CoChat
Sign Up
Cited by 56Impact of Technology on AdolescentsBullying, Victimization, and AggressionChild Development and Digital Technology

Social Media Addiction and Aggressive Behaviors in Early Adolescents: The Mediating Role of Nighttime Social Media Use and Sleep Quality

Open access
Lin, Shanyan, Longobardi, Claudio, Gastaldi, Francesca Giovanna Maria, Fabris, Matteo Angelo·2024·The Journal of Early Adolescence
DOI 10.1177/02724316231160142
Open this paper in CoChat — free accountOpen PDF

Abstract

Social media addiction (SMA) is considered a risk factor for aggressive behaviors in adolescents, while its underlying mechanisms remains unclear. In this study, we investigated the possible mediating roles of nighttime social media use and sleep quality in the relationship between SMA and aggressive behaviors. A sample of 773 Italian secondary school students (49.9% female, age range = 11–15 years) completed a self-report questionnaire that included information on SMA, nighttime social media use, sleep quality, and aggressive behaviors. Results indicated that SMA was positively associated with aggressive behaviors. In addition, SMA was positively related to nighttime social media use and negatively related to sleep quality. Nighttime social media use was associated with a poorer sleep quality, which in turn associated with more aggressive behaviors. Analyses indicated that both nighttime social media use and sleep quality were mediators in the relationship between SMA and aggressive behaviors in adolescents.

AI snapshot

Written from the paper’s full text.

TL;DR

In a cross-sectional sample of 773 Italian early adolescents, social media addiction (SMA) was positively associated with aggressive behaviors, and this link was mediated by greater nighttime social media use and poorer sleep quality. The authors conclude that nighttime social media use and sleep quality may be mechanisms linking SMA to aggression in early adolescence.

Key findings

  • SMA was positively correlated with aggressive behaviors (\(r = .41, p < .001\)), positively correlated with nighttime social media use (\(r = .57, p < .001\)), and negatively correlated with sleep quality (\(r = -.55, p < .001\)).
  • Nighttime social media use was negatively associated with sleep quality (\(r = -.54, p < .001\)) and positively associated with aggressive behaviors (\(r = .38, p < .001\)); sleep quality was negatively associated with aggressive behaviors (\(r = -.36, p < .001\)).
  • All hypothesized indirect effects were significant: SMA → nighttime social media use → aggressive behaviors (estimated effect = .0778, 95% CI [.0318, .1218]); SMA → sleep quality → aggressive behaviors (estimated effect = .0519, 95% CI [.0216, .0859]); and SMA → nighttime social media use → sleep quality → aggressive behaviors (estimated effect = .0280, 95% CI [.0115, .0472]).
  • The direct effect of SMA on aggressive behaviors remained significant after controlling for mediators (\(\beta = .25, p < .001\)). Using a cutoff of 19 on the BSMAS, 10.48% of participants were classified as addicted to social media.

Methods

  • Cross-sectional self-report study of 773 Italian secondary school students (49.9% female; age range 11–15 years; mean age = 12.34, SD = .96) from grades 6–8, recruited by convenience sampling in northern Italy.
  • Measures included the Bergen Social Media Addiction Scale, Nighttime Specific Social Media Use Scale, Adolescent Sleep-Wake Scale, and a 36-item aggression measure covering overt and relational aggression.
  • Mediation was tested with PROCESS macro Model 6 in SPSS 26, using bootstrapping with 5,000 samples to estimate indirect effects and 95% confidence intervals.

Limitations

  • Stated by the authors: the convenience sample from northern Italy limits generalizability to other Italian regions or cultures; sleep quality was self-reported for the past month and may differ from objective sleep measures; the cross-sectional design cannot establish causal or temporal order; all measures were self-report and subject to social desirability, memory, and comprehension effects; and the study examined general aggressive behaviors rather than specific platforms or aggression subtypes.
  • Evident from the text: the mediation model is based on cross-sectional associations, so the reported indirect effects should not be interpreted as causal pathways.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1177/02724316231160142

Cites

56 of 56 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 30 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Social media addiction is negatively associated with sleep quality in this sample.

    “SMA was positively related to nighttime social media use and negatively related to sleep quality.”
    • Also stated in: § Correlations and Descriptive Statistics · plain
      “SMA was positively correlated with nighttime social media use (r = .57,p < .001)and aggressive behaviors (r = 0.41, p < .001)and negatively correlated with sleep quality (r = À.55, p < .001).”
    • Also stated in: § Testing the Mediated Associations · plain
      “sleep quality was negatively predicted by SMA (β = À.35, p < .001)”
    Quote in the paperSubstantial claim
  • Findingown data

    Nighttime social media use is positively associated with aggressive behaviors.

    “Nighttime social media use was negatively associated with sleep quality (r = À.54, p < .001)and positively associated with aggressive behaviors (r = .38,p < .001).”
    • Also stated in: § Testing the Mediated Associations · plain
      “which in turn was positively correlated with aggressive behaviors (β = .14,p < .001).”
    Quote in the paperSubstantial claim
  • Findingown data

    Nighttime social media use and sleep quality serially mediate the link between social media addiction and aggressive behaviors.

