Abstract
Purpose:
A growing concern over social media addiction (SMA) and its impact on academic performance has highlighted a need to understand the underlying mechanisms. While a negative correlation is established, the pathways of influence remain unclear. This study examines the mediating roles of sleep quality and fatigue in the relationship between SMA and academic performance among university students in Somalia.
Methods:
A cross-sectional survey was administered to 566 undergraduate students. The BSMAS, UWES, PSQI, and FAS scales were used to measure SMA, academic performance, sleep quality, and fatigue, respectively. Data were analyzed using Partial Least Square Structural Equation Modeling (PLS-SEM) to test a model of direct and indirect effects.
Results:
The findings revealed a significant direct negative relationship between SMA and academic performance. Sleep quality emerged as a significant mediator; SMA was linked to poorer sleep, which in turn predicted lower academic performance. In contrast, while SMA was associated with higher fatigue, fatigue was not a significant mediator in the relationship between SMA and academic performance.
Conclusions:
SMA appears to harm academic performance through two distinct routes: a direct pathway linked to motivational deficits and self-regulation failure, and an indirect pathway via energy depletion from poor sleep. Fatigue is a consequence but not a causal mechanism for academic decline in our model. Educational institutions should implement interventions targeting both sleep quality and digital self-regulation to mitigate the academic consequences of excessive social media use. Future studies employing longitudinal designs could further clarify the causal dynamics of these relationships.
1 Introduction
The Internet has become vital to millions of people, whether in communication, education, or entertainment. In early 2025, over 5.56 billion individuals, or around 67.9% of the worldwide population, were actively using the Internet, which shows its prevalence and influence (). Social media is almost present everywhere (), Millions of individuals use it worldwide. The adoption of online platforms is increasing with the growth of the global population using the Internet, which shows that digital platforms and the Internet are conjoined. Social networking sites have recently been used for business, dating, politics, education, and communication (). Social media platforms have surprisingly increased along with internet accessibility, simplifying how people communicate, do business, and engage in societal activities. These platforms have shaped public discourse and education environments (). As the internet expands, social networking sites' influence is expected to grow, further redefining how people engage with the world around them. These tools increase communal interactions and people's capital ().
Despite its numerous benefits () to society, and extreme dependency on society, excessive social media usage has been found to cause health issues () among young individuals with frequent online presence, which leads to internet addiction (). Sometimes, social site addiction is defined as an uncontrollable desire or excitement about it (). Excessive social media usage may impact social interactions, educational or professional obligations, and personal Wellbeing (). SMA may potentially cause unhealthy symptoms (; ) depending on its location, its psychotherapy becomes problematic when it coincides with other Clinical and pathological illnesses, i.e., when it has both Brain-degenerative and systemic aspects. Mental dependency on social media is linked to behavioral addiction symptoms (). These uncontrollable negative effects are mainly due to individual's unnecessary use of digital platforms. It is regarded as one of the elements of online addiction (; ; ).
Digital platforms contain the following functions: self-identity, communication, media sharing, online visibility, fostering relations, trust and reputation control, and social grouping. They all enable users to identify a certain digital platform factor and, based on groundbreaking characteristics, bring additional psychological incentives to Internet users (). Digital platform offers multilevel and multidimensional approaches to any organization or social community. They can be described as social networking platforms or mobile ecosystems that can be accessed and used without restrictions (; ). Social networks are essential to learning environments as a main method for communication and community support (). Some SNSs, like Edmodo, are explicitly directed at learning ().
As highlighted above, digital networks have various advantages when applied to learning since they offer easy access to information and break down barriers in communication and telecommunication (), enhance learning through applying group work (), assist learners in improving self-study, boost students' engagement and motivation with their fellow students and teachers, and foster learning that is active and social (). Several studies conducted in Somalia examined how digital addiction disturbs educational activities, which may burden the focus from learning outcomes (). The direct association between SMA and student performance has become the focus of existing research, but it abandons other variables essential to understanding this relationship.
In that regard, this research investigates the mediation role of both Sleep quality and fatigue on the relationship between social media addiction and academic performance. The findings will shed light on the increasing prevalence of social media addiction, a phenomenon believed to hinder students' academic progress and help educational institutions and policymakers take preventive measures against it.
2 Literature review and hypothesis development
2.1 Social media addiction and academic performance
Educational engagement is the students' energy, commitment, interest, and passion for academics. Academic vigor relates to students' passion, persistence, and commitment to perform their best in class, especially when encountering challenges. Commitment implies responsibility, motivation, passion, and readiness for learning, which are characterized by overcoming academic barriers ().
Several studies indicate that students who use the internet more frequently while studying often achieve lower grades. Studies suggest that using these platforms during studying negatively affects academic performance compared to those who refrain from using them. Additionally, different distraction tasks during academic activities reduce focus and cause a decline in educational outcomes (), with increased multitasking worsening performance. Further findings reveal that extroverted students who frequently use digital communication platforms tend to have lower academic achievement, influenced by their self-regulating ability ().
Other factors determining academic performance include student educational background, teaching philosophy, and relationships with colleagues and instructors; students with high self-esteem are more self-possessed in their competence to study and thus more motivated to accomplish their academic achievements, gaining better results than those with low self-esteem. Self-respect, educational involvement, and academic achievement are all related ().
Social media is a major component of modern digital engagement, and they are one of the leading factors of academic distraction. The main cause is the unmanageable time students spend staying online more frequently, especially engaging in non-educational activities on social media. Numerous studies have confirmed that this phenomenon negatively affects students' academic performance ().
Poor academic activities remain one of the most significant effects of SMA addiction among students (). According to the literature, students frequently engage in non-academic activities in social networks: such usage distracts from academic contexts () and may exert a harmful effect on student learning by decreasing concentration levels. Previous research has shown that social networking sites hinder students' academic performance ().
However, they also have social costs. They may lead to poor health, including mental health, poor academic performance, and decreased quality of time spent learning. Furthermore, they may adversely affect students' concentration and attention to learning. This can also result in negative consequences for students who need to control their time for online networking activity ().
H1. Social media addiction has a negative impact on academic performance
2.2 The mediating role of sleep quality
According to the national sleep foundation, healthy adults require 7–9 h of sleep per night, an amount associated with optimal cognitive function, emotional regulation, memory consolidation, and overall brain health (). While the precise duration needed for proper brain function may vary with age, insufficient sleep leads to cognitive impairment (). Central to this cognitive enhancement is the process of memory consolidation, where sleep reinforces neural circuits to ensure future information access (), a concept explained by models like the synaptic homeostasis hypothesis ().
Consequently, disruptions such as sleep deprivation or fragmentation are linked to a wide range of adverse health outcomes, including impaired daytime functioning, mood disturbances, and increased risks for cardiovascular metabolic diseases (; ). These risks apply to both insufficient and excessive sleep durations () and may be particularly detrimental for adolescents due to heightened neural adaptability during this critical developmental period.
In an academic context, poor sleep quality directly contributes to daytime fatigue and reduced memory capacity (). A primary driver of this issue is excessive social media use, which is strongly associated with poor sleep (; ). This pathway, where social media addiction leads to sleep disturbances and psychological imbalance, ultimately undermines academic performance and educational resilience among university students (; ; ).
H2. Sleep quality mediates the relationship between SMA and academic engagement.
