The Impact of Social Media Use on the Academic Performance of University Students: A PRISMA 2020 Systematic Review

d with university records Facebook, YouTube, Instagram, LinkedIn, Twitter Ahmad (2019) Nigeria 100 Cross- sectional survey 4-point Likert scale; frequency of SM use Self-reported grade point aggregate Facebook, WhatsApp, Twitter, Instagram IJCSMT 4.2 Quality Assessment Results Table 2 presents the quality appraisal results for each study. The Malaysian study by Mensah and Nizam (2016) ranked highest, achieving a score of 7 out of 8. This study distinguished itself by employing a multi-predictor regression model, reporting strong reliability measures like Cronbach's alpha, and openly discussing its limitations. Following that, the Bangladeshi study by Rahman and Mithun (2021) received a score of 6 out of 8. Its notable strengths included a large sample size, verification of self-reported CGPA with university records, and the use of a clear scatterplot to illustrate the findings. The Pakistani study by Hasnain et al. (2015) earned a score of 5, with its primary limitation being the use of only a single predictor in its regression analysis, which reduced its explanatory power. Lastly, the Nigerian study by Ahmad (2019) scored 4 out of 8, relying mainly on frequency distributions instead of inferential statistics, and not formally reporting reliability or validity. Table 2. Quality Assessment of Included Studies Study Sample (Domain 1) Instrument Validity (Domain 2) Statistical Rigour (Domain 3) Limitation Transparency (Domain 4) Total (/8) Hasnain et al. (2015) Moderate (1 pt) Moderate (1 pt) Moderate (1 pt) High (2 pts) 5 — Moderate Mensah & Nizam (2016) Moderate (1 pt) High (2 pts) High (2 pts) High (2 pts) 7 — High Rahman & Mithun (2021) High (2 pts) Moderate (1 pt) High (2 pts) Moderate (1 pt) 6 — Moderate Ahmad (2019) Moderate (1 pt) Low (0 pts) Low (0 pts) High (2 pts) 4 — Low 4.3 Risk of Bias Table 3 presents the risk of bias results for each included study. Selection bias was rated High across all four studies, as none used probability-based sampling from a nationally representative frame: three studies used convenience sampling and Ahmad (2019) used random sampling from a single institution only. Information bias was rated High for three studies owing to exclusive reliance on self-reported social media usage and GPA, and Moderate for the Malaysian and Bangladeshi studies where GPA was partly corroborated by institutional records. Confounding bias was rated High for all four studies, as none adequately controlled for socioeconomic status, prior academic achievement, dispositional self-regulation, or teaching quality. Table 3. Risk of Bias Assessment Study Selection Bias Information Bias Confounding Bias Overall Hasnain et al. (2015) High High High High Mensah & Nizam (2016) High Moderate High High Rahman & Mithun (2021) High Moderate High High Ahmad (2019) High High High High IJCSMT 4.4 Synthesis of Findings 4.4.1 Social Media Usage Patterns Almost every university student across these four studies used social media daily often for hours at a stretch. In Bangladesh, Rahman and Mithun (2021) found that 98% of students maintained active Facebook accounts. Even more striking, 46% reported spending between 41 and 55 hours per week just on Facebook. That represents a significant portion of their time, likely cutting into hours intended for studying or other academic tasks. When students discussed their reasons for using social media, the responses were consistent in all four countries: the majority wanted to stay in touch with friends and family (41% in Bangladesh), others used it for entertainment (32%), some sought information (21%), and only a small fraction 6% used it for academic learning. Clearly, social and recreational uses outweighed academic purposes everywhere. This trend was also seen in Pakistan. Hasnain and colleagues (2015) discovered that 45.6% of students had integrated social media into their daily lives. Nearly 30% admitted to frequently losing track of time while browsing or chatting online. In Malaysia, Mensah and Nizam (2016) observed that students prioritized casual socializing over their coursework. In Nigeria, most university students (primarily 18 to 20 years old) spent their time on Facebook, WhatsApp, and Instagram (Ahmad, 2019). Facebook was the leading social media platform in all four countries. In Bangladesh, YouTube ranked second, while WhatsApp and Instagram were particularly popular in Nigeria.In Ghana, Snapchat had a significant presence. 