Next Article in Journal
One-Year Follow-Up of Two Intensive Supplemental Reading and Mathematics Programs in Swedish Elementary School
Next Article in Special Issue
University Dropout in Granada: A Biographical Narrative Study Addressing Student Diversity Based on External Factors
Previous Article in Journal
Task-Centered Analysis of Higher Education Students’ Uses of Generative Artificial Intelligence
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Education Sciences
Volume 15
Issue 12
10.3390/educsci15121677
education-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Predictive Capacity of Social Media Addiction on Academic Engagement in University Students

by
Yosbanys Roque Herrera
1,*,
Santiago Alonso-García
2,
Dennys Vladimir Tenelanda López
3 and
Juan Antonio López Núñez
2
1
Facultad de Ciencias, Escuela Superior Politécnica de Chimborazo, Riobamba 060155, Ecuador
2
Departamento de Organización Escolar y Currículo, Universidad de Granada, 18071 Granada, Spain
3
Facultad de Ciencias de la Salud, Escuela Superior Politécnica de Chimborazo, Riobamba 060155, Ecuador
*
Author to whom correspondence should be addressed.
Educ. Sci. 2025, 15(12), 1677; https://doi.org/10.3390/educsci15121677
Submission received: 4 November 2025 / Revised: 7 December 2025 / Accepted: 9 December 2025 / Published: 12 December 2025
(This article belongs to the Special Issue New Educational Responsibilities in Today’s Society: The Role of Artificial Intelligence, Quality and Inclusion)
Browse Figure
Versions Notes

Abstract

Social media is an essential part of people’s lives worldwide. This study aimed to analyze the predictive capacity of social media addiction on academic engagement among students enrolled in the Faculty of Health Sciences at the National University of Chimborazo during the first academic period of 2023. The Social Media Addiction Questionnaire (ARS) and the Utrecht Work Engagement Scale (UWES-S-17) were applied to 1200 participants during an analytical study. According to the simple linear regression model, 11.2% of the variance in academic engagement levels was explained by social media addiction, with statistical significance (p < 0.05). The multiple linear regression model was significant, although it showed a low capacity to explain and predict the level of academic engagement, considering the dimensions of the level of addiction to social media (obsession, lack of control, and excessive use). The ROC curve parameters showed statistical significance, showing a moderate ability to discriminate insufficient academic commitment. The results serve as a basis for future studies and as a diagnostic basis for establishing policies and strategies in the institution where the research was conducted to increase academic engagement and reduce social media addiction.

1. Introduction

The accelerated development of the Internet since its inception, coupled with the expansion of connection coverage areas and a decrease in the cost of Internet-connected technology, has facilitated the rapid adoption of its use (Marengo et al., 2022; Hugo et al., 2020). In relation to this phenomenon, Yang et al. (2022) argue that the widespread use of virtual communication resources among Internet users facilitates the emergence of new forms of social interaction between individuals.
Consequently, social networks have become an essential part of people’s lives worldwide, with an estimated 4.5 billion users utilizing the services of Facebook, WhatsApp, YouTube, Telegram, TikTok, Instagram, LinkedIn, Pinterest, Twitter, and Snapchat, among others (Miñan Olivos et al., 2023). A meta-regression conducted by Cheng et al. (2021) revealed that addiction to this type of resource had a statistically significant moderating effect on cultural individualism, indicating a higher prevalence of this problem in countries with collectivist social policies.
In 2021, Valencia Ortiz et al. (2021) proposed that people found a way to escape the real (offline) world through social media in virtual space. In this context, they can avoid physical contact during socialization, facilitating the initiation and maintenance of relationships when it is difficult to do so directly. This resource is easily accessible, enables simultaneous communication regardless of the number of contacts and physical distance, allows Internet users to maintain their anonymity, and allows the course of communication to be controlled; this constitutes a risk for adequate psychosocial development in individuals (Acosta Tobón et al., 2022; Aliverdi et al., 2022; Marengo et al., 2022).
In this regard, Cheng et al. (2021) agree that social media addiction represents a health problem that manifests itself through the appearance of discomfort, anxiety, compulsive behavior, sleep disorders, obesity, lower back problems, fatigue, and procrastination, directly affecting people’s professional, academic, economic, and social performance. The presence of social media addiction is more notable among the young and adolescent population (Amador Ortíz, 2021; Ghozali et al., 2024; Marino et al., 2020; Sümen & Evgin, 2021; Li et al., 2018; Uma et al., 2020; Zhao et al., 2022).
Sánchez-Sánchez et al. (2024) established a set of typical characteristics of this addiction:
  • Compulsive behavior of uncontrollable consumption.
  • Decreased capacity for self-control.
  • Urgent need to mitigate a negative or dysphoric emotional state: depression, irritability, or anxiety, among others.
  • A perception that some symptoms, such as mood swings, salience, tolerance problems, relapses, and withdrawal symptoms, can only be alleviated by using this internet resource.
The multidimensional nature of social media addiction was highlighted by Shong and Jia (2024). These authors emphasized the need to implement action plans based on a thorough diagnosis of the situation in each context. In this regard, Basauri Delgado (2023) systematized a set of risk factors through a review of published research results:
  • Presence of other addictions, including online gaming, shopping, sex, drugs, and alcohol, among others.
  • Dependence on cell phone use at all times, without it being necessary.
  • Perception of feeling positive or desirable emotions when connected.
  • Manifestation of psychological symptoms, including negative affect, anxiety, suicidal ideation, attachment problems, sleep latency, frustration, depression, stress, sleep disorders, and need for appreciation from others.
  • Indicators of dependence on social media use based on time spent, frequency of connection, types of emotional reactions, misperceptions of the benefits they offer, and purposes for which they are used (socializing, entertainment, marketing, fun, etc.), among others.
  • Ways in which some cognitive elements are presented, including control, absorption, dissociation, metacognition, distraction, automatic thoughts, cognitive–emotional regulation, and perception of the usefulness of technology.
  • The manifestation of certain psychosocial elements includes attachment to parents, social isolation, a need for emotional support, a sense of not receiving social support, a feeling of hopelessness regarding the effectiveness of socialization, reactive restriction, loneliness, and excessive reliance on group networks for family interaction.
Various researchers have linked addictive behavior related to the use of social media with the postponement of academic activities by students. This affects their academic performance and generates a predisposition to cyberbullying, which manifests as aggression and interpersonal conflicts, in addition to the inappropriate use of information and communications technologies (ICT) in daily life (Salari et al., 2025; Colonio Caro, 2023).
Therefore, procrastination behavior in students is frequently related to addiction to social networks, which prevents adequate time management and is linked to academic commitment (Dominguez-Lara et al., 2020; Trujillo & Noé, 2020).
In this regard, the term “academic engagement” is defined as the magnitude of effort that students exert during the training process, consciously seeking to achieve optimal learning outcomes. This is reflected in the way they overcome the difficulties inherent in curricular and extracurricular activities and tasks (Torres-Escobar & Botero, 2021).
According to Fuster Guillen and Baños Chaparro (2021), adequate commitment among university students fosters their development of autonomy, passion for their studies, critical thinking, and physical and psychological well-being, among other positive effects. This, in turn, reduces university dropout rates and strengthens interpersonal relationships with their peers and teachers.
The state of academic commitment in university students can be influenced by various dimensions, including the effectiveness of self-learning strategies, the level of family support in providing professional training, the quality of teaching staff performance, and the relevance of institutional educational policy guidelines (Perkmann et al., 2021).
The motivation that university students have for their professional training is a factor that influences their levels of academic engagement (Remo Cossu & Awidi, 2022; Haseli Songhori & Salamti, 2024). Therefore, teachers must employ teaching strategies aimed at stimulating intrinsic learning in their students (Ramesh et al., 2023). Likewise, institutional authorities could generate incentives in this regard, such as extracurricular training activities; scholarships; awards for scientific, academic, sports, and cultural merits; and implementation of spaces that develop proactivity and leadership (teaching assistantships, research incubators), among other possible actions (Froment et al., 2023; Shomotova & Ibrahim, 2025).
The level of academic engagement in higher education students can be expressed through various emotional, behavioral, and cognitive dimensions; among which are institutional attachment, participation and attention in class, dedication to learning, interest in ensuring quality in completing academic tasks, and success in their social integration into the university context during curricular and extracurricular activities (Lizarte Simón et al., 2024).
As early as 1978, Rosenshine and Berliner (1978) proposed the importance of exploring all elements that could potentially impact academic engagement, considering this variable to be the primary regulator of school dropout. This position was supported by Tortosa Martínez et al. (2022). In this regard, a model implemented by López Angulo et al. (2021) revealed a direct and negative effect of academic engagement on the intention to drop out of school (β = −0.489; p < 0.001), accounting for 24% of the variability.
The negative effects of addiction on individuals’ commitment to their duties are well-documented; however, few publications report on the effects of excessive social media use on university students’ academic engagement.
The research process presented here started with the following question:
What predictive power does social media addiction have on academic engagement among students enrolled in the Faculty of Health Sciences at the National University of Chimborazo during the first academic term of 2023?
The objective was to analyze the predictive capacity of social media addiction on academic engagement among students enrolled in the Faculty of Health Sciences at the National University of Chimborazo during the first academic term of 2023.
Taking into consideration the proposed objective, the following research hypothesis was established:
The high level of addiction to social media may significantly explain the insufficient academic engagement of university students in health sciences careers.

2. Materials and Methods

The researchers proposed a quantitative study with a non-experimental design, analytical scope, and cross-sectional approach in a health professional training setting at an Ecuadorian university. It was approved by the UNACH Research Committee through Resolution No. 40-CIV-16-02-2022.

2.1. Participants

The entire population was included in this study, which consisted of 1200 students between the ages of 18 and 27, among whom the average age was 21.4 years, and 65.3% were female, enrolled in the first through sixth semesters in the Faculty of Health Sciences at the National University of Chimborazo, Ecuador, during the first academic term of 2023.
The population distribution according to enrollment characteristics was as follows:
  • According to career (Clinical Psychology 181, Medicine 243, Nursing 247, Clinical Laboratory 200, Physical Therapy 165, and Dentistry 164).
  • According to semester (First 248, Second 255, Third 180, Fourth 117, Fifth 165, and Sixth 235).

2.2. Data Collection

During the research process, two instruments were applied: the Social Media Addiction Questionnaire (ARS) and the Utrecht Work Engagement Scale (UWES-S-17), which are described below.
The ARS consists of 24 items that measure three dimensions using a scale of five possible values: (0) never, (1) rarely, (2) sometimes, (3) almost always, and (4) always (Escurra Mayaute & Salas Blas, 2014). The dimensions are as follows:
  • Obsession with the use of social networks, which is associated with an individual’s uncontrollable impulse to connect to the Internet, which can generate anxiety (items 2, 3, 5, 6, 7, 13, 15, 19, 22, and 23).
  • Lack of personal control, based on insufficient capacity for behavioral self-regulation of the compulsion to use social networks (items 4, 11, 12, 14, 20, and 24).
  • Excessive use of social networks, which establishes the extent of a person’s use that exceeds their needs (items 1, 8, 9, 10, 16, 17, 18, and 21).
This instrument was validated by Roque Herrera et al. (2024), who found a high overall internal consistency (α = 0.954), with values across its dimensions ranging from 0.823 to 0.901. They also determined a p-value of 0.000 in Bartlett’s sphericity test, in addition to an overall KMO of 0.966 and figures between 0.853 and 0.930 in its dimensions. In this regard, the results of a systematic bibliographic review conducted by Guzmán Brand and Gélvez García (2023) showed that the ARS was the most widely used instrument in the compiled studies.
The UWES-S-17 includes 17 items that are measured using a seven-value scale: (0) not at all, (1) almost not at all, (2) rarely, (3) sometimes, (4) quite a bit, (5) frequently, and (6) always. These ratings allow for determining the overall scale and that of the three dimensions it measures (Cruzat Aliaga, 2020), as follows:
  • Vigor represents energy, will, persistence, and mental resilience in carrying out academic efforts, including when coping with difficulties (items 1, 4, 8, 12, 15, and 17).
  • Dedication represents the intensity with which a student engages in academic self-preparation with a sense of meaning, inspiration, enthusiasm, pride, and ability to overcome challenges (items 2, 5, 7, 10, and 13).
  • Absorption represents the intensity of concentration in academic self-training, measuring the level of focus on completing the task (items 3, 6, 9, 11, 14, and 16).
The validity of the UWES-S-17 was established by Tristán Monrroy et al. (2021) through a study involving 223 students from the Autonomous University of San Luis Potosí, which found 66.82% of explained variance and a satisfactory model fit (RMSEA = 0.048 and CFI = 0.990).

