Social media have rapidly impacted contemporary society, significantly transforming how people communicate and becoming essential to daily life ( Miñan Olivos et al., 2023). However, when their use is inappropriate, they can generate uncontrollable changes in the management of the time allocated to them, leading to addiction ( Hawi & Samaha, 2017). If they do not remain connected, those navigating these networks may feel like missing out on something momentous ( Trejos-Gil et al., 2023). Furthermore, currently, social media have a massive global presence, with approximately 4.76 billion users ( Shum, 2023). In Peru, statistics corresponding to 2023 reveal that 93.9% of young people between 19 and 24 use Social media more frequently ( Instituto Nacional de Estadística e Informatica (INEI), 2024). Furthermore, as a result of the Covid-19 pandemic, education was directly affected by both teachers and students ( Jaguaco et al., 2022).

Emotional intelligence is fundamental in regulating digital behaviors ( Valencia-Ortiz et al., 2021). It is defined as the inherent capacity of human beings to identify and express their emotions ( Salovey & Mayer, 1990). Likewise, it is related to the correct management of interpersonal relationships ( Idrogo Zamora & Asenjo-Alarcón, 2021). Correct management of emotions allows the individual to increase positive emotions while avoiding negative ones ( Merino-Soto et al., 2019). Furthermore, its lack of development can lead to isolation, which is a characteristic of addiction to Social media ( Arias Cando & Gavilanes Gómez, 2023).

On the other hand, self-esteem constitutes an individual’s concept of himself, covering two fundamental aspects: self-valuation and recognition of particular abilities ( Orth & Robins, 2014). Furthermore, it is a vital factor in the life of a university student, reflected in their academic achievements ( Vargas & Vásquez, 2019). Consequently, self-esteem is primarily relevant online ( Basauri Delgado, 2023). For this reason, addiction to Social media is characterized by excessive and uncontrolled use of these platforms, which can generate harmful effects on the individual, such as social isolation, distortion between reality and the virtual environment, inattention to other activities, and health problems ( Valencia-Ortiz et al., 2021).

Numerous previous studies have addressed emotional intelligence and social media addiction. Studies confirm an inverse relationship between emotional intelligence and social media addiction. Individuals with low emotional intelligence tend to have higher levels of social media addiction ( Saraiva et al., 2018), and this is exacerbated in people with some type of depressive disorder ( İnaltekin & Yağcı, 2024). Studies consistently show a link between lower emotional intelligence and greater problematic social media use ( Piccerillo & Digennaro, 2024). Emotional intelligence helps improve emotional and social skills and acts as a protector against virtual addictions ( Fernández-Martínez et al., 2023). Studies on 505 female participants reported that excessive online social media use showed an inverse relationship with emotional intelligence ( Tyagi & Meena, 2022). Another study in 466 adults reported that the dimensions of emotional intelligence, emotional management, emotional awareness, and relationship management are affected by problematic social media use ( Kahwagi et al., 2022). A study in 400 participants between the ages of 18 and 25 found that younger people with lower emotional intelligence are more likely to develop addictive behaviors of social media use ( Jarrar et al., 2022).

On the other hand, studies also report an inverse relationship between self-esteem and social media addiction. Research confirms that social media addiction negatively predicts self-esteem levels among university students ( Oleas Rodríguez & López-Barranco Pardo, 2024). Social media addiction is associated with lower self-esteem based on social comparison stigmas, body image issues, or emotional problems ( Colak et al., 2023). Likewise, self-esteem also acts as a mediating factor between social media addiction and other constructs such as depression or anxiety ( Hou et al., 2019). Reports on 311 university students confirmed the effect of social comparison on the relationship between social media and self-esteem, which can be negative or positive ( Kavaklı & Ünal, 2021).

Furthermore, in Peru, negative relationships were identified between addiction to Social media and emotional intelligence in a sample of 279 students ( Alarcon & Salas, 2022). Likewise, Akkus ( Akkus, 2021) determined the existence of a negative and significant correlation between self-esteem and addiction to Social media in a population of 246 students. Bányai et al. ( Bányai et al., 2017) discovered a negative correlation between self-esteem and problematic Social media use in 489 Hungarian participants.

On the other hand, the research by Saraiva et al. ( Saraiva et al., 2018) established a significant negative correlation between emotional intelligence and Social media addiction. Similarly, the study by Rivera and Araujo ( Rivera-Véliz & Araujo-Robles, 2021) found a low, although not significant, correlation between emotional intelligence and Social media addiction variables. Consistent with these findings, Sundvik and Davis ( Sundvik & Davis, 2023) determined a low correlation between Social media addiction and emotional intelligence in their research.

This study is relevant because it aims to show how emotional intelligence and self-esteem predict social media addiction in Peruvian students. In Peru, due to the particularities of the educational system, which has been constantly evolving since 2013 ( Saavedra & Gutierrez, 2020), with a large percentage of rural and indigenous education ( Morales Rocha et al., 2024), and with educational inequalities ( Pasquier-Doumer, 2004), it is important to understand how emotional intelligence and self-esteem predict social media addiction in university students.

Table 2 shows that all study variables present univariate normality since the coefficients of asymmetry and kurtosis are within the range ± 2.

Table 3 shows the correlations between the study variables, highlighting the relationships between the dimensions of emotional intelligence, self-esteem, and indicators of social media addiction. A significant positive correlation is observed between self-esteem (AUTO) and the dimensions of emotional intelligence (SEA, OEA, UEO, and ROE), with coefficients between .50 and .53. This indicates that people with higher self-esteem tend to better assess and manage their emotions, both their own and those of others, as well as have a greater capacity for emotional regulation.

On the other hand, variables related to social media addiction (OBS, FAL, and USE) present negative correlations with self-esteem and emotional intelligence. Specifically, social media obsession (OBS) shows negative correlations with SELF (-.34) and with all dimensions of emotional intelligence, with values between -.27 and -.30. This suggests that people with a greater obsession with social media tend to have lower self-esteem and a lower capacity for emotional management. Similarly, lack of personal control over social media use (FAL) also has negative correlations with self-esteem (-.30) and with the dimensions of emotional intelligence, with values between -.29 and -.30. This implies that difficulty controlling the time and frequency of social media use is associated with lower emotional appraisal and regulation.

Table 4 shows that all emotional intelligence dimensions (SEA, OAS, UEO, and ROE) exhibit adequate reliability values. Cronbach’s alpha ranges from 0.708 to 0.8, indicating good internal consistency. Furthermore, the composite reliability (CR) values are higher than 0.8, and the average variance extracted (AVE) is above 0.5, suggesting that the items in each scale adequately explain the variance of their construct.

Regarding self-esteem (SELF), Cronbach’s alpha is 0.811, and the composite reliability reaches 0.852, indicating high reliability. However, the AVE value (0.387) is relatively low, which could suggest that some items do not optimally explain the construct’s variance.

Scales related to social media addiction show even higher levels of reliability. Social media obsession (SMB) presents a Cronbach’s alpha of 0.911, with a composite reliability of 0.936 and an AVE of 0.631, reflecting excellent internal consistency. Similarly, lack of personal control over social media use (LAL) and excessive social media use (USO) have high reliability values, with alphas of 0.859 and 0.938, respectively. The composite reliability for these dimensions exceeds 0.89, and the AVE is greater than 0.5, indicating that the items accurately measure the constructs of social media addiction.

Table 5 presents the results of the linear regressions assessing the impact of self-esteem (SELF) on the dimensions of emotional intelligence. Self-esteem (SELF) is the strongest predictor in all models. This indicates that self-esteem has a positive and significant relationship with all dimensions of emotional intelligence, suggesting that people with higher self-esteem tend to better assess and manage their emotions.

Regarding the R ² values, it is observed that self-esteem and the emotional intelligence dimensions explain between 31.3% and 34% of the variance in each emotional intelligence dimension. This suggests that a considerable proportion of the variability in these emotional skills can be attributed to self-esteem.

The effect size (f ²) for self-esteem varies across models, with the highest effect size for appraisal of others’ emotions (AEO) at 0.065 and the lowest effect size for emotion regulation (ER) at 0.016. This suggests that self-esteem has a particularly strong impact on the perception of others’ emotions, while its influence on emotion regulation is weaker.

Table 6 shows the results of the linear regressions, which analyze the effect of self-esteem (AUTO) and the dimensions of emotional intelligence (SEA, OEA, UEO, and ROE). First, self-esteem (AUTO) is the strongest predictor in all models, with beta coefficients of -0.293 in OBS, -0.225 in FAL, and -0.153 in USO. This indicates that higher self-esteem is associated with lower levels of social media obsession, lack of control, and excessive use. Furthermore, the f ² values for self-esteem are the highest in each model, reinforcing its importance in explaining social media addiction.

Regarding the dimensions of emotional intelligence, it is observed that the assessment of one’s own emotions (SEA) and the use of one’s own emotions (UEO) have weak negative effects on social media addiction, with beta coefficients ranging from -0.072 to -0.136. This suggests that better emotional management is related to a lower level of addiction, although its impact is smaller compared to self-esteem.

On the other hand, appraisal of others’ emotions (OEA) and emotion regulation (ER) did not show significant effects on social media addiction, with beta coefficients close to zero and very low effect sizes (f ² ≈ 0.000). This indicates that these dimensions of emotional intelligence do not significantly predict the tendency toward social media addiction.

In terms of explained variance (R ²), the models explain 17.2% of the variance in OBS, 14.8% in FAL, and 9.2% in USO. These values suggest that, while self-esteem and emotional intelligence influence social media addiction, other factors may also play an important role.

Figure 1 presents a structural equation model (PLS-SEM) that analyzes the relationship between self-esteem, emotional intelligence, and social media addiction. It highlights that self-esteem is the strongest predictor, reducing obsession, lack of control, and excessive social media use. Some dimensions of emotional intelligence also play a role, but with a lesser impact. Furthermore, the model includes factor loadings, which reflect how well the items measure each variable. These loadings are expected to be high (≥0.7), indicating good validity and reliability of the model.

Emotional intelligence and self-esteem are crucial factors for regulating and controlling the use of Social media. They provide people with skills and competencies that allow them to become aware of their behavior in these digital environments ( Garay-Argandona et al., 2021). Veloza ( Veloza Gamba, 2023) maintains that excessive use of Social media negatively affects the development of emotional intelligence and can distort its proper evolution. Martínez et al. ( Martínez-Líbano et al., 2022) report that using Social media negatively affects the health status of university students. For this reason, it is of great relevance to determine whether emotional intelligence and self-esteem are predictors of addiction to Social media.

The findings confirm that self-esteem has a positive relationship with the dimensions of emotional intelligence, meaning that people with higher self-esteem tend to have better assessment, evaluation, and regulation of their emotions. This result is in line with Mayer et al. ( Mayer et al., 2004), where a high degree of emotional attention can negatively affect self-esteem. Good emotional intelligence is associated with better self-esteem ( Calero et al., 2018). Likewise, emotional recovery is related to self-esteem, which ultimately contributes to maintaining a positive image ( Schutte et al., 2002). This relationship has been reported in university students ( Murad, 2021).

On the other hand, self-esteem and emotional intelligence dimensions predict variability in social media addiction. The literature shows that social media addiction decreases self-esteem in university students ( Landa-Blanco et al., 2024). People with low self-esteem are likely to use social media in an extreme way to increase their self-image and self-esteem ( Przepiorka et al., 2024). Higher self-esteem is associated with lower levels of social media addiction ( Mayorca-Martínez et al., 2023). In the case of emotional intelligence, it is inversely related to social media addiction ( Fernández-Martínez et al., 2023). Furthermore, studies show that emotional intelligence can mitigate the relationship between internet addiction and mental health ( Zewude et al., 2024).

The findings confirm that self-esteem has a predictive effect on levels of social media obsession. Theory confirms that low self-esteem can increase people’s vulnerability to developing addictive behaviors online ( Andreassen, 2015). Furthermore, these exposures to virtual content induce negative comparisons that decrease self-esteem and promote negative behaviors in the use of social media ( Vogel et al., 2014). The results confirm the inverse effect of self-esteem on the lack of control in social media use. Studies show that students who have levels of control tend to have higher self-esteem ( Saadat et al., 2012). Likewise, the findings confirm that self-esteem negatively predicts excessive social media use. The literature confirms that higher self-esteem is associated with less addiction ( Rahmawati et al., 2021).

Another finding, albeit with a lesser effect, is that of emotional intelligence in its dimensions of self-assessment of emotions and appropriate use of emotions on social media addiction. College students with low emotional intelligence are prone to problematic social media use ( Kircaburun et al., 2020). Students with poor emotion management skills tend to rely excessively on social platforms ( Hormes et al., 2014).

This study has relevant practical implications, such as the possibility of developing educational programs to control addiction to Social media. Given the constant growth in the use of these platforms, this research constitutes a significant contribution to the existing literature. Social media represent a crucial communication and socialization environment for adolescents. Depending on their use, they can strengthen their ties or contribute to the emergence of specific problems. Given the need to promote responsible and constructive use of these tools,

This study’s limitations could have influenced the results. Furthermore, since this is a self-report, some university students could have responded more honestly due to the time factor. Likewise, the information was acquired virtually, so it could be different when doing it in person. Therefore, conducting more research with representative samples of Peruvian students and other countries is recommended.

We conclude that emotional intelligence and self-esteem are significant constructs to increase or decrease addiction to Social media. According to the results of our research, a moderate negative correlation is observed between Emotional Intelligence and Addiction to Social media. This suggests that people with higher levels of emotional intelligence show less tendency toward Social media addiction. On the other hand, a weak positive correlation is found between Self-esteem and Addiction to Social media. This suggests that people with higher self-esteem may have a slight tendency to show greater addiction to Social media.

This research was approved by the Ethics Committee of the Graduate School, Universidad Peruana Unión on December 11, 2023 (Reference: CE-EPG-000160). It was carried out without specific funding from public, commercial or non-profit organizations. The guidelines for research involving human subjects found in the Declaration of Helsinki will apply. Participants were informed that their participation in this non-experimental research was completely voluntary and that they were free to withdraw at any time without facing sanctions or negative repercussions. Additionally, participants were assured that there were no known risks associated with their participation in the study.

Written Informed consent was obtained from all participants prior to their participation in the study. The full statement for larger samples according to the ethics committee protocol was: The purpose of this questionnaire is to identify factors associated with social media addiction. Your participation is entirely voluntary, and you are not required to complete this survey if you do not wish to. If you decide to participate in this study, please complete the questionnaire. You may also stop completing it at any time if you so choose. I have read the preceding paragraphs and acknowledge that by completing and submitting this questionnaire, I am giving my consent to participate in this study.