental health outcomes? How could future research address this?
6. How can policymakers and mental health professionals responsibly communicate about social media risks and benefits given the current state of evidence?
7. What alternative research approaches or methodologies could provide more insight into the complex relationship between social media use and adolescent well-being?