2.3 The mediating role of fatigue
Fatigue is a complex psychophysiological state characterized by a deterioration in attention, working capacity, and motivation, which has both physical and mental manifestations. This condition is particularly common among university students and is consistently linked to lower academic performance (; ). It is important to recognize that fatigue is a multifaceted phenomenon, existing on a continuum. At one end is acute or temporary exhaustion, a normal physiological response that is typically alleviated by rest (). At the other end is a more persistent, chronic form that builds gradually, is not relieved by relaxation, and can escalate into burnout, a state of emotional exhaustion and withdrawal (; ).
Within this context, a significant modern contributor to student fatigue is Social Media Addiction (SMA). Research shows that students who are excessively engaged in social networking experience a higher neurological burden, reporting more frequent and intense fatigue than their non-addicted peers (). The constant stimulation from appealing online social activities leads to chronic mental tiredness, which in turn results in distraction and impaired cognitive performance.
This overstimulation of the brain can have negative consequences similar to physical overexertion, making even simple academic tasks seem insurmountable (). Considering this evidence, our study proposes the hypothesis that the heightened levels of fatigue induced by SMA serve as a key mediator that reduces engagement in educational activities ().
H3. Fatigue mediates the relationship between SMA and academic engagement.
2.4 The mediation process through sleep quality and fatigue
Sleep disorders like insomnia are prevalent conditions characterized not only by nighttime disruption but also by significant daytime impairments, such as chronic fatigue, which substantially reduces functional capacity (; ). Subjective chronic fatigue is characterized as a medically unexplained multisymptomatic complex associated with persistent burnout, physical pain, and cognitive impairment ().
Chronic Fatigue is defined by intense, lasting exhaustion within 6 months, substantially affecting daily activities. Complications with sleep disturbances are prevalent, affecting up to 95% of patients experiencing restorative sleep (). According to research by fatigue was revealed to be inversely linked to sleep length, time awake at night, and sleep efficiency in both excellent and poor sleepers.
This link between poor sleep and daytime exhaustion is particularly relevant for university students, a population uniquely vulnerable due to high academic workloads and nocturnal habits that negatively impact their quality of life (). The issue is compounded by high rates of digital addiction, as excessive use of online platforms is known to diminish sleep effectiveness and mental Wellbeing, consequently harming academic outcomes ().
The Developmental-Contextual Model of Digital Demands and Resources (DC-DDR) offers a framework to understand this dynamic through its “energy-draining” process. This model posits that when digital activities consume excessive time and psychological energy, they displace restorative routines like sleep, creating an imbalance that leads to exhaustion and reduced academic performance (). Therefore, grounded in this theoretical framework, our study hypothesizes that Social Media Addiction negatively influences academic performance through the mediating pathways of decreased sleep effectiveness and increased fatigue.
H4. The correlation between SMA and academic performance is mediated by sleep quality and fatigue, which are serial mediators.
2.5 Theoretical framework: developmental-contextual model of digital demands and resources (DC-DDR)
The Developmental-Contextual Model of Digital Demands and Resources (DC-DDR) offers an integrative framework to examine digital ups and downs for academic and psychological achievement. Applied in the framework of developmental psychology and digital behavior research, DC-DDR emphasizes how exposure to digital demands, such as extended screen time on digital platforms, influences cognitive, emotional, and behavioral appraisals of personal and contextual resources (; ). The DC-DDR model has been previously employed to identify the impact of digital engagement in various functional domains, including fatigue, sleep interference, and academic disengagement (; ).
The model explains that stressing academic tasks weakens academic interest and motivation by consuming cognitive resources, negatively affecting the quality of sleep, and causing chronic fatigue (; ). Two central mechanisms underpin the DC-DDR model: the Energy-Depleting and Motivational processes. The energy depletion results when technology-mediated activities demand time and effort, suggesting that time is used for enjoyable or necessary pursuits like sleep and focused academic work, resulting in mental fatigue. The motivational process underlines that if students do not have the psychological and behavioral capital to effectively self-regulate their digital engagement, lost focus and demotivation lead to lower academic outcomes for students ().
Previous studies have focused on the direct effects of digital platforms' presence on academic performance, ignoring the mediating role of sleep efficiency and exhaustion levels. However, recent studies indicate the need to examine these mediating paths to better understand educational disaffiliation (). Other papers have also pointed to self-regulation as another moderating factor that mediates the adverse effects of digital demands on academic achievement ().
Using the DC-DDR model's theoretical framework, this research will enhance theoretical and empirical identification of the relationship between digital demands, sleep efficiency, fatigue, and their impact on students' engagement. This approach emphasizes that to enhance the learning process in the digital context, innovative learning strategies should blend with appropriate time, adequate provisions for the regulation of the device, and sufficient break time (). The proposed theoretical model is shown in Figure 1, which illustrates the hypothesized relationships between social media addiction, sleep quality, fatigue, and academic performance.
Figure 1
3 Materials and methods
3.1 Study design and target populations
To examine the predicted model using empirical data, the study population was undergraduate students from four universities (Jamhuriya, Jazeera, University of Somalia, and City University) in Mogadishu-Somalia. Quantitative design was employed to explore the influence of addiction to Social Media sites on educational accomplishment through Sleep quality and Fatigue. These systematic methods significantly contribute to understanding the underlying consequences of addiction to extreme digital platforms on educational outcomes ().
As previously conducted on the consequences of excessive Social media usage on educational performance in Somalia, this research will emphasize both the implications of networking sites for academic performance and the underlying effects (Mediation): Sleep quality and Fatigue. Some of the selected undergraduate students from higher education institutions were surveyed.
For this study, a convenience sampling technique was employed to recruit students from four universities in Somalia. This method is commonly used in exploratory social science research, as it allows for efficient data collection from accessible and willing participants (; ). It was chosen due to its practicality in a setting where obtaining official, comprehensive student lists for random sampling is administratively challenging and may raise privacy concerns.
Data were collected using a structured questionnaire distributed via Google Forms. The survey link was primarily shared through established WhatsApp groups managed by student representatives and department administrators at the participating universities. These groups function as major channels for academic and social communication among students.
Prior approval was obtained from university authorities to access these groups for participant recruitment. A standardized message outlining the study's purpose, voluntary participation, and anonymity was posted alongside the survey link. Data collection occurred for 3 months (Nov 2024–January 2025), during which periodic reminder messages were sent to encourage participation and improve response rates.
This approach facilitated practical access while minimizing sampling bias.
The demographic characteristics of the final sample, including each university representative sample and their response rates, are summarized in Table 1. Therefore, we evaluated the reliability of all items employing Cronbach's alpha(α), and the overall items indicated over 0.7, which is the standard threshold ().
Table 1
| Variable | Item | Frequency | Percent (%) |
|---|---|---|---|
| Age | ≥18 and ≤ 20 | 363 | 64.1 |
| ≥21 and ≤ 23 | 172 | 30.4 | |
| ≥24 and ≤ 26 | 27 | 4.8 | |
| ≥27 and above | 4 | 0.7 | |
| Total | 566 | 100 | |
| Gender | Male | 357 | 63.1 |
| Female | 209 | 36.9 | |
| Total | 566 | 100 | |
| Qualification | Computer and IT | 457 | 80.7 |
| Business and Management | 39 | 6.9 | |
| Engineering | 23 | 4.1 | |
| Health Sciences | 47 | 8.3 | |
| Total | 566 | 100 | |
| Distribution by University | University Name | Representative Sample | Response Rate (%) |
| Jamhuriya University | 420 | 225 (54%) | |
| Jazeera University | 170 | 170 (61%) | |
| University of Somalia | 113 | 133 (68%) | |
| City University | 58 | 58 (46%) | |
| Total | 990 | 566 (57%) | |
| Internet usage | < 1 h | 204 | 36 |
| ≥1 to ≤ 2 h | 196 | 34.6 | |
| ≥3 to ≤ 4 h | 60 | 10.6 | |
| ≥5 to ≤ 6 h | 60 | 10.6 | |
| More than 6 h | 46 | 8.1 | |
| Total | 566 | 100 |
Sample frequency, percentage (N = 566).