4.4.2 Statistical Evidence on Social Media Use and Academic Performance Table 4 summarizes the key statistics from all four studies. Each study discovered the same result: students who spend more time on social media generally have lower academic performance. This supports the time-displacement hypothesis, which suggests that time spent on social media reduces the time available for studying. Table 4. Key Statistical Findings Across Included Studies Study n r β R2 / ΔR2 p-value Key Mediating / Moderating Variables Hasnain et al. (2015) 171 −0.239 −0.317 ΔR2 = .052 < .01 None tested (single-predictor bivariate model) Mensah & Nizam (2016) 102 — Time app.: −.303; Health add.: +.351; Nat. use: +.257; Friend conn.: +.168 R2 = .505 < .001 Time appropriateness, time duration, nature of usage, friend-people connection, health addiction, security/privacy IJCSMT Rahman & Mithun (2021) 200 Neg. trend (R2 = .374) N/A (scatter) R2 = .374 Not reported Weekly hours on SM vs. CGPA (extreme values: 15– 17 hrs → CGPA 3.89–3.93; 58–61 hrs → CGPA 2.67– 2.95) Ahmad (2019) 100 N/A N/A N/A Descriptive only Platform type, purpose of use, behaviour change (dual positive/negative effects reported) Hasnain et al. (2015) discovered in Pakistan that social media use was weakly and negatively correlated with GPAr = 0.239 (p < .01), β = 0.317 (p < .01)and accounted for just 5.2% of the variance (ΔR2 = .052). They acknowledged this small effect and noted that many other factors influence academic performance. In Malaysia, Mensah and Nizam (2016) conducted a multiple regression with six predictors, explaining a much larger proportion: 50.5% of the variance (R2 = .505; F = 16.142; p < .001). Time appropriateness emerged as the strongest negative predictor (β = −0.303, p < .001). Conversely, health addiction was the most significant positive predictor (β = +0.351, p < .001), while nature of usage (β = +0.257, p = .002) and friend-people connection (β = +0.168, p = .033) also had positive impacts. Time spent and privacy concerns were not significant predictors. In Bangladesh, Rahman and Mithun (2021) used a scatter plot to reveal a strong negative pattern (R2 = .374): students who spent the least time per week on social media (15–17 hours) had the highest CGPAs (3.89 3.93), whereas heavy users (58–61 hours) recorded the lowest CGPAs (2.67 2.95). Meanwhile, in Nigeria, Ahmad (2019) used frequency distributions and found both negative and positive effects 77% of students reported that social media affected their writing and speaking, and 88% believed it changed their behavior. 4.4.3 Mediating and Moderating Variables Mensah and Nizam’s 2016 study in Malaysia stand out for its clear emphasis on what really influences how social media influences students. They make a distinct distinction between time appropriateness meaning when students are on social media, such as during class or study time versus outside those periods and time duration, or simply the total hours spent online. Here is the key point: it is not the sheer number of hours that hurts academic results, but the timing. Students who browse social media during classes or study sessions actually see their performance decline, while just spending a lot of time online is not as harmful. For those creating interventions, this highlights the importance of context and timing, rather than just imposing limits on screen time. The study also points to a troubling connection between compulsive social media use and health. Students who cannot put their devices down often disrupt their sleep and eating patterns, which affects both their physical health and their ability to concentrate and do well academically. This is not just a theory O’Keeffe and Pearson (2011) found the same trend. However, it is not all bad news. Mensah and Nizam also discovered that the purpose of social media use makes a difference. When students use social media for academic reasons like organizing study groups, exchanging notes, or contacting lecturers their academic performance actually benefits. This supports social capital theory: leveraging social networks to build relationships and collaborate can enhance academic achievement (Pasek et al., 2009; Wang et al., 2011). IJCSMT Across all four studies, one consistent finding emerges: the reasons behind, timing of, and people involved in students’ social media use have a much greater influence on academic results than just the amount of use. Those who primarily use social media for entertainment or casual chats with friends experience the greatest academic setbacks. However, students who use these platforms for peer support and collaborative learning tend to keep up academically or even excel. 5. Discussion 5.1 Cross-Contextual Consistency The main finding from this systematic review is unmistakable: in four very different developing- country settings, increased social media use is consistently linked to lower academic achievement. It made no difference which social media platforms were most popular, how affluent the country was, or how strong the universities were. In every study, the results pointed the same way students who devote more time to social media tend to perform worse academically. This trend appeared everywhere the study was examined, supporting the time-displacement hypothesis: essentially, when students invest time and mental effort in social media, they have less remaining for their studies. This pattern appears to hold across a variety of higher education systems in the developing world. These findings are in line with what researchers have observed in wealthier countries (see Karpinski & Duberstein, 2009; Madge et al., 2009; Kubey et al., 2001), but they also extend the existing evidence. Now, we have systematic results from universities in South Asia and Sub- Saharan Africa regions where reliable data on this issue have been scarce until now. 5.2 Divergence in Effect Magnitude All the studies point in the same direction, but the effect size varies greatly. In Pakistan, social media factors accounted for only 5.2% of the difference in academic performance, while in Malaysia, the figure jumped to 50.5%. That is a massive difference almost tenfold. What explains this gap? It has to do with how each study constructed its models. The researchers in Pakistan considered just a single variable, whereas the Malaysian team used six carefully selected predictors, each reflecting a different aspect of social media involvement. This highlights how critical it is to use multi-variable models in this area and how relying too much on simple, bivariate results can be misleading. In the future, researchers should follow the Malaysian approach: break social media use into its main components and examine them individually. The Nigerian study (Ahmad, 2019) did not use the strongest statistical techniques, but it still provides valuable insight. It found both positive and negative effects. Students benefited when they participated in academic forums, study groups, or followed official school pages. This double- sided pattern fits with social capital theory and is consistent with the Malaysian findings, where both how students used social media and their social connections had positive effects. 5.3 Theoretical Implications The findings reveal a bigger picture where time-displacement, internet addiction, social capital, and self-regulation theories all come together. These theories do not contradict each other they are connected. Social media can take away from study time, and excessive use may become a compulsion that negatively affects students’ health. However, when students use social media for academic reasons, it helps them develop social capital and improves their academic performance. Self-regulation is at the core of these relationships, influencing how these effects unfold. For policy makers, the message is clear: the focus should be on helping students develop self-regulation skills IJCSMT and monitoring when they use social media. Simply imposing blanket restrictions on screen time will not be effective. 5.4 Implications for Practice 5.4.1 Digital Literacy and Self-Regulation Education Universities in developing countries should reconsider how they support students. Rather than simply advising students to reduce social media use, orientation programs and pastoral care ought to teach digital literacy and self-regulation, using real evidence about effective strategies. Studies indicate that it is not the amount of time students spend online that matters most, but whether they use social media at appropriate times. Therefore, universities should assist students in establishing clear boundaries between study time and online time. Provide them with practical tools scheduled social media breaks, apps to monitor screen time, and possibly device-free zones for studying. This method gives students practical skills they can apply, not just rules they are likely to ignore. 5.4.2 Institutional Social Media Strategy Universities should take initiative with their official social media platforms. Since students are already active online, channeling that activity toward academic purposes makes sense. Rahman and Mithun (2021) highlight that more than 70% of Bangladeshi students follow their university’s official social media pages, and almost as many reports that these platforms make accessing academic materials easier. When instructors share course materials, assignment guidance, or additional resources in these groups, they connect with students’ daily routines and genuinely support their academic progress. 5.4.3 Student Wellbeing University counselling and wellbeing services should address compulsive social media use, as it is directly linked to disrupted sleep, irregular eating, and increased anxiety factors that Mensah and Nizam (2016) associate with academic performance. Students exhibiting unhealthy social media habits ought to have access to digital wellbeing programs, peer support, and clear counselling referral options. 5.5 Limitations of This Review Study-level limitations. All the studies included here used a cross-sectional design, so we cannot determine whether heavy social media use leads to lower grades or if students with lower grades are more likely to use social media. It is possible that social media is distracting students, but it is also possible that those already having trouble in class turn to their phones for support. The available studies do not clarify this. To truly understand the relationship, we need longitudinal panel data or randomized controlled trials that follow students over time. Measurement limitations. In every study, social media use was measured through self-reports. This poses an issue. Research in digital health consistently finds that people tend to underestimate their screen time when asked. Self-reported data does not align with the actual usage recorded by their devices. Researchers should start using objective data from sources like Apple Screen Time or Google Digital Wellbeing, and pair that with verified academic records, rather than relying on participants’ memory or honesty. Sampling limitations. All these studies drew participants from a single institution in one country. The study from Pakistan included seven institutions from two cities, but even that does not IJCSMT represent an entire nation. To make broader claims, research needs to include samples from multiple schools, across various regions, and from a range of programs and academic disciplines. Confounding. None of these studies accounted for possible confounding variables. They did not control for factors such as socioeconomic status, prior academic performance, self-regulation, access to devices or high-speed internet, or the quality of teaching and institutional support. Without adjusting for these factors, it is impossible to isolate the unique effect of social media on academic achievement. Review-level limitations. This review included just four studies, found through a targeted rather than systematic search, with no registered protocol. It also did not conduct a thorough search of grey literature. With such a small number of studies, a meta-analysis was not possible; meaning the statistical strength of these conclusions is limited. These gaps should be kept in mind when considering the findings of this review. 6. Conclusion This systematic review, conducted in accordance with the PRISMA 2020 guidelines, examines four empirical studies from Pakistan, Malaysia, Bangladesh, and Nigeria. The results are remarkably consistent: students who spend more time on social media generally have lower academic performance. All four studies report the same trend. Increased social media usage is associated with lower GPA or CGPA, with regression coefficients ranging from β = −0.303 to β = −0.317. Depending on the analytical model, social media use accounts for between 5.2% and 50.5% of the variance in academic outcomes. Facebook is the most popular platform in all four countries, and most students use social media primarily for entertainment or socializing rather than academic purposes. However, the situation is more nuanced. It is not just the amount of time spent online that matters the way students use social media is also important. When students use these platforms to interact with classmates, form study groups, share resources, or communicate with instructors, the effect on academic performance can be neutral or even positive. Therefore, rather than banning or restricting access to social media, universities should encourage students to leverage these platforms in ways that enhance their learning. Blanket bans are not only impractical; they may also be counterproductive. That said, the current evidence has limitations. The studies rely on cross-sectional data and self- reported surveys, and they do not always control for other variables that could affect the outcomes. This introduces a significant risk of bias. More robust research is needed longitudinal studies, objective measures of social media use, verified academic records, and larger, more representative samples to truly understand how social media influences academic performance and to identify effective interventions. Social media is an integral part of university students’ daily lives and is unlikely to disappear. The main challenge for universities, educators, and policymakers is to help students develop the self- regulation and digital literacy needed to make social media a tool for academic achievement rather than a source of distraction. IJCSMT References Ahmad, S. A. (2019). Social media and students' academic performance in Nigeria. Asian Journal of Education and e-Learning, 7(1), 29–36. Akhtar, N. (2013). Relationship between internet addiction and academic performance among university undergraduates. Educational Research and Reviews, 8(19), 1793–1796. Akubugwo, I. G., & Burke, M. (2013). Effect of social media on postgraduate students during academic lectures and library session: A case study of Salford University Manchester, United Kingdom. IOSR Journal of Research & Method in Education, 3(6), 44–50. Alwagait, E., Shahzad, B., & Alim, S. (2015). Impact of social media usage on students' academic performance in Saudi Arabia. Computers in Human Behavior, 51, 1092–1097. Boyd, D., & Ellison, N. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), 1–11. Drury, G. (2008). Social media: Should marketers engage and how can it be done effectively? Journal of Direct, Data and Digital Marketing Practice, 9, 274–277. Hasnain, H., Nasreen, A., & Ijaz, H. (2015). Impact of social media usage on academic performance of university students. In Proceedings of the 2nd International Research Management & Innovation Conference (IRMIC 2015), Langkawi, Malaysia. Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J., & Welch, V. A. (Eds.). (2022). Cochrane handbook for systematic reviews of interventions (version 6.3). The Cochrane Collaboration. www.training.cochrane.org/handbook Jeong, T. G. (2005). The effect of internet addiction and self-control on achievement of elementary school children. Korean Journal of Yeolin Education, 5(3). Karpinski, A. C., & Duberstein, A. (2009). A description of Facebook use and academic performance among undergraduate and graduate students. Paper presented at the Annual Meeting of the American Educational Research Association, San Diego, CA. Kubey, R., Lavin, M., & Barrows, J. (2001). Internet use and collegiate academic performance decrements: Early findings. Journal of Communication, 51(2), 366–382. Madge, C., Meek, J., Wellens, J., & Hooley, T. (2009). Facebook, social integration and informal learning at university. Learning, Media and Technology, 34(2), 141–155. Mensah, S. O., & Nizam, I. (2016). The impact of social media on students' academic performance: A case of Malaysia tertiary institution. International Journal of Education, Learning and Training, 1(1), 14–21. Nalwa, K., & Anand, A. P. (2003). Internet addiction in students: A cause of concern. CyberPsychology & Behavior, 6(6), 653–656. O'Keeffe, G. S., & Pearson, K. C. (2011). The impact of social media on children, adolescents and families. Pediatrics, 127(4), 800–804. Owusu-Acheaw, M., & Larson, A. G. (2015). Use of social media and its impact on academic performance of tertiary institution students: A study of students of Koforidua Polytechnic, Ghana. Journal of Education and Practice, 6(6), 94–101. Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., ... & Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. British Medical Journal, 372, n71. https://doi.org/10.1136/bmj.n71 Pasek, J., More, E., & Hargittai, E. (2009). Facebook and academic performance: Reconciling a media sensation with data. First Monday, 14(5). Paul, J. A., Baker, H. M., & Cochran, J. D. (2012). Effect of online social networking on student academic performance. Computers in Human Behavior, 28(6), 2117–2127. IJCSMT Rahman, S., & Mithun, M. N. A. S. (2021). Effect of social media use on academic performance among university students in Bangladesh. Asian Journal of Education and Social Studies, 20(3), 1–12. Safko, L., & Brake, D. K. (2009). The social media bible: Tactics, tools, and strategies for business success. John Wiley & Sons. Talaue, G. M., Saad, A. A., Rushaidan, N. A., Hugail, A. A., & Fahhad, A. A. (2018). The impact of social media on academic performance of selected college students. International Journal of Advanced Information Technology, 8(5), 27–34. Wang, Q., Chen, W., & Liang, Y. (2011). The effects of social media on college students. The Alan Shawn Feinstein Graduate School, Johnson & Wales University. Wells, G. A., Shea, B., O'Connell, D., Peterson, J., Welch, V., Losos, M., & Tugwell, P. (2013). The Newcastle-Ottawa Scale for assessing the quality of nonrandomised studies in meta-analyses. Ottawa Hospital Research Institute. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp