Data Analysis Essay: Social Media Use and Sleep

Introduction

Adolescents and young adults are the highest users of social media. This population group experiences many risk factors which affect their sleep. During puberty, circadian rhythms change meaning adolescents go to bed later; adolescents start school early meaning they get little sleep during the morning and then spend time using screens late at night. This causes a harmful cycle of chronic sleep deprivation which can cause issues within the public. Lack of sleep can cause problems with their academic performance, mood regulation, memory consolidation and growth. Any factors which cause sleep disruption should be considered. An increasing amount of research shows that social media can have a negative effect on sleep. This causes less sleep, reduced sleep efficiency and poor quality sleep, and that is shown to have a negative impact on sleep, by reducing sleep time, sleep efficiency and sleep quality. The amount of social media use is increasingly associated with reduced sleep time, delayed bedtime and sleep quality. This essay compares the statistical evidence of recent studies published by MDPI, to address three questions. The first question is whether the amount of social media use is a predictor of sleep difficulties later in life. Secondly, what is the strength of the association between problematic or addictive use and sleep outcomes? Third, is there a graded dose-response relationship between the severity of dependence and poorer mental health, and is disturbed sleep a pathway to poorer mental health? The essay concludes that general use has a small association, problematic use has a strong, graded and clinically significant association with poor sleep.

The studies were included due to their recent publication, peer review, statistical transparency, and complementarity to each other. In their two-wave longitudinal survey of 2021 secondary school students in the Netherlands (M = 13.86 years, SD = 1.25), Van den Eijnden et al. (2021) assessed the students’ baseline tendencies toward social media use and problematic use as well as their sleep patterns and sleep quality with a one-year follow-up using a six-item measure of frequency of use, the Social Media Disorder scale, a one-item measure of bedtime and the Groningen Sleep Quality Scale. Çelik and Güler (2025) included the investigation of social media addiction and its possible effects on the sleep of adolescents using the Social Media Addiction Scale for Adolescents and Cleveland Adolescent Sleepiness Questionnaire. The number of participants in the study was 309 Turkish students (10-18 years of age) and Pearson’s correlation and linear regression analysis were used to identify possible correlations. Hernández-Mariano et al. (2025) investigated the social media addiction, sleep quality (Pittsburgh Sleep Quality Index; PSQI) and depressive symptoms (Center for Epidemiologic Studies Depression Scale; CESD-7) among 638 students enrolled in undergraduate nursing programs from three public universities in Mexico using logistic regression and counterfactual causal mediation analysis. The studies are joined together in two ways—first, longitudinally, studies of adolescent and young adult; second, cross-sectionally, studies involving adolescent and studies involving young adult.

Statistical Findings Across the Studies

The longitudinal Dutch data are the most convincing regarding direction of effect, and indicate that the frequency of use is most important for sleep timing. Baseline use of social media was a significant predictor of delayed bedtime one year later (β = 0.14, p < .001) and problematic use of social media predicted delayed bedtime one year later (β = 0.07, p < .01) (Van den Eijnden et al., 2021). The effects are small, but became apparent after a full year and after earlier sleep characteristics and demographic variables were accounted for, making them less likely to be a “snapshot” effect. Importantly, neither measure was a significant predictor of the perceived sleep quality in these controls. It seems, at least in the short-term, it is not how young people evaluate their sleep that suffers, but when they go to sleep. Parental rules as a protective factor were also studied in the same study. The effects of social media use on bedtime and sleep quality were buffered by strict rules regarding internet and smartphone use 1 hour before bedtime, but only for adolescents who had low levels of social media use. When users were very engaged, parental control was no longer able to ensure sleep. Rule setting is thus a beneficial early intervention, but is not a cure if heavy use has already occurred.

The results of the study conducted in Turkey showed that 37.9% of the adolescents had sleep problems and that social media addiction had a positive and significant correlation with daytime sleepiness (r = 0.61, p < .05) which is considered a strong correlation according to traditional standards. The adolescents’ sleep problems were reported as 37.9% in the Turkish study, and daytime sleepiness showed a positive and significant correlation with social media addiction (r = 0.61, p < .05) which is considered as a strong correlation according to traditional standards. The regression results indicated that social media addiction explained 37.5% of the variance in sleepiness (R² = 0.375) and each one unit increase in addiction scores resulted in an increase of 0.79 in sleepiness (B = 0.79). The descriptive statistics have confirmed the concern. The average sleepiness score was 39.86 (SD = 11.99), and the highest subscale was insomnia at school (M = 14.61, SD = 4.56). This discovery directly suggests a risk to concentration, learning and school function as the negative effects of poor sleep occur during the time when they are supposed to be learning (Çelik & Güler, 2025).