    “This suggests that the chain mediation effect of nighttime social media use and sleep quality in the relationship between SMA and aggressive behaviors was also significant (estimated effect = .028,95% CI = [.012,.047]).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The causal ordering in the mediation model holds, with social media addiction preceding nighttime social media use, sleep quality, and aggressive behaviors.
  • Findingown data

    Nighttime social media use mediates the link between social media addiction and aggressive behaviors.

    “Analyses indicated that both nighttime social media use and sleep quality were mediators in the relationship between SMA and aggressive behaviors in adolescents.”
    • Also stated in: § Testing the Mediated Associations · hedged
      “The indirect effect of nighttime social media use between SMA and aggressive behaviors was significant (estimated effect = .078,95% CI = [.032,.122]),suggesting that nighttime social media use was a significant mediator.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The causal ordering in the mediation model holds, with social media addiction preceding nighttime social media use, sleep quality, and aggressive behaviors.
  • Findingown data

    In early adolescents, social media addiction is positively associated with aggressive behaviors.

    “Results indicated that SMA was positively associated with aggressive behaviors.”
    • Also stated in: § Correlations and Descriptive Statistics · plain
      “SMA was positively correlated with nighttime social media use (r = .57,p < .001)and aggressive behaviors (r = 0.41, p < .001)”
    • Also stated in: § Discussion · plain
      “Our results showed that SMA was positively associated with aggressive behaviors in adolescents.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 8
  • May combine two claims. 5

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Monte Carlo power analysis indicated that with 773 participants, power for indirect effects ranged from .91 to .99.

    “Monte Carlo power analysis for indirect effects was performed, and the result showed that with a sample size of 773, the power ranged from .91 to .99.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A longitudinal study found that media use, especially Internet and online games, could undermine the development of empathic and prosocial behaviors, probably due to reduced face-to-face interactions.

    “a longitudinal study (Fitzpatrick & Boers, 2021) found that media use, especially the exposure to the Internet and online games, could undermine the development of empathic and prosocial behaviors, probably due to reduced face-to-face interactions.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Fitzpatrick & Boers, 2021

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Social media interactions tend to increase arousal in adolescents, affecting sleep behavior and potentially decreasing sleep quality.

    “In addition, interactions on social media tend to increase arousal in adolescents affecting various aspects of sleep behavior and may thus decrease sleep quality (Lin et al., 2021).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lin et al., 2021

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The aggressive behaviors instrument showed good internal consistency in this sample (Cronbach's alpha = .91).

    “In the present study, this instrument showed good internal consistency: Cronbach's alpha = .91.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Objective and subjective sleep measures may reflect different characteristics of sleep, and these differences may impact the final model.

    “On one hand, objective and subjective sleep measures may reflect different chiaracteristics of sleep and different aspects of the sleep experience (Gregory & Sadeh, 2012) and that these differences may impact the final model of the variables studied in this research.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Gregory & Sadeh, 2012

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    These difficulties are particularly pronounced for social media addiction at-risk adolescents, who have higher need to belong and greater fear of missing out.

    “These difficulties might be particularly pronounced for SMA at-risk adolescents, considering that a higher need to belong and a greater fear of missing out appear to characterize these adolescents (Fabris et al., 2021;Marengo et al., 2021).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Fabris et al., 2021

      Not checked in the free run.

    • cites Marengo et al., 2021

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    Self-report sleep quality for the past month may only reflect general or mid/long-term relationships among variables.

    “On the other hand, measuring the subjective-recall general sleep quality in the past month may only reflect the general or mid/long-term relationships of the studied variables.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Adolescents at risk for social media addiction may use more social networks at night, which results in lower sleep quality.

    “Therefore, the SMA at-risk adolescents may use more social networks during the night, resulting in lower sleep quality.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationown data

    The current sample's mean values on the main variables were similar to those in previous studies, suggesting the sample was representative.

    “The mean score on SMA in the current sample is 12.68 (SD = 4.89), which is similar to the value found in other studies with the Italian youth sample (e.g., M = 12.11, SD = 4.94, Monacis et al., 2017).”
    • Also stated in: § Correlations and Descriptive Statistics · hedged
      “In addition, the mean values of other variables examined were similar to those found in previous studies (e.g., LeBourgeois et al., 2005;Scott & Woods, 2018), suggesting that this was a representative sample.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Monacis et al., 2017

      Not checked in the free run.

    • cites LeBourgeois et al., 2005

      Not checked in the free run.

    • cites Scott & Woods, 2018

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Including gender and age as covariates did not significantly change the mediation model.

    “Student's gender and age were included as controlled variables but they not illustrated in the figure to make the picture concise, and there is no significant difference between the models with or without these controlled variables.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The use of only self-report instruments is a limitation due to social desirability, memory, and text comprehension.

    “Fourth, only selfreport instruments were used in the study, which is a limitation given the effects of social desirability, memory, and text comprehension.”
    Quote in the paperNot a substantial claim
11 of 11

  • The paper relies on this without support.

    Little research has examined the underlying mechanisms, such as mediating factors, of the relationship between social media addiction and aggressive behaviors.

    “Although a growing body of work has addressed the relationship between SMA (and other Internet-related addictions) and aggressive behaviors in adolescents, little has been studied in terms of its underlying mechanisms (e.g., the mediating factors).”

  • The paper relies on this without support.

    Few studies have specifically examined social media use, as opposed to other Internet activities, in relation to sleep quality.

    “while fewer studies have specifically examined social media use (Woods & Scott, 2016).This is a limitation considering that social media use is one of the main online activities among adolescents”

  • The paper relies on this without support.

    Very little is known about the possible relationship between social media addiction and nighttime social media use and their impacts on sleep quality.

    “however, we have very little information about the possible relationship between SMA and nighttime social media use and their potential impacts on sleep quality.”

  • The paper relies on this without support.

    No study has specifically examined social media addiction and its relationship with adolescents' nighttime social media use and sleep quality.

    “However, no study has specifically examined SMA and its relationship with adolescents' nighttime social media use and sleep quality.”

  • The paper relies on this without support.

    Little research has been conducted on possible mediating factors in the association between aggression and social media addiction or other Internet-related addictions.

    “Although there is evidence of an association between aggression and SMA or other Internetrelated addictions, little research has been conducted on the possible mediating factors involved.”

  • The paper relies on this without support.

    The causal ordering in the mediation model holds, with social media addiction preceding nighttime social media use, sleep quality, and aggressive behaviors.

    “we hypothesized that SMA directly predicted more aggressive behaviors in adolescents, and that SMA could positively and indirectly influence aggressive behaviors through more nighttime social media use and poor sleep quality.”

  • The paper relies on this without support.

    Adolescents at risk for social media addiction use the Internet and social media more than other adolescents.

    “SMA adolescents tend to use the Internet and social media more than others, and it is possible that this behavior increases during free time, such as at night before bedtime.”
7 of 7

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Cybervictimization remains a class phenomenon and is often the online extension of a pre-existing offline form of victimization.

    “Although cybervictimization takes place online, it remains a class phenomenon, which is often the online extension of a pre-existing offline form of victimization [45,50,51].”

    From Cybervictimization and Online Sexual Harassment: Prevalence, Association, and Predictors (2024) · cites this paper as [50]

    The section “Social Media Addiction and Aggressive Behaviors in Adolescents” says social media use may increase the risk of victimization, but does not say cybervictimization is a class phenomenon or an extension of prior offline victimization. Searches for those terms and for offline bullying among classmates found no passage in our copy establishing the claimed relationship. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Mean gender code conflicts with participant counts

    Minor

    Gender is coded 0 for female and 1 for male. With 386 females among 773 participants, the mean is 387/773 = 0.5006, which rounds to .50, not .51.

  • Gender standard deviation is impossible for its coding

    Minor

    A variable coded only 0 or 1 cannot have a standard deviation of .52; even its maximum possible sample standard deviation with 773 observations is approximately .5003.

  • Age standard deviation differs between text and table

    Minor

    The participants' age standard deviation is reported as .96 in the text but .95 in Table 1, both to two decimal places.

3 of 3

Cites

Flagged first, then by citations

Research works printed but not linked: 4

  1. Chu X., 2021, Current Psychology, 1
  2. Estévez López E., 2018, Psicothema, 30, 66
  3. Griffiths M. D., 2017, Education and Health, 35, 49
  4. Mérelle S., 2017, Clinical Neuropsychiatry: Journal of Treatments Evaluation, 14, 11

Cited by

Sorted by citations

Related

Authors
Lin, S., Longobardi, C., Gastaldi, F. G. M., & Fabris, M. A.
Affiliations
University of Turin (IT)
Corresponding author
Longobardi, Claudio
ORCID
Venue
The Journal of Early Adolescence
Publisher
SAGE Publications
Published
2023-02-27
Volume
44
Issue
1
Pages
41-58
ISSN
  • 0272-4316 (ISSN-L)
  • 1552-5449
DOI
10.1177/02724316231160142
OpenAlex
W4322495768
Semantic Scholar
794e47df6075f802bad778d98687b8ee54f7ba52
Cited by
41
References
53
Citations by year
2026: 7 · 2025: 12 · 2024: 19 · 2023: 3
FWCI
17.69
Type
journal article
Language
en
Open access
Licence
journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
Locations
  • The Journal of Early Adolescence (publishedVersion)
  • Institutional Research Information System University of Turin (University of Turin) (submittedVersion, other-oa)
Topics
  • Social Sciences › Sociology and Political Science › Impact of Technology on Adolescents
  • Social Sciences › Psychology › Social Psychology › Bullying, Victimization, and Aggression
  • Social Sciences › Education › Child Development and Digital Technology
Keywords
social media addiction, aggression, sleep quality, early adolescents, mediation, high school students, adolescent mental health

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-08.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-08.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review