3.2 Measurements used
A. Bergen Social Media Addiction (BSMAS), which was initially referred to as the Bergen Facebook Addiction Scale BFAS and later renamed The Bergen Social Media Addiction (), is used to measure the level of Social Addiction (). The word social media here refers to Facebook, X (Formerly Twitter), Snapchat, and many others. The scale contains six items that measure the addiction level with a 5-point Responses were measured on a 5 Likert scale from 1 = Very rarely, 2 = Rarely, 3 = Sometimes, 4 = Often, 5 = Very Often, thus yielding a composite score from 6 to 30 in total, with a higher score indicating a greater degree of Social Media Addiction (). Students with a total score equal to or >19 were considered to have SMA.
B. The Utrecht Work Engagement Scale (UWES-9S) was employed to assess the academic performance (). The scale, which consists of three main categories that measure academic vigor, dedication, and absorption, was adopted. This UWES-9S was measured at a 7-point Likert from 0 (never) to 6 (always) ().
C. Fatigue Assessment Scale (FAS) was employed to assess fatigue (). The FAS contains ten elements, and each is rated on a 5-point Likert scale from 1 (never) to 5 (always), with advanced scores revealing advanced stages of subjective fatigue. This scale has been widely used as a real self-reporting tool to determine subjective fatigue (; )
D. The Pittsburgh Sleep Quality Index (PSQI) was assessed to examine the efficiency of sleep (). The scale consists of seven items, with subjective sleep quality, sleep duration, sleep duration, sleep latency, sleep disturbances, habitual sleep efficiency, sleeping medication use, and daytime dysfunction, measured by a Likert scale from 0 to 3. The overall score on the measure is the cumulative of the scores for these seven elements, ranging from 0 to 21 ().
Therefore, considering contemporary awareness about SMA and its effect on academic performance, this research employs structured data collection methods to limit data bias sources and minimize measurement errors. Realizing how students engage in digital behaviors and the impacts they could stimulate, quantitative research was used to examine the moderating functions of sleep efficiency and fatigue between SMA and academic participation.
3.3 Ethical approval
To ensure the research was conducted ethically, approval was sought from the Jamhuriya University of Science Research Ethics Committee (JUREC) with a reference number of JUREC0102/CGS315/052024. Such approval guarantees compliance with international ethical standards within the framework of the Declaration of Helsinki. Furthermore, all participants agreed to use their data through their faculty deans, which ensured their willingness to participate and the confidentiality of their information.
Furthermore, the research design was implemented using SPSS 25 for coordinating demographic analysis, giving researchers a precise approach to identifying participant characteristics such as age, gender, academic qualification, and internet usage. In addition, PLS-SEM was used to determine direct, indirect, and mediating influences, thus providing a clear picture of the effects of digital platform addiction affecting educational engagement through sleep efficiency and tiredness.
Due to these ethical and methodological precautions, this study enhances the knowledge of digital addiction and academic achievement. It contributes to education policies and social student welfare measures with real-life data.
4 Result
The results of the PLS-SEM analysis are presented below. The analysis begins with the validation of the measurement model, followed by the testing of the structural model and the research hypothesis ().
4.1 Respondent's profile
Table 1 presents the demographic distribution of the 566 undergraduate students surveyed from four universities in Mogadishu, Somalia. The majority of participants (64.1%) were between 18 and 20 years old, followed by those aged 21–23 years (30.4%). The sample was predominantly male (63.1%) over female (36.9%). Regarding academic discipline, students from the faculty of Computer and IT constituted the largest group at 80.7%. The remaining were from health Sciences (8.3%), Business and Management (6.9%), and Engineering (4.1%). For daily internet use, 36% of students were online for < 1 h, and 8.1% for more than 6 h.
4.2 Measurement model
The first step of using PLS-SEM is to assess the measurement model. It contains reliability and validity and maintains internal steadiness for convergent and discrimination (Figure 2, Table 2). All the items were above the standard threshold of 0.708, except AC4, AC5, AC8, F4, F3, F10, SP3, SP5, and SP9, which were removed due to their low loadings. Cronbach's alpha (α) and composite reliability (CR) were conducted to construct internal consistency reliability, and the CR value was higher than the standard threshold of 0.70.
Figure 2
Table 2
| Item | Indicator | Outer loadings | α | CR | AVE | VIF |
|---|---|---|---|---|---|---|
| AC | AC1 | 0.897 | 0.924 | 0.709 | ||
| AC2 | 0.775 | |||||
| AC3 | 0.888 | 2.507 | ||||
| AC6 | 0.895 | |||||
| AC7 | 0.852 | |||||
| FT | FT1 | 0.855 | 0.890 | 0.536 | ||
| FT2 | 0.715 | |||||
| FT5 | 0.767 | 1.724714 | ||||
| FT6 | 0.677 | |||||
| FT7 | 0.728 | |||||
| FT8 | 0.751 | |||||
| FT9 | 0.791 | |||||
| SM | SM1 | 0.883 | 0.907 | 0.551 | ||
| SM2 | 0.771 | |||||
| SM3 | 0.762 | |||||
| SM4 | 0.777 | 1.724714 | ||||
| SM5 | 0.656 | |||||
| SM6 | 0.787 | |||||
| SM7 | 0.789 | |||||
| SM8 | 0.631 | |||||
| SP | SP1 | 0.860 | 0.891 | 0.538 | ||
| SP10 | 0.710 | |||||
| SP2 | 0.732 | 1.757857 | ||||
| SP4 | 0.757 | |||||
| SP6 | 0.743 | |||||
| SP7 | 0.742 | |||||
| SP8 | 0.740 |
Outer loadings, validity, reliability, and collinearity (VIF).
α, Cronbach's Alpha; CR, Composite reliability; AVE, average extracted; VIF, variance inflation factor; AC, Academic Performance; FT, Fatigue; SM, social media; SP, Sleep Quality.
The average variance extracted was also tested and exceeded the recommended standard of 0.5. The item's discriminant validity was also measured using the heterotrait–monotrait ratio (HTMT) criterion, Fornell–Larcker Criterion, and Cross loadings (). Tables 3–5 indicate that all items meet the required thresholds, thereby confirming the establishment of discriminant validity.
Table 3
| Item | AC | FT | SM | SP |
|---|---|---|---|---|
| AC | ||||
| FT | 0.092 | |||
| SM | 0.143 | 0.447 | ||
| SP | 0.146 | 0.408 | 0.275 |
Discriminant validity—HTMT criterion.
Table 4
| Construct | AC | FT | SM | SP |
|---|---|---|---|---|
| AC | 0.842 | |||
| FT | −0.043 | 0.732 | ||
| SM | −0.129 | 0.397 | 0.742 | |
| SP | 0.118 | 0.369 | 0.256 | 0.734 |
Discriminant validity—Fornell–Larcker criterion.
Table 5
| Construct | AC | FT | SM | SP |
|---|---|---|---|---|
| AC1 | 0.792 | −0.052 | −0.082 | 0.061 |
| AC2 | 0.775 | 0.032 | −0.046 | 0.145 |
| AC3 | 0.888 | −0.085 | −0.169 | 0.063 |
| AC6 | 0.895 | −0.068 | −0.119 | 0.102 |
| AC7 | 0.852 | 0.004 | −0.104 | 0.126 |
| FT1 | −0.002 | 0.688 | 0.226 | 0.247 |
| FT2 | −0.112 | 0.715 | 0.264 | 0.225 |
| FT5 | −0.065 | 0.767 | 0.308 | 0.262 |
| FT6 | −0.022 | 0.677 | 0.280 | 0.271 |
| FT7 | −0.008 | 0.728 | 0.300 | 0.303 |
| FT8 | 0.013 | 0.751 | 0.326 | 0.267 |
| FT9 | −0.030 | 0.791 | 0.317 | 0.308 |
| SM1 | −0.068 | 0.285 | 0.748 | 0.174 |
| SM2 | −0.072 | 0.302 | 0.771 | 0.177 |
| SM3 | −0.117 | 0.256 | 0.762 | 0.193 |
| SM4 | −0.184 | 0.201 | 0.777 | 0.149 |
| SM5 | −0.056 | 0.287 | 0.656 | 0.184 |
| SM6 | −0.084 | 0.338 | 0.787 | 0.198 |
| SM7 | −0.148 | 0.337 | 0.789 | 0.207 |
| SM8 | −0.044 | 0.313 | 0.631 | 0.221 |
| SP1 | 0.147 | 0.291 | 0.153 | 0.711 |
| SP10 | 0.052 | 0.274 | 0.162 | 0.710 |
| SP2 | 0.014 | 0.333 | 0.284 | 0.732 |
| SP4 | 0.096 | 0.317 | 0.220 | 0.757 |
| SP6 | 0.110 | 0.243 | 0.181 | 0.743 |
| SP7 | 0.088 | 0.168 | 0.130 | 0.742 |
| SP8 | 0.119 | 0.188 | 0.115 | 0.740 |
Discriminant validity—cross loadings.
The bold values indicate Discriminant validity – Cross Loadings indicate the factor loadings of each measurement item on its intended construct.
4.3 Structural model assessment
The structural model results are illustrated in Figure 3, highlighting the path coefficients, significance levels, and explained variances for each construct. After the validation of the validity and reliability measurement, the first step is to evaluate the model (). The next step is to evaluate the variance inflation factor (VIF) through multicollinearity. As we assessed the data, we found AC8 above the threshold and removed it. We re-assessed and was not found any multicollinearity; all values were within the standard threshold (). The next phase is to assess the accuracy of the proposed hypothesis using bootstrapping 10,000 sub-sample procedures.
Figure 3
The direct relationships between the study variables are presented in Table 6. We evaluated the direct effects of SMA on educational performance (ACP), Poor Sleep (SQ), and Fatigue (FT). Finally, we found the direct effects of SMA on academic performance, sleep quality, and fatigue. As the result in Table 6 shows, SMA is negatively and significantly correlated with ACP (β = −0.155, t = 3.152, P = 0.002), with SMA is certainly and substantially associated with SQ (β = 0.256, t = 6.168, P = 0.000), with SMA is positively and significantly correlated with FT (β = 0.324, t = 8.151, P = 0.000).
Table 6
| Hypothesis | Model | β | STDEV | t | P | Result |
|---|---|---|---|---|---|---|
| H1: | SMA → ACP | −0.155 | 0.049 | 3.152 | 0.002 | Accepted |
| H2: | SMA → SQ | 0.256 | 0.042 | 6.168 | 0.000 | Accepted |
| H3: | SMA → FT | 0.324 | 0.040 | 8.151 | 0.000 | Accepted |
| H4: | SQ → ACP | 0.174 | 0.046 | 3.756 | 0.000 | Accepted |
| H5: | SQ → FT | 0.286 | 0.040 | 7.125 | 0.000 | Accepted |
| H6: | FT → ACP | −0.045 | 0.047 | 0.967 | 0.334 | Rejected |
Direct relationships.
Moreover, we learned that SQ is directly positively, and significantly associated with ACP (β = 0.174, t = 3.756, P = 0.000). Similarly, SQ is direct, positively, and significantly associated with FT (β = 0.286, t = 7.125, P = 0.000). Hence, H1, H2, H3, and H4, and H5 were supported. Thus, H6 investigated the direct relationship with FT and ACP since we found it negatively impacted but (β = −0.045, t = 7.125, P = 0.000) was not significant and meaningfully impacts Academic performance. The mediation analysis provides indirect effects of SMA on educational performance (ACP), sleep quality (SQ), and fatigue (FT). The mediation analysis results are summarized in Table 7, showing the indirect pathways and their statistical significance.
Table 7
| Hypothesis | Constructs | β | SD | t | P | Result |
|---|---|---|---|---|---|---|
| H7: | SMA → SQ → FT | 0.073 | 0.015 | 4.836 | 0.000 | Supported |
| H8: | SMA → FT → ACP | −0.015 | 0.015 | 0.958 | 0.338 | Not Supported |
| H9: | SMA → SQ → ACP | 0.045 | 0.014 | 3.164 | 0.002 | Supported |
| H10: | SMA → SQ → FT → ACP | −0.003 | 0.004 | 0.902 | 0.367 | Not Supported |
| H11: | SQ → FT → ACP | −0.013 | 0.014 | 0.915 | 0.360 | Not Supported |
Mediation analysis.
The investigators found that SQ (Sleep Quality) Facilitates the correlation between digital platform habit (SMA) and fatigue (FT), and H7 is supported (β = 0.073, t = 4.836, P = 0.000). Similarly, SQ (Sleep Quality) Facilitates the correlation between digital platforms and educational engagement (ACP). Therefore, H9 was supported (β = 0.045, t = 3.164, P = 0.002). However, the indirect effects of H8 (SMA → FT → ACP), H10 (SMA → SQ → FT → ACP), H11 (SQ → FT → ACP) were rejected. This indicates that fatigue alone or together with SMA nor SQ does not affect ACP and was not significant, respectively (β = −0.015, t = 0.958, P = 0.338), (β = −0.003, t = 0.902, P = 0.367), (β = −0.013, t = 0.915, P = 0.360). The model's descriptive power was assessed using the coefficient of determination (R2).
The results show that the predictors explained 4.2% of the variance in academic performance (ACP), 6.6% in fatigue (FT), and 23.5% in SQ (Table 8). These findings suggest that the model provides meaningful descriptive power for SP. Furthermore, the relevance (Q2) of the constructs was evaluated. As shown in Table 8, the Q2 values for ACP (0.011), FT (0.060), and SP (0.149) are all above the standard threshold of 0, confirming the relevance of the model. Notably, SQ demonstrates stronger predictive power, while ACP and FT present opportunities for enhancement by incorporating additional influencing factors.
Table 8
| Construct | R2 | R-square adjusted | Q2 |
|---|---|---|---|
| AC | 0.042 | 0.037 | 0.011 |
| FT | 0.235 | 0.232 | 0.060 |
| SQ | 0.066 | 0.064 | 0.149 |
Model's explanatory power.
5 Discussion
This research investigated the impact of Social Media Addiction (SMA) on the academic performance of students in Somali universities, examining sleep efficiency and exhaustion as mediating variables. Our findings confirmed a significant negative relationship between SMA and academic engagement (H1). This result aligns with a substantial body of literature demonstrating that excessive social media use corresponds with reduced commitment to studies, lower participation in educational activities, and ultimately, poorer academic outcomes.
These findings show that individuals who excessively use SMA and encounter addictive behaviors find it inspiring to stay focused on their academic activities; these distractions caused by SMA not only interfere with learners' ability to focus but also reduce their mental perseverance, which is significant for maintaining engagement in academic performance. Similar results were found by , , , , , and .
Therefore, the results align with other countries' evidence and strengthen SMA's influence on academic performance. In the second hypothesis (H2), the study found that SMA is negatively related to sleep quality and mediates SMA and academic engagement. These results also support previous studies conducted by (; ; ; ; ; ; ), which has supported that extended use of social media, particularly at night, affects sleep.
These disruptions to learning and development may impact pace, behavior, and organization, causing learners to lose focus or motivation. Thus, the third hypothesis (H3) indicates that sleep quality significantly mediates the relationship between SMA and academic performance. Improving sleep could be the solution to minimizing the effects of SMA on academic performance, which also supports previous studies conducted by () and .
Furthermore, in hypotheses 4 (H4) and 5 (H5), SMA remarkably correlates with fatigue. This study shows that students who are overexposed to SMA experience higher levels of mental fatigue and emotional distress due to heavy social media engagement. Additionally, sleep quality mediates the relationship between SMA and fatigue. Lack of quality sleep was associated with increased fatigue, which aligns with the findings demonstrating that disruptions in sleep result in daytime tiredness, decreased concentration, and lowered performance at academic tasks.
These indications also support the previous studies conducted by , , and and implies that students who are addicted to social media will likely suffer from fatigue linked to compromised sleep, which will affect their functioning and academic activity. Most significantly, hypotheses about the serial mediation pathway (SMA → sleep quality → fatigue → academic engagement) were not supported.
This is similar to the previous study by and who observed no correlation between social fatigue and academic performance. According to their research, fatigue results from digital overwhelm, but it is not necessarily reflected in lost academic functioning. Its negative effects, however, can be offset through other cognitive or behavioral interventions like improvements in study skills and self-regulation in digital learning environments. Likewise, this study's lack of a strong indirect effect indicates that fatigue alone may not fully explain academic disengagement in the presence of social media addiction.
Our finding of a significant negative correlation between Social Media Addiction (SMA) and academic performance is well explained by the Developmental-Contextual Model of Digital Demands and Resources (DC-DDR). The model's two main mechanisms, energy depletion and motivational mechanism, support our results, particularly the mediating role of sleep quality.
The energy-depleting mechanism suggests that technology consumes cognitive resources. Our results provide strong support for this model as SMA predicts poorer sleep quality, which in turn lowers academic performance (supporting H9). This demonstrates how SMA disrupts restorative sleep, a critical process for cognitive function and academic readiness. Interestingly, while sleep quality mediates SMA and fatigue (supporting H7), fatigue itself did not emerge as a significant mediator in our proposed model (rejecting H8, H10, H11).
Consequently, the primary academic harm in our findings comes from the cognitive consequences of poor sleep, rather than subjective fatigue. This aligns with the findings of and . The motivational process of the DC-DDR framework explains the direct negative link between SMA and academic performance. This model highlights failure in self-regulation, where the immediate reward of social media competes with the delayed gratification of academic success.
As the framework suggests, when students lack the psychological and behavioral capacity to effectively self-regulate (), their focus is captured by social media platforms, leading to demotivation and disengagement observed in our findings. This motivational conflict operates independently of the sleep pathway and explains why SMA has its own detrimental effect.
5.1 The study limitations
The study has several limitations that warrant consideration. Though we emphasized that the statistically significant path from SMA through sleep quality to academic performance is a key finding from our study, regardless of the overall R2 value. However, we acknowledge that academic performance is likely influenced by additional variables that were beyond the scope of our study. The primary contribution of our study is not in explaining the total variance in academic performance, but in successfully testing a specific theoretical pathway proposed by the DC-DDR framework.
The 4.2% variance explained in our study is consistent with findings from similar research on social media and academic outcomes, which often report modest effect sizes due to the complex, multifactorial nature of academic behavior in higher education settings (; ), suggesting a broader investigation into other factors that contribute to academic progress. This study utilized a cross-sectional design, which constrains the capacity to derive causal inferences, rather than exploring the association of the IV to DV.
Moreover, employing self-reported metrics may add biases stemming from subjective impressions and the accuracy of reporting. A notable limitation of this study is the overrepresentation of participants from the Faculty of Computer and IT, who accounted for 80.7% of the respondents. Despite efforts to distribute the survey equally across faculties, the response rate from non-IT disciplines was relatively low, likely due to differences in screen usage and accessibility. As a result, the findings may not be fully generalizable to students from other academic backgrounds.
Future research should aim for more balanced disciplinary representation to improve generalizability. Additionally, employing longitudinal or experimental methods could help establish causal links, while incorporating objective measures alongside self-reports could reduce bias. Finally, expanding the sampling to include a wide range of academic faculties and institutions would further enhance the representation.
6 Conclusion
The study investigated the mechanisms through which social media addiction (SMA) impacts academic performance among undergraduate students in Somalia, testing sleep quality and fatigue as mediators. Our findings suggest that SMA harms academic performance via two distinct pathways. First, there is a direct negative relationship, suggesting that SMA creates a motivational conflict that diverts attention from academic goals. Second, and more significantly, SMA has a powerful indirect effect by degrading sleep quality, which in turn diminishes academic performance. Critically, while SMA was linked to higher fatigue, fatigue was not a significant mediator in our model. This key finding suggests that cognitive impairment from poor sleep is a more direct driver of academic decline than the subjective feeling of being tired. These results support our DC-DDR model, confirming that SMA acts through both motivational and energy-depletion mechanisms.
6.1 Recommendations and future direction
The findings of this study offer practical implications for university administrators, educators, and mental health professionals in Somalia and similar contexts. Instead of advocating for unrealistic measures like digital detoxes or technology bans, approaches that are often unsustainable for digitally native students, our results emphasize more practical, educational, and supportive interventions. Universities should develop targeted awareness campaigns that highlight the connection between late-night social media use, poor sleep, and academic underperformance. Programs focusing on sleep strategies, such as setting a digital wellbeing plan and avoiding screens before bedtime, can help encourage healthier habits.
The study also recommends integrating digital literacy and self-regulation skills into student orientation programs. These initiatives should aim to equip students with the tools to manage their technology use responsibly, without demonizing social media. Training could focus on monitoring screen time, recognizing compulsive use, and fostering focused, productive work. In addition to that, the faculty also has a role to play. They need to employ interesting and interactive teaching approaches to keep the students attentive in class and reduce their tendency to escape by browsing their social media accounts. However, universities must offer counseling and support services to students who may have developed social media dependency or are under pressure due to the large workload. Introducing such support systems would ensure that the faculty members assist students in obtaining the necessary support to manage the new digital behaviors.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available upon reasonable request from corresponding author.
Ethics statement
The studies involving humans were approved by Ethical Committee, Jamhuriya University of Science and Technology. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
MK: Investigation, Writing – original draft, Data curation, Software, Formal analysis, Visualization, Funding acquisition, Resources, Conceptualization, Project administration, Supervision, Validation, Methodology, Writing – review & editing. AO: Writing – review & editing, Validation, Supervision, Visualization. AA: Formal analysis, Software, Methodology, Conceptualization, Writing – review & editing. MM: Conceptualization, Software, Writing – review & editing, Methodology, Formal analysis.
Funding
The author(s) declare that no financial support was received for the research and/or publication of this article.
Acknowledgments
The author would like to acknowledge and express gratitude to his supervisor, Dr. Abdifetah Ibrahim Omar, Dr. Nur Rashid Ahmed, for their invaluable guidance and rigorous oversight throughout the development of this paper.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Gen AI was used in the creation of this manuscript.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
Abi-JaoudeE.NaylorK. T.PignatielloA. (2020). Smartphones, social media use and youth mental health. Can. Med. Assoc. J., 192, E136–E141. 10.1503/cmaj.190434
2
Abu-SnienehH. M.ArouryA. M. A.AlsharariA. F.Al-GhabeeshS. H.EsailehA. A. (2020). Relationship between sleep quality, using social media platforms, and academic performance among university students. Perspect. Psychiatr. Care56, 415–423. 10.1111/ppc.12450
3
AlageelA. A.AlyahyaR. A.BahatheqA.AlzunaydiY.AlghamdiN. A.AlrahiliR. A.et al. (2021). Smartphone addiction and associated factors among postgraduate students in an Arabic sample: a cross-sectional study. BMC Psychiatry21:302. 10.1186/s12888-021-03285-0
4
AlapinI.FichtenC. S.LibmanE.CretiL.BailesS.WrightJ.et al. (2000). How is good and poor sleep in older adults and college students related to daytime sleepiness, fatigue, and ability to concentrate?J. Psychosom. Res.49, 381–390. 10.1016/S0022-3999(00)00194-X
5
Al-BarashdiH.BouazzaA.JaburN. (2015). Smartphone addiction among university undergraduates: a literature review. J. Sci. Res. Rep.4, 210–225. 10.9734/JSRR/2015/12245
6
Al-DhanhaniA.MizouniR.OtrokH.Al-RubaieA. (2015). Analysis of collaborative learning in social network sites used in education. Soc. Netw. Anal. Min.5:65. 10.1007/s13278-015-0303-z
7
Al-GarniA.AlamriH.AsiriW.AbudasserA.AlawashizA.BadawiF.et al. (2024). Social media use and sleep quality among secondary school students in Aseer Region: a cross-sectional study. Int. J. Gener. Med.17, 3093–3106. 10.2147/IJGM.S464457
8
Ali HomaidA. (2022). Problematic social media use and associated consequences on academic performance decrement during COVID-19. Addict. Behav.132:107370. 10.1016/j.addbeh.2022.107370
9
AlkhalafA. M.TekianA.ParkY. S. (2018). The impact of WhatsApp use on academic achievement among Saudi medical students. Med. Teach.40, S10–S14. 10.1080/0142159X.2018.1464652
10
AmanoA.AhmedG.NigussieK.AsfawH.FekaduG.HikoA.et al. (2023). Internet addiction and associated factors among undergraduate students of Jimma University; Jimma, South West Ethiopia, institutional based cross-sectional study. BMC Psychiatry23:721. 10.1186/s12888-023-05197-7
11
AminM. H. J.AlhadiH. A.MohamedA. E. A.YacoubH. K. E. T.KhalifaR. M. H.DafallahI. A. E.et al. (2024). Impact of smartphone addiction on health status, mental Wellbeing, and sleep quality among medical students in Sudan. BMC Psychiatry24:965. 10.1186/s12888-024-06377-9
12
AndradeC. (2021). The Inconvenient Truth About Convenience and Purposive Samples. Indian J. Psychol. Med.43, 86–88. 10.1177/0253717620977000
13
AndreassenC. S.BillieuxJ.GriffithsM. D.KussD. J.DemetrovicsZ.MazzoniE.et al. (2016). The relationship between addictive use of social media and video games and symptoms of psychiatric disorders: a large-scale cross-sectional study. Psychol. Addict. Behav.30, 252–262. 10.1037/adb0000160
14
BanksS.Van DongenH. P. A.MaislinG.DingesD. F. (2007). Behavioral and Physiological Consequences of Sleep Restriction. J. Clin. Sleep Med.03, 519–528. 10.5664/jcsm.26918
15
BányaiF.ZsilaÁ.KirályO.MarazA.ElekesZ.GriffithsM. D.et al. (2017). Problematic social media use: results from a large-scale nationally representative adolescent sample. PLoS ONE12:e0169839. 10.1371/journal.pone.0169839
16
BaumeisterR. F.VohsK. D.TiceD. M. (2007). The strength model of self-control. Curr. Dir. Psychol. Sci.16, 351–355. 10.1111/j.1467-8721.2007.00534.x
17
BekaluM. A.McCloudR. F.ViswanathK. (2019). Association of social media use with social wellbeing, positive mental health, and self-rated health: disentangling routine use from emotional connection to use. Health Educ. Behav.46, 69S−80S. 10.1177/1090198119863768
18
BergC. A.UpchurchR. (2007). A developmental-contextual model of couples coping with chronic illness across the adult life span. Psychol. Bull.133, 920–954. 10.1037/0033-2909.133.6.920
19
BuysseD. J.ReynoldsC. F.MonkT. H.BermanS. R.KupferD. J. (1989). The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res.28, 193–213. 10.1016/0165-1781(89)90047-4
20
CadimaR.OjedaJ.MonguetJ. M. (2025). Social networks and performance in distributed learning communities. Educ. Technol. Soc.15, 296–304. Available online at: https://www.jstor.org/stable/jeductechsoci.15.4.296
21
ÇiftciN.SarmanA.ÇobanM. (2025). The relationship between social media addiction, insomnia, and depression in adolescents. Psychol. Health Med.14, 1–16. 10.1080/13548506.2025.2465659
22
Cortés-RamosA.Torrecilla GarcíaJ. A.Landa-BlancoM.Poleo GutiérrezF. J.Castilla MesaM. T. (2021). Activism and social media: youth participation and communication. Sustainability13:10485. 10.3390/su131810485
23
DaiB.LinY.LaiX.HeJ.DuM.HouX.et al. (2024). The effects of self-esteem and parental phubbing on adolescent internet addiction and the moderating role of the classroom environment: a hierarchical linear model analysis. BMC Psychiatry24:55. 10.1186/s12888-023-05489-y
24
De BruinE. J.Van RunC.StaaksJ.MeijerA. M. (2017). Effects of sleep manipulation on cognitive functioning of adolescents: a systematic review. Sleep Med. Rev.32, 45–57. 10.1016/j.smrv.2016.02.006
25
De VriesJ.MichielsenH. J.Van HeckG. L. (2003). Assessment of fatigue among working people: a comparison of six questionnaires. Occup. Environ. Med.60, i10–i15. 10.1136/oem.60.suppl_1.i10
26
DharmawanA.SetiadiT.KarliI. (2024). The paradox of switching social medias: a mixed-method study of students' academic performance and social fatigue. Simulacra7, 182–194. 10.21107/sml.v7i2.26754
27
Digital Around the World (2025). DataReportal—Global Digital Insights. Available online at: https://datareportal.com/global-digital-overview (Accessed August 6, 2025).
28
DixonS. J. (2024). Topic: Social media. Statista. Available online at: https://www.statista.com/topics/1164/social-networks/ (Accessed August 6, 2025).
29
DongW.TangH.WuS.LuG.ShangY.ChenC.et al. (2024). The effect of social anxiety on teenagers' internet addiction: the mediating role of loneliness and coping styles. BMC Psychiatry24:395. 10.1186/s12888-024-05854-5
30
Erduran TekinÖ. (2024). Academic self-discipline as a mediating variable in the relationship between social media addiction and academic achievement: mixed methodology. Humanit. Soc. Sci. Commun.11:1096. 10.1057/s41599-024-03633-x
31
EtikanI. (2016). Comparison of convenience sampling and purposive sampling. Am. J. Theor. Appl. Stat.5:1. 10.11648/j.ajtas.20160501.11
32
EversK.ChenS.RothmannS.DhirA.PallesenS. (2020). Investigating the relation among disturbed sleep due to social media use, school burnout, and academic performance. J. Adolesc.84, 156–164. 10.1016/j.adolescence.2020.08.011
33
Fekih-RomdhaneF.SassiH.CheourM. (2021). The relationship between social media addiction and psychotic-like experiences in a large nonclinical student sample. Psychosis13, 349–360. 10.1080/17522439.2020.1861074
34
FengZ.DiaoY.MaH.LiuM.LongM.ZhaoS.et al. (2022). Mobile phone addiction and depression among Chinese medical students: the mediating role of sleep quality and the moderating role of peer relationships. BMC Psychiatry22:567. 10.1186/s12888-022-04183-9
35
Fortier-BrochuÉ.Beaulieu-BonneauS.IversH.MorinC. M. (2010). Relations between sleep, fatigue, and health-related quality of life in individuals with insomnia. J. Psychosom. Res.69, 475–483. 10.1016/j.jpsychores.2010.05.005
36
FournierL.SchimmentiA.MusettiA.BoursierV.FlayelleM.CataldoI.et al. (2023). Deconstructing the components model of addiction: an illustration through “addictive” use of social media. Addict. Behav.143:107694. 10.1016/j.addbeh.2023.107694
37
GezginciE.GulecH.YildirimG. (2025). The effect of TikTok use on internet addiction of college students in Somalia: a cross-sectional study. Int. J. Environ. Health Res.0, 1–11. 10.1080/09603123.2024.2448488
38
Gillen-O'NeelC.HuynhV. W.FuligniA. J. (2013). To study or to sleep? The academic costs of extra studying at the expense of sleep. Child Dev.84, 133–142. 10.1111/j.1467-8624.2012.01834.x
39
HaandR.ShuwangZ. (2020). The relationship between social media addiction and depression: a quantitative study among university students in Khost, Afghanistan. Int. J. Adolesc. Youth25, 780–786. 10.1080/02673843.2020.1741407
40
HairJ.AlamerA. (2022). Partial Least Squares Structural Equation Modeling (PLS-SEM) in second language and education research: guidelines using an applied example. Res. Methods Appl. Linguist. 1:100027. 10.1016/j.rmal.2022.100027
41
HamidS.WaycottJ.KurniaS.ChangS. (2015). Understanding students' perceptions of the benefits of online social networking use for teaching and learning. Internet High. Educ.26, 1–9. 10.1016/j.iheduc.2015.02.004
42
HashemiA.NooriA. Q.OrfanS. N.AkramyS. A.Mohd RameliM. R. (2024). Undergraduate students' perception of smartphone addiction and its impact on themselves and their academic performance: a case study. Cogent Educ. 11:2340845. 10.1080/2331186X.2024.2340845
43
HenselerJ.RingleC. M.SarstedtM. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. Acad. Market. Sci.43, 115–135. 10.1007/s11747-014-0403-8
44
HirshkowitzM.WhitonK.AlbertS. M.AlessiC.BruniO.DonCarlosL.et al. (2015). National sleep foundation's sleep time duration recommendations: methodology and results summary. Sleep Health1, 40–43. 10.1016/j.sleh.2014.12.010
45
HoL. Y. W.LaiC. K. Y.NgS. S. M. (2021). Measuring fatigue following stroke: the Chinese version of the fatigue assessment scale. Disabil. Rehabil.43, 3234–3241. 10.1080/09638288.2020.1730455
46
HofmannW.FrieseM.WiersR. W. (2008). Impulsive versus reflective influences on health behavior: a theoretical framework and empirical review. Health Psychol. Rev.2, 111–137. 10.1080/17437190802617668
47
HuangL.ZhangJ.DuanW.HeL. (2023a). Peer relationship increasing the risk of social media addiction among Chinese adolescents who have negative emotions. Curr. Psychol.42, 7673–7681. 10.1007/s12144-021-01997-w
48
HuangP-. C.LatnerJ. D.O'BrienK. S.ChangY-. L.HungC-. H.ChenJ-. S.et al. (2023b). Associations between social media addiction, psychological distress, and food addiction among Taiwanese university students. J. Eat. Disord.11:43. 10.1186/s40337-023-00769-0
49
JiangF.VanDykeR. D.ZhangJ.LiF.GozalD.ShenX.et al. (2011). Effect of chronic sleep restriction on sleepiness and working memory in adolescents and young adults. J. Clin. Exp. Neuropsychol.33, 892–900. 10.1080/13803395.2011.570252
50
KillgoreW. D. S. (2010). “Effects of sleep deprivation on cognition,” in Progress in Brain Research (Vol. 185) (Amsterdam: Elsevier), 105–129.
51
KimS. Y.KimM-. S.ParkB.KimJ-. H.ChoiH. G. (2017). The associations between internet use time and school performance among Korean adolescents differ according to the purpose of internet use. PLoS ONE12:e0174878. 10.1371/journal.pone.0174878
52
Landa-BlancoM.GarcíaY. R.Landa-BlancoA. L.Cortés-RamosA.Paz-MaldonadoE. (2024). Social media addiction relationship with academic engagement in university students: the mediator role of self-esteem, depression, and anxiety. Heliyon10:e24384. 10.1016/j.heliyon.2024.e24384
53
LatifianM.AarabiM. A.EsmaeiliS.AbdiK.RahebG. (2024). The role of internet addiction and academic resilience in predicting the mental health of high school students in Tehran. BMC Psychiatry24:420. 10.1186/s12888-024-05853-6
54
MadaiahM.SeshaiyengarC.SureshP.MunipapannaS.SonnappaS. (2016). Study to assess the effects of social networking sites on medical college students. Int. J. Commun. Med. Public Health3:1204–1208. 10.18203/2394-6040.ijcmph20161385
55
MahdiuonR.SalimiG.RaeisyL. (2020). Effect of social media on academic engagement and performance: perspective of graduate students. Educ. Inform. Technol.25, 2427–2446. 10.1007/s10639-019-10032-2
56
MaksniemiE.HietajärviL.KetonenE. E.LonkaK.PuukkoK.Salmela-AroK. (2022). Intraindividual associations between active social media use, exhaustion, and bedtime vary according to age—A longitudinal study across adolescence. J. Adolesc.94, 401–414. 10.1002/jad.12033
57
MalikA.DhirA.KaurP.JohriA. (2020). Correlates of social media fatigue and academic performance decrement: a large cross-sectional study. Inform. Technol. People34, 557–580. 10.1108/ITP-06-2019-0289
58
MancaS.RanieriM. (2017). Implications of social network sites for teaching and learning. Where we are and where we want to go. Educ. Inform. Technol.22, 605–622. 10.1007/s10639-015-9429-x
59
MichielsenH. J.VriesD. e.Van HeckJ. (2003). Psychometric qualities of a brief self-rated fatigue measure. J. Psychosom. Res.54, 345–352. 10.1016/S0022-3999(02)00392-6
60
NaganeM.SugeR.WatanabeS-. I. (2016). Time or retiring and sleep quality may be predictors of academic performance and psychosomatic disorder in university students. Biol. Rhythm Res.47, 329–337. 10.1080/09291016.2015.1126076
61
NguyenT.TranH.NguyenG.NguyenY.DinhH. (2025). Narcissism, social media addiction, self-esteem, and haxeco traits: exploring influences on life satisfaction among generation Z. Psychol. Res. Behav. Manage.18, 419–434. 10.2147/PRBM.S447067
62
PekpazarA.Kaya AydinG.AydinU.BeyhanH.AriE. (2021). Role of instagram addiction on academic performance among Turkish university students: mediating effect of procrastination. Comput. Educ. Open2:100049. 10.1016/j.caeo.2021.100049
63
RingleC. M.da SilvaD.BidoD. S. (2014). Modelagem de Equações Estruturais com Utilização do Smartpls. Revista Brasileira de Market.13, 56–73. 10.5585/remark.v13i2.2717
64
RosenL. D.Mark CarrierL.CheeverN. A. (2013). Facebook and texting made me do it: media-induced task-switching while studying. Comput. Human Behav.29, 948–958. 10.1016/j.chb.2012.12.001
65
RussellC.WeardenA. J.FaircloughG.EmsleyR. A.KyleS. D. (2016). Subjective but not actigraphy-defined sleep predicts next-day fatigue in chronic fatigue syndrome: a prospective daily diary study. Sleep39, 937–944. 10.5665/sleep.5658
66
SalariN.ZareiH.RasoulpoorS.GhasemiH.Hosseinian-FarA.MohammadiM.et al. (2025). The impact of social networking addiction on the academic achievement of university students globally: a meta-analysis. Public Health Pract.9:100584. 10.1016/j.puhip.2025.100584
67
SamahaM.HawiN. S. (2017). Corrigendum to “Relationships among smartphone addiction, stress, academic performance, and satisfaction with life” [Computers in Human Behavior 57 (April 2016) 321–325]. Comput. Human Behav.75:1016. 10.1016/j.chb.2016.12.001
68
SarbazvatanH.AminiA.AminisaniN.ShamshirgaranS. M. (2017). Sleep quality and academic progression among students of Tabriz University of Medical Sciences, Northwest of Iran. Res. Dev. Med. Educ.6, 29–33. 10.15171/rdme.2017.006
69
SchaufeliW. B.BakkerA. B.SalanovaM. (2006). The measurement of work engagement with a short questionnaire: a cross-national study. Educ. Psychol. Meas., 66, 701–716. 10.1177/0013164405282471
70
ScholzeA.HeckerA. (2023). Digital job demands and resources: digitization in the context of the job demands-resources model. Int. J. Environ. Res. Public Health20:6581. 10.3390/ijerph20166581
71
ShahnawazM. G.RehmanU. (2020). Social networking addiction scale. Cogent Psychol.7:1832032. 10.1080/23311908.2020.1832032
72
ShankarA.CharumathiS.KalidindiS. (2011). Sleep duration and self-rated health: the national health interview survey 2008. Sleep, 34, 1173–1177. 10.5665/SLEEP.1232
73
ShenJ.BarberaJ.ShapiroC. M. (2006). Distinguishing sleepiness and fatigue: focus on definition and measurement. Sleep Med. Rev.10, 63–76. 10.1016/j.smrv.2005.05.004
74
TamuraN.KomadaY.InoueY.TanakaH. (2022). Social jetlag among Japanese adolescents: association with irritable mood, daytime sleepiness, fatigue, and poor academic performance. Chronobiol. Int.39, 311–322. 10.1080/07420528.2021.1996388
75
TiesingaL. J.DassenT. W. N.HalfensR. J. G.Van Den HeuvelW. J. A. (1999). Factors related to fatigue; priority of interventions to reduce or eliminate fatigue and the exploration of a multidisciplinary research model for further study of fatigue. Int. J. Nurs. Stud., 36, 265–280. 10.1016/S0020-7489(99)00022-X
76
Turhan GürbüzP.ÇobanÖ. G.ErdoganA.KopuzH. Y.AdanirA. S.ÖnderA.et al. (2021). Evaluation of internet gaming disorder, social media addiction, and levels of loneliness in adolescents and youth with substance use. Subst. Use Misuse56, 1874–1879. 10.1080/10826084.2021.1958856
77
UpadhayayN.GuragainS. (2017). Internet use and its addiction level in medical students. Adv. Med. Educ. Pract.8, 641–647. 10.2147/AMEP.S142199
78
WalkerM. P. (2009). The role of sleep in cognition and emotion. Ann. N. Y. Acad. Sci., 1156, 168–197. 10.1111/j.1749-6632.2009.04416.x
79
WalshJ. L.FielderR. L.CareyK. B.CareyM. P. (2013). Female college students' media use and academic outcomes: results from a longitudinal cohort study. Emerg. Adulthood1, 219–232. 10.1177/2167696813479780
80
WooK. S.BongS. H.ChoiT. Y.KimJ. W. (2021). Mental health, smartphone use type, and screen time among adolescents in South Korea. Psychol. Res. Behav. Manage.14, 1419–1428. 10.2147/PRBM.S324235
81
YooS-. S.HuP. T.GujarN.JoleszF. A.WalkerM. P. (2007). A deficit in the ability to form new human memories without sleep. Nat. Neurosci.10, 385–392. 10.1038/nn1851
82
YoungK. S. (2004). Internet addiction: a new clinical phenomenon and its consequences. Am. Behav. Sci.48, 402–415. 10.1177/0002764204270278
83
YusoffS.YusoffR.Md NohN. H. (2017). Blended learning approach for less proficient students. Sage Open7:2158244017723051. 10.1177/2158244017723051
84
ZhuangJ.MouQ.ZhengT.GaoF.ZhongY.LuQ.et al. (2023). A serial mediation model of social media addiction and college students' academic engagement: the role of sleep quality and fatigue. BMC Psychiatry23:333. 10.1186/s12888-023-04799-5
Summary
Keywords
social media addiction, academic performance, sleep quality, fatigue, Somali Universities Students
Citation
Khalaf MAA, Omar AI, Abdulle AW and Mohamud MA (2025) Social media addiction and academic engagement: the role of sleep quality and fatigue among university students in Somalia. Front. Educ. 10:1614746. doi: 10.3389/feduc.2025.1614746
Received
19 April 2025
Accepted
31 July 2025
Published
03 September 2025
Volume
10 - 2025
Edited by
Ayşe Bulut, Bozok University, Türkiye
Reviewed by
Dewi Rosiana, Bandung Islamic University, Indonesia
Zihadur Rahman, Uttara University, Bangladesh
Updates
Copyright
© 2025 Khalaf, Omar, Abdulle and Mohamud.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Mohamed A. A. Khalaf khalaf@just.edu.so
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.