The Mexican multicenter study adds to the line of evidence by including an element of dose-response and testing the mediation, whereas Hernández-Mariano et al. (2025) identified 51.7% of nursing students as sleep-disturbed and 27.5% as depressed. Indeed, with increasing levels of dependence, the tendency for sleep disorders was also elevated, gradually rising from 36.8% among non-dependent students to 69.2% of those with high dependence. Thus, with short-term sleep loss acting as a mediator, problematic social media use was a risk factor for sleep disorders that, in turn, contributed to the development of mental health problems. But getting better sleep would take off about one-third of the risk, and the other two-thirds indicate that sleep is just part of the equation.

Discussion, Implications, and Limitations

When the three studies are compared, there are patterns, despite the differences in country, age group, and methodology: general frequency of use is associated with later bedtimes, addictive use is strongly associated with sleepiness and disturbance, and the association increases as the level of dependence increases. This finding further bolsters confidence that the connection is not a one-off finding from one study. There is one apparent contradiction that needs to be resolved, however. A significant prediction of sleep quality was not found by Van den Eijnden et al. (2021) but by Çelik and Güler (2025) for sleepiness. The results in fact complement each other as the Dutch study took into account the quality of sleep prior to the study, while the Turkish study looked at daytime sleepiness, which is a cumulative effect of both short sleep and late sleep, irrespective of how satisfied users are with their sleep. The lesson to be learned from the above is that the results of a study are determined by the outcome of sleep selected.

A plausible mechanism for such a pattern to arise exists. Screen light in the evening can inhibit melatonin production and disrupt circadian rhythms; stimulating and open-ended information can raise arousal and lead to late night bedtime procrastination. This resulting loss of sleep adds up to daytime sleepiness. The mediation analysis supports that explanation since sleep disruption interferes with the neuroendocrine regulation and emotional control that would normally help in guarding against depressive symptoms (Hernández-Mariano et al., 2025). The implications of the findings are of both prevention and intervention. Prevention might require early intervention, before the onset of the development of the heavier users, since frequency of use predicted delayed bedtimes, and parental restrictions only protected lighter users (Van den Eijnden et al., 2021). The same applies to universities, as the mediation analysis indicated that a third of the risk of problematic social media use leading to depression can be absorbed by sleep, meaning that sleep hygiene interventions in university students could help reduce this pathway (Hernández-Mariano et al., 2025). The results also help to identify some of the simple interventions that Çelik and Güler (2025) recommended to tackle daytime sleepiness, such as education of adolescents about the negative effects of excessive and uncontrolled social media usage and parental limits on social media use.

The evidence, however, has its limitations which need to be discussed. Two of the three studies used were of the cross-sectional type and therefore, there is limited certainty of cause and effect. In addition, the longitudinal study had a loss of approximately 30% of its sample and those who dropped out were heavier users and were likely poorer sleepers as well, which may have introduced a bias. Lastly, all three studies have been self-report and are specific to their cultural and educational settings. More research is needed to further explore the boundaries of the association and to test the causal relationship using objective measures and longer follow-up periods and experimental restrictions.

Conclusion

The relationship between social media use and sleep is consistently reported, but differs depending on how social media use is measured. Excessive use of social media is associated with later bedtimes, excessive social media use is strongly related to daytime sleepiness, and the higher the level of dependence, the greater the risk of sleep disturbance and depression in a graded fashion, with sleep disturbance accounting for approximately one-third of the link between problematic use and depressive symptoms. The evidence has relevance for sleep and digital-wellness interventions at the university level, parental guidance for younger teens, and prevention. Sleep protection does not mean you have to ban social media, which has its merits, but rather it means you have to understand the numbers and convert them into behavioral change; and small changes to nighttime routines can add up to hours of sleep and better mood and academic performance.