2.3. Procedure

Throughout this study, the researchers adhered to the corresponding ethical principles. Participants provided their informed consent, ensuring respect for their autonomy; the study also obtained the appropriate institutional authorization for the application of the instrument. The data were organized using SPSS 23.0 software, and the results were used solely for scientific and academic purposes, with no malicious intent.

2.4. Statistical Analysis Procedure

The data were initially processed using descriptive statistical methods, which involved relative frequencies that allowed the study variables to be characterized within the study’s context.
The analysis of the predictive capacity of social media addiction on academic engagement was carried out using a simple linear regression model (which analyzed only the two main variables), following the methodology implemented by Bermejo Salmon (2020). Additionally, a multiple linear regression model was applied using a linear function with the independent variables, which included the global level of addiction to the use of social media and its respective dimensions (obsession, lack of control in its use, and excessive use). This model made it possible to categorize the participants into a level or group based on the values of the dependent variable.
This analysis assessed the degree of impact, and the factors that affect the level of addiction to social networks, using the following tests and inferential analyses: R2 of Cox and Snell and R2 of Nagelkerke (as indicators of goodness of fit), ANOVA (to establish the predictive statistical significance of the independent variables), collinearity analysis (which specifies the presence of biases and the tolerance of the model), and COR curve analysis (Fuster Rico et al., 2024) (to determine the discrimination capacity for students with insufficient academic commitment).

3. Results

When characterizing the level of academic engagement, it was found that only 4.2% of participants fell below the average level, while the latter was predominant, encompassing 40.5%. Regarding the level of social media addiction, the majority of students scored within the average level in the overall assessment (93.7%) and by dimension (between 52.4% and 55.7%); however, excessive social media use was a notable exception, with 63.5% scoring at a high level.

3.1. Simple Linear Regression Analysis Predicting Academic Engagement Status with Respect to Social Media Addiction Level

The correlation analysis, before structuring the linear regression models, reflected the existence of a statistically significant relationship (p < 0.01) between the main study variables and their respective dimensions (level of social media addiction and level of academic engagement). This relationship was negative and of mild to moderate intensity, with Pearson’s r values ranging from −0.101 to −0.297.
The results suggest that 11.2% of the academic engagement level can be explained by the level of addiction to social media (Table 1).
The model is statistically highly significant (Table 2) in predicting the state of academic engagement (p = 0.000 < 0.001).
The VIF value is less than 10, indicating the absence of collinearity; with VIF = 1.000, the model can be considered virtually bias-free. Likewise, the tolerance value was greater than 0.2, but the predictive capacity was negative, as indicated by the beta value (Table 3).
The analysis of the regression coefficients corroborated the statistical significance of the model in predicting the state of the dependent variable (p = 0.000 < 0.01), in addition to establishing that as addiction to social networks increases, academic commitment decreases by 0.023 units (Table 3).

3.2. Multiple Linear Regression Analysis Predicting Academic Engagement Status with Respect to Dimensions of Social Media Addiction

The verification of the predictive capacity of the social media addiction variable on the state of academic commitment motivated the multiple linear analysis to consider the dimensions of this relationship.
The results suggest that 6.9% of the variance in the state of academic engagement level can be explained by the levels of obsession with social media, lack of control of social media use, and excessive social media use (Table 4).
The model is statistically highly significant (Table 5) in predicting the state of academic engagement (p = 0.000 < 0.001).
Analysis of the regression coefficients confirmed the statistical significance of the model in predicting the state of the dependent variable (p < 0.05) using the three independent variables related to social media addiction (obsession, lack of control in use, and excessive use).
The VIF value indicates the absence of collinearity, with values of less than 10 for all predictor variables. Furthermore, the mean VIF value was 1.990, suggesting some caution regarding potential biases, although this value is not far from 1.000. Likewise, the tolerance value was greater than 0.2 for all predictor variables (Table 6).
The predictive capacity was negative in all cases according to the beta value, and it was also possible to establish the following (Table 6):
  • As the level of obsession with social networks increased, academic engagement decreased by 0.207 units, holding the other predictor variables in the model constant.
  • As the level of lack of control of social media use increased, academic engagement decreased by 0.155 units, holding the other predictor variables in the model constant.
  • As the level of excessive social media use increased, academic engagement decreased by 0.139 units, holding the other predictor variables in the model constant.
The proportions with the highest values of variance were not concentrated in a single factor, corroborating the absence of collinearity (Table 7), DW values above 4.0 were observed.
To verify the predictive capacity of social media addiction and its dimensions on insufficient academic engagement (nominal dichotomous qualitative variable), ROC curve analysis was performed (Figure 1). This allowed the following cut-off points to be established: obsession with social networks (cut-off point = 17.92; sensitivity = 0.784; specificity = 0.520); lack of use control of social networks (cut-off point = 14.58; sensitivity = 0.686; specificity = 0.329); excessive use of social networks (cut-off point = 18.75; sensitivity = 0.725; specificity = 0.460); global addiction to the use of social networks (cut-off point = 52.08; sensitivity = 0.725; specificity = 0.414).
The results show that all values of the area under the curve were greater than 0.5 for social media addiction and its studied dimensions, ranging from 0.715 to 0.736; therefore, they offer a moderate ability to discriminate students with insufficient academic commitment. Likewise, statistical significance was determined for the ROC curve parameters (p = 0.000) (Table 8).

4. Discussion

Consistent with the theoretical positioning of the authors of this study, Hrivnák and Jarábková (2022) acknowledged that the appropriate use of social media for academic purposes has a high beneficial potential for students. These authors researched 776 postgraduate students from different universities in Slovakia, observing a positive impact on the development of cooperation between researchers and academic engagement. In this regard, Estevez et al. (2023) reported that peer support on social media can influence stress through emotional engagement, thereby serving as a pathway to academic engagement. Therefore, it is important to point out that teachers should take advantage of the potential of this Internet resource during the teaching–learning process, instilling its responsible use. This deduction can also be supported from the results of a study on 308 students from Shiraz University, Iran, which showed an indirect and positive effect of digital informal learning on the relationship between students’ digital competence and their academic engagement, which was also supported by the results (β = 0.22; p = 0.0001) (Heidari et al., 2021).
Likewise, when investigating academic engagement among Argentine university students during the COVID-19 pandemic, Rigo (2021) observed, through the statements obtained in the qualitative phase, that participants positively valued social media as a means of collaborative work, allowing them to share information and interact while completing teaching tasks and activities. This benefitted their involvement, dedication, and satisfaction with learning, in addition to increasing their interest in the professional career they were pursuing; therefore, the appropriate use of social media can benefit students’ academic engagement.
Through modeling, Acosta Gonzaga (2023) established the effects of academic engagement on behavioral, emotional, metacognitive, cognitive, and learning involvement, as well as a statistically significant (p < 0.01) and positive correlation between metacognitive engagement and academic performance. These findings corroborate the importance of academic engagement in students’ academic success.
However, Amador Ortíz (2021) found a statistically significant association between academic failure and obsession, as well as excessive use of social media (p < 0.05). Fifty-eight percent of participants with a high level of social media addiction experienced academic failure, which is a finding that is indirectly consistent with those of the present study. This does not coincide with the findings of the present study, since more than 90% of the population had an average level of social media addiction, suggesting that efforts should focus on the dimension related to excessive use.
On the other hand, Bautista Quispe et al. (2023) found a moderate, negative, and statistically significant correlation between social media addiction and its dimensions with academic self-regulation (r = −0.410; p < 0.05), a factor that influences educational engagement. Likewise, Tanhan et al. (2024) developed a model that yielded a predictive capacity of social media addiction on cognitive flexibility, with a negative and significant effect (F = 64.26; R2 = 0.14; p < 0.01), explaining 14% of the variance. These results are indirectly consistent with those obtained in the present research, which showed a negative and significant relationship with respect to academic engagement (p < 0.01), showing that decreasing levels of addiction to social networks can improve the way in which students approach their learning.
Coinciding with the results obtained about the significant (p < 0.01) predictive capacity of the dimensions of social media addiction, Yana Salluca et al. (2022) also found that social media addiction was strongly and positively correlated with academic procrastination, an indicator of low academic engagement (r = 0.710; p = 0.01). Similarly, Caratıquıt and Caratıquıt (2023) reported a positive, moderate, and significant relationship between social media addiction and academic procrastination (β = 0.442, p < 0.001), as well as a negative and low-intensity relationship with academic performance (β = −0.113, p = 0.008). Thus, educational actions aimed at eliminating addiction to social networks will also decrease academic procrastination in university students. In accordance with this, Junco (2012) established that frequent Facebook checking (β = −0.068, p < 0.01) and video game abuse (β = −0.118, p < 0.001) were negative predictors of academic engagement. Likewise, constant chatting on Facebook was a negative predictor of the time students spent preparing for class.
In Nantong University in Jiangsu, China, Mou et al. (2024) developed a sequential model that found a statistically significant pathway: social anxiety → social media addiction → academic engagement → academic performance. This pathway was also significant (R 2 = −0.013; 95% CI: between −0.018 and −0.008), accounting for 6.8% of the total effect. Therefore, they deduced that social media addiction played a partial mediating role in the relationship between social anxiety and academic performance. These findings are very similar to those obtained in the present study in the multiple linear regression model (6.9% explanation, p = 0.000); in addition to reinforcing the support regarding the negative impact of social media addiction and low academic commitment on student performance.
Problematic internet use includes the misuse of social media as an important dimension. This variable showed significant predictive power for academic engagement in the present study (p < 0.001). Accordingly, in another study involving 4852 adolescents from various regions of mainland China, Liu et al. (2022) found a significant negative correlation between problematic internet use and academic engagement (p < 0.05). Their mediating model explained 35.9% of the variance in academic engagement, revealing a direct effect and an indirect effect on the following trajectories: problematic internet use → depression → insomnia → academic engagement; problematic internet use → depression → academic engagement; and problematic internet use → anxiety → academic engagement.
In accordance with the results obtained, two indirect indicators considered as potential unwanted consequences of social media addiction were analyzed in two investigations. Polanco et al. (2023) determined that victimization during peer relationships negatively predicted academic engagement (β = −0.17; p < 0.05) in 204 students from two public schools. Peker (2024) found that time management had a significant predictive capacity for academic engagement (β = 0.52; SE = 0.06; p < 0.001) among 242 university students majoring in Psychology and Sociology. These data suggest that programs are needed to promote the appropriate and ethical use of social networks as a means of interaction between peers in order to avoid aggressive communication behaviors that can lead to online harassment, which reduces the effectiveness of collaborative academic activities through this medium.
Trejos Gil et al. (2023) investigated a population of 381 university students from Medellín, Colombia, using a model, and found that the variant including the factors of anxiety, connection, and social networks explained 11.17% of the variance in engagement. This agrees with the results obtained in this study (11.2%; p = 0.000), which indicate that measures to promote the appropriate use of social networks among students can be implemented as part of pedagogical strategies aimed at increasing their academic engagement.

Limitations and Future Perspectives

The main limitation of this study was the model’s adjustment to the dimensions of a single instrument from which the data were obtained, and thus, from which both main variables were measured. It is recommended that future studies employ other valid instruments that include additional factors, thereby increasing the number of possible variants.
By assessing future prospects, the results obtained can serve as a basis for subsequent studies and as a diagnostic basis for establishing policies and strategies by the authorities of the institution where the research was conducted, with the aim of reducing social media addiction and increasing academic engagement, which will positively impact school dropout rates.

5. Conclusions

The results show a predominance of average levels of academic engagement and social media addiction. The researchers established a statistically significant correlation between these two variables and their respective dimensions, which was negative and of mild to moderate intensity in all cases.
The simple linear regression model was statistically significant and showed a moderate capacity to explain and predict the state of academic engagement based on social media addiction, with no collinearity observed.
The multiple linear regression model was also statistically significant, reflecting a low explanatory and predictive capacity for academic engagement, considering social media obsession, lack of social media use control, and excessive social media use as independent variables, and indicating that there was no collinearity.
The ROC curve parameters demonstrated statistical significance, and a moderate ability to discriminate against insufficient academic engagement was established.
This model made adjustments to the dimensions of a single instrument, from which the data were obtained. In future studies valid instruments must be employed that include additional factors, increasing the number of possible variants.
The results serve as a basis for future studies and as a diagnostic basis for establishing policies and strategies in the institution where the research was conducted to increase academic engagement and reduce social media addiction.
The main theoretical contribution of the research consists of identifying social media addiction as a variable capable of affecting academic engagement in university students.
Thus, the results indicate that higher education institutions should implement measures to promote the appropriate use of social networks among students as part of pedagogical strategies aimed at increasing their academic engagement.

Author Contributions

Conceptualization, Y.R.H. and D.V.T.L.; methodology, Y.R.H. and S.A.-G.; validation, Y.R.H., S.A.-G. and J.A.L.N.; formal analysis, Y.R.H.; investigation, Y.R.H. and S.A.-G.; resources, Y.R.H. and D.V.T.L.; data curation, Y.R.H.; writing—original draft preparation, Y.R.H. and D.V.T.L.; writing—review and editing, Y.R.H., S.A.-G. and J.A.L.N.; supervision, S.A.-G. and J.A.L.N.; project administration, Y.R.H.; funding acquisition, Y.R.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research has not received external funding. The APC was funded by the Escuela Superior Politécnica de Chimborazo.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of UNACH Research Committee (protocol code 40-CIV-16-02-2022 and date of approval: 16 February 2022).

Informed Consent Statement

Informed consent was obtained from all subjects involved in this study.

Data Availability Statement

The data supporting the results are available upon request from the corresponding author. An anonymized dataset will be provided to anyone who requests it for academic purposes and agrees to a confidentiality agreement, subject to applicable ethical requirements.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Acosta Gonzaga, E. (2023). The effects of self-esteem and academic engagement on university students’ performance. Behavioral Sciences, 13(4), 348. [Google Scholar] [CrossRef] [PubMed]
  2. Acosta Tobón, S. A., Ramírez Arroyave, A. Y., Saldarriaga Álvarez, Y. A., Uribe Ortiz, J., & Ruiz Mejía, N. (2022). Uso de redes sociales y autoimagen en adolescentes de instituciones educativas en Antioquia, Colombia. Revista Senderos Pedagógicos, 13(1), 23–41. [Google Scholar] [CrossRef]
  3. Aliverdi, F., Farajidana, H., Tourzani, Z. M., Salehi, L., Qorbani, M., Mohamadi, F., & Mahmoodi, Z. (2022). Social networks and internet emotional relationships on mental health and quality of life in students: Structural equation modelling. Boston Medical Center-Psychiatry, 22(1), 451. [Google Scholar] [CrossRef]
  4. Amador Ortíz, C. M. (2021). Influencia de la adicción a las redes sociales en la reprobación escolar de estudiantes de nivel licenciatura. CienciaUAT, 16(1), 62–72. [Google Scholar] [CrossRef]
  5. Basauri Delgado, M. A. (2023). Factores asociados a la adicción a redes sociales en universitarios: Una revisión sistemática y metaanálisis. Persona, 26(1), 11–56. [Google Scholar] [CrossRef]
  6. Bautista Quispe, J. A., Velazco Reyes, B., Estrada Araoz, E. G., Córdova-Rojas, L. M., & Ascona García, P. P. (2023). Adicción a las redes sociales y procrastinación académica en adolescentes de educación básica regular. Revista Universidad y Sociedad, 15(3), 509–517. [Google Scholar]
  7. Bermejo Salmon, M. (2020). Tratamiento del nivel de competencias laborales desde la regresión lineal simple. Retos de la Dirección, 14(1), 264–277. [Google Scholar]
  8. Caratıquıt, K. D., & Caratıquıt, L. J. C. (2023). Influence of social media addiction on academic achievement in distance learning: Intervening role of academic procrastination. Turkish Online Journal of Distance Education, 24(1), 1–19. [Google Scholar] [CrossRef]
  9. Cheng, C., Lau, Y., Chan, L., & Luk, J. W. (2021). Prevalence of social media addiction across 32 nations: Meta-analysis with subgroup analysis of classification schemes and cultural values. Addictive Behaviors, 117, 106845. [Google Scholar] [CrossRef]
  10. Colonio Caro, J. D. (2023). Revisión sistemática sobre la adicción a las redes sociales en adolescentes latinoamericanos entre el 2020–2022. Propósitos y Representaciones, 11(2), e1759. [Google Scholar] [CrossRef]
  11. Cruzat Aliaga, V. (2020). Validación de la escala de engagement académico de Utrecht en estudiantes de una universidad chilena [Master’s thesis, Universidad de Talca]. Available online: https://repositorio.utalca.cl/repositorio/items/4ea10d79-4cde-4f8b-806b-39f4480f433e (accessed on 7 September 2025).
  12. Dominguez-Lara, S., Sánchez-Villena, A. R., & Fernández-Arata, M. (2020). Propiedades psicométricas de la UWES-9S en estudiantes universitarios peruanos. Acta Colombiana de Psicología, 23(2), 7–23. [Google Scholar] [CrossRef]
  13. Escurra Mayaute, M., & Salas Blas, E. (2014). Construcción y validación del cuestionario de adicción a redes sociales (ARS). Liberabit, 20(1), 73–91. [Google Scholar]
  14. Estevez, G., O’Neal, C., & Cerrato, S. (2023). Impact of stress on later academic engagement among emergent bilingual students: Grit and academic support as protective factors. School Psychology Review, 54(2), 170–183. [Google Scholar] [CrossRef]
  15. Froment, F., De-Besa Gutiérrez, M., & Gil Flores, J. (2023). Efecto del apoyo a la autonomía sobre la satisfacción académica: La motivación y el compromiso académico como variables mediadoras. Revista de Investigación Educativa, 41(2), 479–499. [Google Scholar] [CrossRef]
  16. Fuster Guillen, F. G., & Baños Chaparro, J. (2021). Análisis Bayesiano del compromiso académico en estudiantes de psicología: Diferencias según sexo y edad. Revista de Investigación en Psicología, 24(1), 5–18. [Google Scholar] [CrossRef]
  17. Fuster Rico, A., Vicent, M., Gonzálvez, C., Pérez Marco, M., Granados Alós, L., & Suriá, R. (2024). Empleando análisis de perfiles y curvas ROC para examinar la relación entre el perfeccionismo y la autoeficacia académica en estudiantes de Educación Secundaria. Revista de Psicodidáctica, 29(2), 139–146. [Google Scholar] [CrossRef]
  18. Ghozali, G., Fatma, F., & Oktaviani, L. W. (2024). Relationship between social media addiction and anxiety among junior high school students. Indonesian Journal of Global Health Research, 6(S4), 191–202. [Google Scholar] [CrossRef]
  19. Guzmán Brand, V. A., & Gélvez García, L. E. (2023). Adicción o uso problemático de las redes sociales online en la población adolescente. Una revisión sistemática. Psicoespacios, 17(31), 1–22. [Google Scholar] [CrossRef]
  20. Haseli Songhori, M., & Salamti, K. (2024). The linkage between university students’ academic engagement and academic support: The mediating role of psychological capital. Iranian Journal of Educational Sociology, 7(2), 72–84. [Google Scholar] [CrossRef]
  21. Heidari, E., Mehrvarz, M., Marzooghi, R., & Stoyanov, S. (2021). The role of digital informal learning in the relationship between students’ digital competence and academic engagement during the COVID-19 pandemic. Journal of Computer Assisted Learning, 37(4), 1154–1166. [Google Scholar] [CrossRef]
  22. Hrivnák, M., & Jarábková, J. (2022). Drivers of academic engagement and university–industry collaboration in conditions of Slovakia. Administrative Sciences, 12(4), 128. [Google Scholar] [CrossRef]
  23. Hugo, F. J., Jiménez, C. E., Holovatyi, M., & Lara, P. E. (2020). El impacto de las redes sociales en la administración de las empresas. RECIMUNDO: Revista Científica de la Investigación y el Conocimiento, 4(1), 173–182. [Google Scholar] [CrossRef]
  24. Junco, R. (2012). The relationship between frequency of Facebook use, participation in Facebook activities, and student engagement. Computers & Education, 58(1), 162–171. [Google Scholar] [CrossRef]
  25. Li, J. B., Mo, P., Lau, J., Su, X. F., Zhang, X., Wu, A., Mai, J. C., & Chen, Y. X. (2018). Online social networking addiction and depression: The results from a large-scale prospective cohort study in Chinese adolescents. Journal of Behavioral Addictions, 7(3), 686–696. [Google Scholar] [CrossRef]
  26. Liu, S., Zou, S., Zhang, D., Wang, X., & Wu, X. (2022). Problematic internet use and academic engagement during the COVID-19 lockdown: The indirect effects of depression, anxiety, and insomnia in early, middle, and late adolescence. Journal of Affective Disorders, 309, 9–18. [Google Scholar] [CrossRef]
  27. Lizarte Simón, E. J., Gijón Puerta, J., Galván Malagón, M. C., & Khaled Gijón, M. (2024). Influence of self-efficacy, anxiety and psychological well-being on academic engagement during university education. Education Sciences, 14(12), 1367. [Google Scholar] [CrossRef]
  28. López Angulo, Y., Cobo Rendón, R. C., Pérez Villalobos, M. V., & Díaz Mujica, A. E. (2021). Social support, autonomy, academic commitment, and drop out intention in first year undergraduate students. Formación Universitaria, 14(3), 139–148. [Google Scholar] [CrossRef]
  29. Marengo, D., Fabris, M. A., Longobardi, C., & Settanni, M. (2022). Smartphone and social media use contributed to individual tendencies towards social media addiction in Italian adolescents during the COVID-19 pandemic. Addictive Behaviors, 126, 107204. [Google Scholar] [CrossRef] [PubMed]
  30. Marino, C., Lenzi, M., Canale, N., Pierannunzio, D., Dalmasso, P., Borraccino, A., Cappello, N., Lemma, P., Vieno, A., Nardone, P., & Spinelli, A. (2020). Problematic social media use: Associations with health complaints among adolescents. Annali dell’Istituto Superiore di Sanita, 56(4), 514–521. [Google Scholar] [CrossRef]
  31. Miñan Olivos, G. S., Flores Capcha, J. C., Piñas Illanes, A. R., Suria Huincho, Y. S., Toledo Figueroa, W. K., Garcia Zarzosa, J. A., & Mejia Huerta, G. C. (2023). Academic stress and social network addiction in Peruvian university students. Salud, Ciencia y Tecnología, 3, 519. [Google Scholar] [CrossRef]
  32. Mou, Q., Zhuang, J., Wu, Q., Zhong, Y., Dai, Q., Cao, X., Gao, Y., Lu, Q., & Zhao, M. (2024). Social media addiction and academic engagement as serial mediators between social anxiety and academic performance among college students. BMC Psychology, 12, 190. [Google Scholar] [CrossRef]
  33. Peker, M. (2024). Willing, able, and engaged: Roles of action-state orientation, intrinsic academic motivation, and time management on academic engagement. Current Psychology, 43(20), 18444–18454. [Google Scholar] [CrossRef]
  34. Perkmann, M., Salandra, R., Tartari, V., McKelvey, M., & Hughes, A. (2021). Academic engagement: A review of the literature 2011–2019. Research Policy, 50(1), 104114. [Google Scholar] [CrossRef]
  35. Polanco, C., Paskewich, B. S., Leff, S. S., & Waasdorp, T. E. (2023). Relational peer victimization as a predictor of academic engagement. Journal of Child and Family Studies, 32(7), 1882–1894. [Google Scholar] [CrossRef]
  36. Ramesh, A., Brown, K. Y., Juarez, P. D., Tabatabai, M., & Matthews Juarez, P. (2023). Curricular interventions in medical schools: Maximizing community engagement through communities of practice. Annals of Family Medicine, 21(2), 61–67. [Google Scholar] [CrossRef]
  37. Remo Cossu, I., & Awidi, N. (2022). Can we use online technology to rejig the traditional laboratory experience to improve student engagement? Higher Education Pedagogies, 7(1), 1–19. [Google Scholar] [CrossRef]
  38. Rigo, D. Y. (2021). Percepciones de estudiantes sobre compromiso académico y clima del aula: Tendencias pasadas, presentes y futuras sobre una educación superior en transición. ENSAYOS, Revista de la Facultad de Educación de Albacete; Ensayos, 36(2), 77–96. [Google Scholar]
  39. Roque Herrera, Y., Alonso García, S., Criollo Criollo, A. R., Dávalos Villegas, M. X., & Rincón Soto, I. B. (2024). Validación confirmatoria del Cuestionario de Adicción a Redes Sociales en estudiantes de carreras de la salud. RISTI—Revista Ibérica de Sistemas e Tecnologias de Informação, (E70), 603–614. Available online: https://www.risti.xyz/issues/ristie70.pdf (accessed on 5 September 2025).
  40. Rosenshine, B. V., & Berliner, D. C. (1978). Academic engaged time. British Journal of Teacher Education, 4(1), 3–16. [Google Scholar] [CrossRef]
  41. Salari, N., Zarei, H., Hosseinian-Far, A., Rasoulpoor, S., Shohaimi, S., & Mohammadi, M. (2025). The global prevalence of social media addiction among university students: A systematic review and meta-analysis. Journal of Public Health, 33, 223–236. [Google Scholar] [CrossRef]
  42. Sánchez-Sánchez, A. M., Sánchez-Sánchez, F. J., & Ruiz-Muñoz, D. (2024). Factores de riesgo en el uso del teléfono móvil y de las redes sociales en los estudiantes universitarios. Doxa Comunicación. Revista Interdisciplinar De Estudios De Comunicación Y Ciencias Sociales, 38, 19–39. [Google Scholar] [CrossRef]
  43. Shomotova, A., & Ibrahim, A. (2025). Higher education student engagement, leadership potential and self-perceived employability in the United Arab Emirates. Studies in Higher Education, 50(6), 1206–1232. [Google Scholar] [CrossRef]
  44. Shong, P. N., & Jia, Y. F. (2024). A multidimensional view of fear of missing out as a mediator between the need to belong and problematic smartphone use. Computers in Human Behavior Reports, 13, 100352. [Google Scholar] [CrossRef]
  45. Sümen, A., & Evgin, D. (2021). Social media addiction in high school students: A cross-sectional study examining its relationship with sleep quality and psychological problems. Child Indicators Research, 14(6), 2265–2283. [Google Scholar] [CrossRef]
  46. Tanhan, F., Özok, H. İ., Kaya, A., & Yıldırım, Y. (2024). Mediating and moderating effects of cognitive flexibility in the relationship between social media addiction and phubbing. Current Psychology, 43, 192–203. [Google Scholar] [CrossRef]
  47. Torres-Escobar, G. A., & Botero, L. (2021). Factores asociados al compromiso académico en universitarios de carreras virtuales: Revisión de las principales teorías e instrumentos. Panorama, 15(28), 54–76. [Google Scholar] [CrossRef]
  48. Tortosa Martínez, B. M., Pérez-Fuentes, M. d. C., & Jurado, M. d. M. M. (2022). Addressing leadership effectiveness for student academic engagement: A systematic review. School Leadership & Management, 42(4), 366–380. [Google Scholar] [CrossRef]
  49. Trejos Gil, C. A., Tordecilla Pájaro, C. C., & Mena Mena, D. (2023). Adicción a la red social Tik Tok en jóvenes universitarios. Revista de Comunicación y Salud, 14, 1–18. [Google Scholar] [CrossRef]
  50. Tristán Monrroy, B. V., Flores Rueda, I. C., Sánchez Macías, A., & Briano Turrent, G. C. (2021). Student academic engagement in times of COVID-19: Challenges and opportunities for online teaching. Formación Universitaria, 14(6), 193–202. [Google Scholar] [CrossRef]
  51. Trujillo, K., & Noé, M. (2020). La Escala de Procrastinación Académica (EPA): Validez y confiabilidad en una muestra de estudiantes peruanos. Revista de Psicología y Educación, 15(1), 98–107. [Google Scholar] [CrossRef]
  52. Uma, E., Lee, C., Shapiai, S., Binti, A., Soe, H., & Varghese, E. (2020). Academic procrastination and self-efficacy among a group of dental undergraduate students in Malaysia. Journal of Education and Health Promotion, 9, 326. [Google Scholar] [CrossRef]
  53. Valencia Ortiz, R., Cabero Almenara, J., & Garay Ruiz, U. (2021). Adicción a las redes sociales en estudiantes mexicanos: Percepciones de discentes y docentes. Revista Tecnología, Ciencia y Educación, (19), 103–122. [Google Scholar] [CrossRef]
  54. Yana Salluca, M., Adco Valeriano, D. Y., Alanoca Gutiérrez, R., & Casa Coila, M. D. (2022). Adicción a las redes sociales y la procrastinación académica en adolescentes peruanos en tiempos de coronavirus COVID-19. Revista Electrónica Interuniversitaria de Formación del Profesorado, 25(2), 129–143. [Google Scholar] [CrossRef]
  55. Yang, S. Y., Wang, Y. C., Lee, Y. C., Lin, Y. L., Hsieh, P. L., & Lin, P. H. (2022). Does smartphone addiction, social media addiction, and/or internet game addiction affect adolescents’ interpersonal interactions? Healthcare, 10(5), 963. [Google Scholar] [CrossRef] [PubMed]
  56. Zhao, J., Jia, T., Wang, X., Xiao, Y., & Wu, X. (2022). Risk factors associated with social media addiction: An exploratory study. Frontiers in Psychology, 13, 837766. [Google Scholar] [CrossRef] [PubMed]
Education 15 01677 g001
Figure 1. Representation of the ROC curve.
Figure 1. Representation of the ROC curve.
Education 15 01677 g001
Table 1. Summary of the simple linear regression model.
Table 1. Summary of the simple linear regression model.
ModelRR SquaredCorrected R-SquaredTypical Error of the Estimate
10.334 a0.1120.1110.84408
Note: a Predictor variables: (constant), addiction to the use of social networks.
Table 2. ANOVA a analysis of the results corresponding to the simple linear regression model.
Table 2. ANOVA a analysis of the results corresponding to the simple linear regression model.
ModelSum of SquaresglMean SquareFNext.
Regression107,2021107,202150,4660.000 b
Residual853,53511980.712
Total960,7381199
Note: a Dependent variable: overall academic engagement. b Predictor variables: (constant), social media addiction.
Table 3. Coefficients a of the simple linear regression model.
Table 3. Coefficients a of the simple linear regression model.
ModelUnstandardized CoefficientsStandardized CoefficientstSig.Confidence Interval B (95%)Collinearity Statistics
BTypical Error.BetaLower LimitUpper LimitToleranceVIF
Constant58890.097--60,5380.00056986080----
Addiction to the use of social networks−0.0230.002−0.334−12,2660.000−0.027−0.01910001000
Note: a Dependent variable: overall academic engagement.
Table 4. Summary of model b multiple linear regression results.
Table 4. Summary of model b multiple linear regression results.
ModelRR SquaredCorrected R-SquaredTypical Error of the EstimateDurbin-Watson
10.262 a0.0690.0660.835571931
Note: a Predictor variables: (constant), level of excessive social media use, level of obsession with social media, level of lack of control of social media use. b Dependent variable: level of academic engagement.
Table 5. ANOVA results a corresponding to the multiple linear regression model.
Table 5. ANOVA results a corresponding to the multiple linear regression model.
ModelSum of SquaresglMean SquareFSig.
1Regression61,633320,54429,4250.000 b
Residual835,02711960.698
Total896,6591199
Note: a Dependent variable: level of academic engagement. b Predictor variables: (constant), level of excessive social media use, level of obsession with social media, and level of lack of control of social media use.
Table 6. Coefficients a of the multiple linear regression model.
Table 6. Coefficients a of the multiple linear regression model.
Model (Level Dimensions of Social Media Addiction)Unstandardized CoefficientsStandardized CoefficientstSig.Confidence Interval B (95%)Collinearity Statistics
BTypical ErrorBetaLower LimitUpper LimitToleranceVIF
(Constant)49560.136--36,3940.00046895224----
Obsession−0.2070.068−0.119−30390.002−0.340−0.0730.5051981
Lack of control in use−0.1550.068−0.096−22770.023−0.289−0.0210.4422260
Excessive use−0.1390.058−0.088−23860.017−0.253−0.0250.5781730
Note: a Dependent variable: level of academic commitment.
Table 7. Collinearity diagnostics a of the multiple linear regression model.
Table 7. Collinearity diagnostics a of the multiple linear regression model.
Dim.EigenvaluesCondition IndexProportions of Variance in the Level of Dimensions of Addiction to Social Media Use
ConstantObsessionLack of Use ControlExcessive Use
1394410000.000.000.000.00
20.02512,5370.870.030.200.02
30.01914,5500.010.290.050.88
40.01317,5650.110.680.750.10
Note: Dim. (dimension). a Dependent variable: level of academic commitment.
Table 8. ROC curve parameters.
Table 8. ROC curve parameters.
Variables Resulting from the ContrastAreaTypical ErrorSig. (**)95% CI
LowerSuperior
Obsession0.7150.0430.0000.6300.800
Lack of use control0.7360.0460.0000.6460.825
Excessive use0.7200.0450.0000.6320.808
Global addiction0.7260.0450.0000.6380.814
Notes: CI (confidence interval); Sig. (statistical significance); ** (statistical significance at the p = 0.01 level).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Roque Herrera, Y.; Alonso-García, S.; Tenelanda López, D.V.; López Núñez, J.A. Predictive Capacity of Social Media Addiction on Academic Engagement in University Students. Educ. Sci. 2025, 15, 1677. https://doi.org/10.3390/educsci15121677

AMA Style

Roque Herrera Y, Alonso-García S, Tenelanda López DV, López Núñez JA. Predictive Capacity of Social Media Addiction on Academic Engagement in University Students. Education Sciences. 2025; 15(12):1677. https://doi.org/10.3390/educsci15121677

Chicago/Turabian Style

Roque Herrera, Yosbanys, Santiago Alonso-García, Dennys Vladimir Tenelanda López, and Juan Antonio López Núñez. 2025. "Predictive Capacity of Social Media Addiction on Academic Engagement in University Students" Education Sciences 15, no. 12: 1677. https://doi.org/10.3390/educsci15121677

APA Style

Roque Herrera, Y., Alonso-García, S., Tenelanda López, D. V., & López Núñez, J. A. (2025). Predictive Capacity of Social Media Addiction on Academic Engagement in University Students. Education Sciences, 15(12), 1677. https://doi.org/10.3390/educsci15121677

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop