Abstract
Purpose
This study examined the association between general self-efficacy and social media addiction among young adult undergraduate nursing students in Indonesia.
Methods
This cross-sectional study included 120 Indonesian undergraduate nursing students aged 21–23 years. Social media addiction and self-efficacy were assessed using the Bergen Social Media Addiction Scale and the General Self-Efficacy Scale, respectively. Data were analyzed using hierarchical multiple linear regression.
Results
The mean social media addiction score was 19.11 (standard deviation=2.11), and the mean self-efficacy score was 24.60 (standard deviation=4.20). Self-efficacy showed a weak but statistically significant negative correlation with social media addiction (r=−.21, p=.022). Sociodemographic predictors entered in Step 1 of the hierarchical regression analysis explained 7.3% of the variance in social media addiction (R2=.073). Sex was a significant predictor (β=−0.19, p=.038), with male students reporting lower scores than female reference group. The final model explained 10.8% of the variance in social media addiction (R2=.108; F(5, 114)=2.77, p=.021).
Conclusion
Higher self-efficacy was weakly associated with lower social media addiction. Given the limited variance explained and the omission of other well-established correlates of social media addiction, these findings do not support general self-efficacy as a standalone protective mechanism.
Keywords: Internet addiction disorder, Mental health, Psychiatric nursing, Self-efficacy, Young adult
INTRODUCTION
In recent years, social media use has become an increasingly integral part of university students’ lives. This is primarily because digital environments facilitate communication, information sharing, and social interaction in both academic and non-academic settings [1]. However, problematic social media use has been associated with academic difficulties, sleep disturbance, emotional dysregulation, and reduced quality of life in young adults [2]. More specifically, the increasing level of social media use among nursing students has become a significant public health concern and an emerging issue in nursing practice settings. Problematic use has been associated with reduced attention during clinical training, impaired academic performance, sleep disturbances, and emotional dysregulation [3]. These behavioral consequences may interfere with clinical judgment, professional concentration, and safe patient care during clinical placements. Therefore, understanding the factors associated with problematic social media use is important for developing preventive and health-promoting strategies in nursing education and practice settings. In this study, the term “problematic social media use” is adopted as the primary construct label. Related expressions such as “social media addiction and excessive social media use” are retained only when quoting instrument names (Bergen Social Media Addiction Scale) or the terminology used in the original cited studies, and should be considered as referring to the same underlying construct. In addition to its academic and clinical implications, problematic social media use during young adulthood also intersects with core developmental tasks, making it a relevant focus for health-promoting and preventive nursing interventions aimed at safeguarding psychological well-being during this transitional life stage.
According to previous studies, young adulthood is a developmental phase characterized by greater independence, identity exploration, and digital engagement. During this phase, young adults experience educational stress, changing social roles, and increasing personal responsibilities. These overlapping academic, social, and personal challenges may increase dependence on the Internet for communication, coping, and social interaction. Therefore, problematic social media use may be common among this demographic. Identifying the factors associated with problematic use is important for developing targeted preventive and health-promoting nursing interventions [4]. In this developmental period, self-efficacy may serve as a psychological resource that supports adaptive coping, adjustment, and well-being. Its association with problematic social media use is hence relevant to both academic functioning and broader preventive and developmental nursing practice.
Previous studies have identified several psychological, social, and environmental correlates of social media addiction [5]. These findings suggest that complex interrelationships between personal and situational factors influence problematic social media use [6]. Self-efficacy, defined as an individual’s belief in their ability to manage actions and challenges, has also been suggested as a potential correlate of digital self-regulation. However, social media behavior is multidimensional and cannot be explained by a single psychological factor [7]. In a different study, Bandura emphasized that self-efficacy beliefs are domain-specific [8]. Accordingly, the General Self-Efficacy Scale (GSES) assesses perceived capability across situations rather than self-efficacy specific to social media use. The construct reflects an individual’s perceived ability to cope with stressful or demanding situations [9] and may represent a broader psychological resource relevant to behavioral regulation [9]. The present results show an association between generalized self-efficacy and social media addiction, not the effect of domain-specific self-efficacy on digital regulation.
Most studies on social media addiction have been conducted among Western or East Asian participants, with only a few involving Indonesian university students. This gap in the literature raises questions about the extent to which previous findings apply across diverse cultural contexts. The sociocultural environment in Indonesia, characterized by strong peer connectedness, collectivist social values, and widespread smartphone use, may influence social media engagement differently compared with other regions. The country’s collectivist orientation is rooted in longstanding sociocultural and religious traditions, including the philosophy of gotong royong (mutual cooperation), strong extended-family and communal structures, and religious and customary (adat) values that emphasize interdependence and communal harmony. These traditions continue to shape interpersonal expectations, socialization patterns, and normative behavior among young Indonesians, including modes of engagement with social media within peer and family networks. Available Indonesian studies have primarily examined the prevalence and socio-ecological correlates of social media use and related disorders among students, with limited attention to generalized self-efficacy as a psychological correlate [1]. The influence of cultural and technological contexts suggests that factors associated with social media addiction may vary across populations. Therefore, a theoretically grounded study within the Indonesian sociocultural context is required to identify relevant psychological correlates [10]. In collectivist societies, social expectations, peer influence, and group norms can shape individual behavior. These influences may also contribute to beliefs about one’s ability to regulate behavior, including digital media use. From a social cognitive perspective, self-efficacy develops through personal experiences as well as social modeling and interpersonal interactions. This shows that strong peer connectedness and collectivist norms may influence young adults’ perceptions of their ability to regulate social media engagement, providing a potential link between sociocultural context and individual self-efficacy beliefs.
Accordingly, the present study addressed the following questions: (1) Do social media addiction scores differ according to sex, age, and educational level? (2) What are the levels of self-efficacy and social media addiction among Indonesian university students? (3) Is self-efficacy correlated with social media addiction? (4) Is self-efficacy independently associated with social media addiction after adjusting for sex, age, and educational level?
METHODS
Ethical statements: This study was approved by the Health Research Ethics Committee of Universitas Muhammadiyah Ponorogo (Protocol No. 643/ER/KEPK/2025) and conducted in accordance with the Declaration of Helsinki. Electronic consent was obtained from all participants aged 20 years or older before enrollment. Participation was voluntary, and anonymity and confidentiality were maintained throughout the study.
1. Study Design
This study used a cross-sectional correlational design to investigate the relationship between self-efficacy and social media addiction in young Indonesian adults. The study was reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guideline for cross-sectional studies [11].
2. Population and Sample Size
Participants were undergraduate nursing students aged 21–23 years recruited from 9 universities in East Java province, Indonesia. Eligibility criteria included current enrollment in a nursing program, active social media use, and the ability to complete an online questionnaire. Students who reported psychiatric or neurological conditions or were unable to complete the questionnaire appropriately were excluded. A total of 120 participants with complete data were included in the final analysis. Sample-size adequacy for the hierarchical multiple regression was evaluated using both predictor-based heuristics and formal power analysis. Conventional rules of thumb recommended the use of 10–15 observations per predictor. Educational level was represented by 2 dummy variables, resulting in 5 predictors in the final model: sex, age, 2 educational-level dummy variables, and self-efficacy. The recommended minimum sample size was 50–75 participants [12], while the final sample comprised 120 participants. A sensitivity analysis was performed using G*Power ver. 3.1 (Heinrich-Heine-Universität Düsseldorf) for linear multiple regression with fixed effects and an R² increase. Using N=120, α=.05, and 1 tested predictor among 5, the analysis provided 80% power to detect an incremental effect of f2=0.07 for self-efficacy beyond the covariates. The observed incremental effect was smaller (f2=0.04), corresponding to an achieved power of approximately .58. The sample size was sufficient to detect incremental effects of approximately f2=0.07 but provided limited power to detect the smaller effect observed in the study.
3. Measurements
1) Social media addiction
Social media addiction was assessed using the Bergen Social Media Addiction Scale (BSMAS) [13]. The instrument comprised 6 items scored on a 5-point Likert scale ranging from 1 (very rarely) to 5 (very often), yielding total scores of 6–30 [14]. Higher scores indicated more problematic social media use. The total BSMAS score was used as a continuous variable in the statistical models. In the original validation study, the scale showed good internal consistency (Cronbach’s α=.88) [13]. A reliability generalization meta-analysis reported a pooled α of .84 (95% confidence interval [CI], .83–.85) across BFAS/BSMAS studies [14]. In the present study, Cronbach’s α for BSMAS was recorded at .84.
2) Self-efficacy
Self-efficacy was assessed using the 10-item GSES developed by Schwarzer and Jerusalem [15]. Each item was scored on a 4-point Likert scale ranging from 1 (not at all true) to 4 (exactly true), yielding total scores of 10–40. This scale measured self-efficacy across a wide range of challenging situations, with higher scores signifying greater perceived capabilities. The GSES was used to assess generalized rather than social media-specific self-efficacy. The original validation showed good internal consistency, with Cronbach’s α ranging from .76 to .90 across samples from 23 nations (most values in the high .80s) [15]. The Indonesian version of the GSES demonstrated sound, unidimensional construct validity among university students [16]. In the present study, GSES showed good internal consistency (Cronbach’s α=.81).
3) Sociodemographic characteristics
Sociodemographic characteristics included sex, age, and educational level, which were treated as covariates in the analysis. A dummy variable was created for sex (0=female, 1=male), and age was entered as a continuous variable. Academic enrollment was classified according to the current educational level of the participants. These covariates were selected on conceptual and empirical grounds rather than arbitrarily. Problematic social media use is multifactorial and has been shown to vary with demographic and academic characteristics in student populations [6,17,18]. Sex, age, and educational level were included as covariates to account for potential demographic confounding. In line with social cognitive theory [19], self-efficacy was designated as the primary personal-level predictor, while the sociodemographic variables served as adjustment factors.
4. Data Collection
The questionnaire was developed using Google Forms (Google LLC) for self-administered data collection. The survey was disseminated through official university communication channels, campus student organization networks, and academic servers. Before participation, prospective participants were shown an information page outlining the study aims and data protection measures, and only those who provided electronic consent could proceed to the questionnaire. Data were collected from February to June 2025, and responses were simultaneously transferred to a secure, password-protected database accessible only to members of the study team.
5. Data Analysis
Statistical analyses were conducted using IBM SPSS ver. 26.0 (IBM Corp.). Descriptive statistics were used to summarize participants’ sociodemographic characteristics and the main study variables. Independent-samples t-tests and one-way analysis of variance were used to examine differences in social media addiction across sociodemographic categories. Pearson’s correlation coefficient was used to assess the association between self-efficacy and social media addiction. The homogeneity-of-variance assumption for the independent-samples t-test was evaluated using Levene’s test (F(1, 118)=0.37, p=.547), indicating that the equal-variance assumption was met.
A hierarchical multiple linear regression analysis was used to explore the predictors of social media addiction. Step 1 included the control variables—sex, age, and educational level—while Step 2 added the pivotal independent variable (self-efficacy). Changes in the proportion of explained variance (ΔR2) were tested to determine whether self-efficacy still contributed independently after controlling for sociodemographic variables. Tests for normality, linearity, homoscedasticity, and multicollinearity were conducted before the regression analysis, and all assumptions were met, with statistical significance set at p<.05.
RESULTS
1. General Characteristics of Participants
A total of 120 respondents were included in the analysis, the majority of whom were women (90.8%), as detailed in Table 1. In terms of educational level, 51.7% were in their third year, 29.2% were in their fourth year, and 19.2% had been enrolled for more than 4 years. Meanwhile, 45%, 27.5%, and 27.5% of participants were 22, 21, and 23 years old, respectively.
Table 1.
Distribution of the demographic characteristics of the participants (N=120)
| Characteristic | Category | No. (%) |
|---|---|---|
| Sex | Female | 109 (90.8) |
| Male | 11 (9.2) | |
| Educational level | Third-year student | 62 (51.7) |
| Fourth-year student | 35 (29.2) | |
| Extended senior student | 23 (19.2) | |
| Age (yr) | 21 | 33 (27.5) |
| 22 | 54 (45.0) | |
| 23 | 33 (27.5) |
Percentages may not total 100 due to rounding.
2. Differences in Social Media Addiction by General Characteristics
No statistically significant differences in total BSMAS scores were observed by age or educational level (both p>.05), whereas a significant difference was observed by sex. Female participants had higher mean±standard deviation (SD) BSMAS scores (19.25±2.09) than males (17.80±1.89), and the difference was significant (t(118)=2.20, p=.029) (Table 2). Mean social media addiction scores did not differ by age group (21-year-olds: 18.40±2.19; 22-year-olds: 19.50±1.98; 23-year-olds: 19.19±2.10) (F(2, 117)=2.90, p=.059). Similarly, educational level did not significantly contribute to the variance in social media addiction (F(2, 117)=1.01, p=.368), and the mean scores were 19.13±2.11, 19.41±2.02, and 18.61±2.21 for third-year, fourth-year, and extended senior students, respectively. Given the pronounced sex imbalance in this sample (90.8% female vs. 9.2% male, n=11), the observed difference should be interpreted with caution. The small male subgroup may generate an unstable and imprecise estimate of the sex difference, so this finding should not be given undue weight until it is replicated in samples with a more balanced sex distribution.
Table 2.
Differences in BSMAS scores according to general characteristics (N=120)
| Characteristic | No. of participants | Mean BSMAS score±SD | t/F | p |
|---|---|---|---|---|
| Sex | 2.20a) | .029 | ||
| Female | 109 | 19.25±2.09 | ||
| Male | 11 | 17.80±1.89 | ||
| Age (yr) | 2.90b) | .059 | ||
| 21 | 33 | 18.40±2.19 | ||
| 22 | 54 | 19.50±1.98 | ||
| 23 | 33 | 19.19±2.10 | ||
| Educational level | 1.01b) | .368 | ||
| Third-year student | 62 | 19.13±2.11 | ||
| Fourth-year student | 35 | 19.41±2.02 | ||
| Extended senior student | 23 | 18.61±2.21 |
BSMAS, Bergen Social Media Addiction Scale; SD, standard deviation.
a)By independent t-test. b)By one-way analysis of variance.
3. Levels of Self-efficacy and Social Media Addiction
The mean±SD social media addiction score was 19.11±2.11 out of a possible range of 6–30. As shown in Table 3, the mean self-efficacy score was 24.60±4.20 out of a possible range of 10–40.
Table 3.
Descriptive statistics of the primary variables (N=120)
| Variable | Scale range | Mean±SD |
|---|---|---|
| Social media addiction | 6–30 | 19.11±2.11 |
| Self-efficacy | 10–40 | 24.60±4.20 |
SD, standard deviation.
4. Correlation between Self-efficacy and Social Media Addiction
Self-efficacy had a weak negative correlation with social media addiction (r=–.21, p=.022), as detailed in Table 4. This indicated that higher self-efficacy was associated with lower social media addiction.
Table 4.
Correlation between self-efficacy and social media addiction (N=120)
| Variable | Self-efficacy (GSES) | Social media addiction (BSMAS) |
|---|---|---|
| Self-efficacy (GSES) | 1 | |
| Social media addiction (BSMAS) | −.21* | 1 |
BSMAS, Bergen Social Media Addiction Scale; GSES, General Self-Efficacy Scale.
*p=.022.
5. Hierarchical Regression Predicting Social Media Addiction
Hierarchical multiple linear regression was used to determine whether self-efficacy was independently associated with social media addiction after adjustment for demographic characteristics, including sex, age, and educational level (Table 5). In Step 1, sex, age, and educational level accounted for 7.3% of the variance (R2=.073), but the overall model did not reach statistical significance (F(4, 115)=2.26, p=.066). Among the control variables, sex was a significant individual predictor (β=–.19, p=.038), indicating that female students reported higher addiction scores after adjusting for age and educational level. Notably, neither age nor educational level was significantly associated with social media addiction.
Table 5.
Hierarchical multiple linear regression predicting social media addiction (N=120)
| Variable | Step 1 | Step 2 | ||||
|---|---|---|---|---|---|---|
| B (95% CI) | β | p | B (95% CI) | β | p | |
| Sex (male vs. femalea)) | −1.373 (−2.670 to −0.076) | −.19 | .038 | −1.296 (−2.576 to −0.016) | −.18 | .047 |
| Age (yr) | 0.365 (−0.140 to 0.870) | .13 | .155 | 0.345 (−0.153 to 0.842) | .12 | .173 |
| Educational level | ||||||
| Fourth-year student vs. third-year studenta) | 0.280 (−0.584 to 1.145) | .06 | .522 | 0.265 (−0.586 to 1.117) | .06 | .538 |
| Extended senior vs. third-year studenta) | −0.545 (−1.543 to 0.453) | −.10 | .281 | −0.515 (−1.499 to 0.469) | −.10 | .302 |
| Self-efficacy (GSES) | - | - | −0.094 (−0.182 to −0.006) | −.19 | .036 | |
| Model summary | ||||||
| R2 | - | .073 | - | .108 | ||
| Adjusted R2 | - | .041 | - | .069 | ||
| ΔR2 | - | - | - | .035 | .036 | |
B, unstandardized regression coefficient; β, standardized regression coefficient; CI, confidence interval; GSES, General Self-Efficacy Scale; R2, coefficient of determination; ΔR2, change in R2 from Step 1 to Step 2.
a)Reference category: Step 1 included sex, age, and educational level (represented by 2 dummy variables, with third-year students as the reference category). Step 2 added self-efficacy (GSES). Sex was coded with female students as the reference group and male students as the comparison group.
In Step 2, self-efficacy was added to the model and proved to be a significant negative predictor of social media addiction (β=–.19, p=.036). Sex also remained a significant predictor (β=–.18, p=.047). Moreover, the addition of self-efficacy increased the proportion of variance explained from 7.3% to 10.8% (R2=.108; ΔR2=.035), with the final model reaching statistical significance (F(5, 114)=2.77, p=.021).
Notably, self-efficacy remained independently associated with social media addiction after adjustment for sex, age, and educational level. However, the model did not include several established psychosocial correlates of problematic social media use, such as impulsivity, fear of missing out (FOMO), depression, anxiety, loneliness, peer norms, and sleep disturbances. The self-efficacy estimate should therefore be interpreted as an adjusted association rather than a unique or comprehensive effect. In the final model, self-efficacy was negatively associated with social media addiction (B=–0.094; 95% CI, –0.182 to –0.006; β=–.19, p=.036).
DISCUSSION
After controlling for sociodemographic factors, self-efficacy was negatively associated with social media addiction (β=−.19, p=.036). However, several established correlates of problematic social media use, including impulsivity, emotional dysregulation, FOMO, depressive symptoms, peer norms, loneliness, and sleep disturbances, were not included in the model. Their omission may have introduced omitted-variable bias and contributed to residual confounding. The results therefore provide tentative evidence of a small negative association between general self-efficacy and social media addiction among Indonesian young adults, but they do not establish a causal or unique effect of self-efficacy.
Beyond statistical significance, the practical magnitude of this association warrants consideration. The standardized coefficient for self-efficacy (β=−.19) and its incremental contribution to explained variance (ΔR2=.035, 3.5%) indicated a small effect by conventional criteria. Moreover, the final model explained only 10.8% of the variance in social media addiction, suggesting that generalized self-efficacy was a modest correlate rather than a standalone factor for reducing problematic use. In nursing practice, self-efficacy may therefore be more appropriately addressed as part of multicomponent, multilevel interventions incorporating educational, social, and environmental strategies rather than as a primary intervention target. Therefore, the 3.5% incremental variance should be interpreted cautiously, particularly given the established correlates omitted from the model.
The mean BSMAS score in the present study was 19.11, corresponding to an average item score of approximately 3.19 on the 5-point response scale. Although estimating the prevalence of social media addiction was not the primary aim, this score suggests that social media-related addictive tendencies were not negligible among the participating nursing students. The Indonesian version of the BSMAS has demonstrated satisfactory psychometric properties among university students, supporting its use in assessing the risk of problematic social media use in this population [20]. For context, a recent study involving nursing students reported a mean BSMAS score of 17.73±5.61, although differences in the populations and study contexts limit direct comparison [21]. The relatively elevated score observed in the present study may therefore warrant attention, particularly given university students’ extensive use of digital platforms for academic, interpersonal, and informational purposes. Nevertheless, the mean BSMAS score should not be interpreted as establishing the prevalence or a clinical diagnosis of social media addiction. Thus, further research using validated diagnostic or risk thresholds in Indonesian nursing student populations is needed.
The relatively small proportion of explained variance observed in this study should also be considered in the context of the study design. The model was designed to examine the independent association of self-efficacy within a limited set of variables rather than to provide a comprehensive explanation of social media addiction. Previous studies have identified multiple determinants of problematic social media use, including impulsivity, emotional regulation, peer influence, social support, and exposure to digital media [17], but these factors were not incorporated into the current model. The association with self-efficacy thus represents 1 component of a broader multifactorial process and should not be interpreted as a comprehensive explanation of problematic social media use.
Another limitation was the possibility of omitted-variable bias. Meta-analyses have shown that problematic social media use tends to be associated with a battery of psychosocial variables such as impulsivity, depressive and anxiety symptoms, emotional dysregulation, FOMO, peer norms, and sleep disturbances [17]. When relevant variables are omitted from a regression model, the coefficient of an included variable may be confounded and not reflect its unique effect. This occurs because the observed association may partly reflect the influence of omitted variables. Consequently, the association between self-efficacy and social media addiction observed in this study (β=−.19) might be an overestimate. Confirmatory studies incorporating a broader set of well-recognized psychosocial predictors are therefore needed to determine the unique contribution of self-efficacy beyond that of stronger and more proximal predictors.
Despite these limitations, the negative association between self-efficacy and social media addiction is consistent with previous reports of a positive association between self-efficacy and self-regulation in digital environments [22]. Previous studies of college samples have also identified a relationship between higher self-efficacy, less problematic Internet use, and better control of online behavior [23]. Together, these results suggest that self-efficacy has a limited association with social media addiction and cannot, on its own, explain problematic digital behavior. The relatively low explanatory power of the model, compared with models incorporating multiple psychosocial predictors, further supports a multifactorial interpretation. From this perspective, social media addiction may reflect the interplay of intrapersonal, interpersonal, and environmental factors [24], meaning that self-efficacy is better viewed as a contributing factor than as the sole determinant of problematic social media use [25,26].
Through a social cognitive lens, self-efficacy refers to people’s belief in their ability to regulate their behavior and manage situational demands [27]. Greater self-efficacy in using social media platforms could enhance confidence in limiting time online and resisting digital distractions [28]. However, the small effect size observed in this study showed that behavioral regulation in digital environments cannot be explained by efficacy beliefs alone [29]. Social cognitive theory posits that personal, behavioral, and environmental factors interact to influence behavior. This interaction helps explain why self-efficacy accounted for only a small proportion of the variance in social media addiction [18]. Thus, the results provide support for theoretical models, suggesting that maladaptive social media use is an outcome of multilevel processes rather than a single macro-level factor [30].
Social and cultural contexts in Indonesia may also help explain the modest explanatory power of self-efficacy for social media use among young adults. Social norms, peer connectedness, collectivist values, and expectations regarding online presence shape digital behavior beyond individual psychological factors. Beliefs about personal capability may therefore be influenced by prevailing social expectations and norms surrounding digital engagement. The modest explanatory power observed in the present study (R2=.108) supports a culturally sensitive perspective when examining psychological predictors of digital behavior in the Indonesian context. These results also have implications for nursing interventions targeting digital health among young adults. Although self-efficacy is insufficient as a primary intervention target, it can be incorporated into broader preventive strategies that promote digital self-regulation, coping, and healthy technology use. In other words, interventions combining educational, social, and environmental strategies are more appropriate than approaches focused solely on individual characteristics. Embedding psychological strategies within multilevel public health interventions offers a more comprehensive approach to addressing problematic social media use.
Despite the key findings presented here, the results should be interpreted with caution due to several limitations concerning the study’s design, measurement, sampling, and analytic scope. (1) In terms of design, the cross-sectional approach precludes causal inference about the relationship between self-efficacy and social media addiction. (2) With respect to measurement, reliance on self-report data may introduce reporting bias. (3) The GSES captures generalized rather than domain-specific self-efficacy, which may underestimate the role of efficacy beliefs in regulating social media use. Future studies should explore whether domain-specific efficacy, alongside psychological and contextual antecedents, generates more adequate explanatory models. (4) Regarding sampling, participants were restricted to Indonesian nursing students, limiting generalizability to other age groups and cultural contexts. The sample also had a pronounced sex imbalance (90.8% female, 9.2% male), so caution is needed when generalizing the results to male and female students or making comparisons between them. (5) Sex-stratified or oversampling strategies could improve precision for male students. (6) In terms of analytical scope, the final model explained 10.8% of the variance in social media addiction (R2=.108) and had approximately 58% power to detect the small incremental effect of self-efficacy (f2=0.04), indicating a need for larger samples in future studies. Furthermore, the model adjusted only for sociodemographic factors and excluded established correlates of problematic social media use, including impulsivity, depression, anxiety, emotional dysregulation, FOMO, peer norms, and sleep disturbances. (7) This omission may introduce omitted-variable bias and limit the interpretation of self-efficacy as an independent predictor. Overall, these limitations suggest potentially opposing sources of bias. The use of a generalized measure may have underestimated the association, whereas the omission of stronger and more proximal correlates may have resulted in overestimation. The net direction of bias therefore cannot be determined, and the observed association should be considered provisional. However, this study provides preliminary, theoretically grounded evidence on generalized self-efficacy and social media addiction among Indonesian university students and supports further study using larger samples, domain-specific measures, and more comprehensive models incorporating relevant psychosocial determinants.
CONCLUSION
Self-efficacy had a weak negative association with social media addiction among Indonesian university students. However, given the limited explanatory power of the model and the exclusion of several established predictors, this result should not be interpreted as evidence of a unique or causal protective effect. Instead, self-efficacy should be considered alongside psychological, social, and behavioral determinants of social media addiction.
Future studies should use longitudinal designs to clarify the temporal relationship between self-efficacy and problematic social media use and incorporate a broader range of psychosocial factors. Examining family, peer, and institutional contexts, alongside mixed-method approaches, could provide a more comprehensive understanding of social media addiction. The resulting evidence could inform the development of evidence-based nursing interventions to promote digital health.
Footnotes
Authors’ contribution
Conceptualization: all authors. Methodology: all authors. Formal analysis: SGM, LH. Data curation: all authors. Writing–original draft: all authors. Writing–review & editing: SGM, NSA. Final approval of published version: all authors.
Conflict of interest
The authors declare no conflict of interest.
Funding
No funding was received for this study.
Data availability
The data that support the results of this study are not publicly available due to ethical restrictions and participant privacy considerations. De-identified data may be obtained from the corresponding author upon reasonable request, subject to approval by the relevant ethics committee and institutional regulations.
Acknowledgements
The authors thank all participants and the institutions involved for their support and cooperation during the study.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this study, the author(s) did not use any generative artificial intelligence (AI) or AI-assisted technologies.
REFERENCES
- 1.Zainita UH, Martha E, Amelia T, Safitri D. Beyond a digital habit: socio-ecological perspectives on social media disorder among Indonesian college students. AIMS Public Health. 2025;12(4):1128–1145. doi: 10.3934/publichealth.2025057. https://doi.org/10.3934/publichealth.2025057 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sun W, Chao M. Exploring the influence of excessive social media use on academic performance through media multitasking and attention problems: a three-dimension usage perspective. Educ Inf Technol. 2024;29(18):23981–24003. doi: 10.1007/s10639-024-12811-y. https://doi.org/10.1007/s10639-024-12811-y . [DOI] [Google Scholar]
- 3.Banodha H, Saini P. Consequences of social media addiction: a triadic perspective on mental, social, and physical health. Int J Indian Psychol. 2025;13(3):4086–4100. doi: 10.25215/1303.373. https://doi.org/10.25215/1303.373 . [DOI] [Google Scholar]
- 4.Ramjan LM, Maneze D, Krstevska E, Pathrose SP, Lewis P. Educational programs to improve nurses’ knowledge, attitude, and practice in adolescent and young adult health: an integrative review. Nurse Educ Today. 2022;118:105510. doi: 10.1016/j.nedt.2022.105510. https://doi.org/10.1016/j.nedt.2022.105510 . [DOI] [PubMed] [Google Scholar]
- 5.Stsiampkouskaya K, Joinson A, Piwek L, Ahlbom C. Emotional responses to likes and comments regulate posting frequency and content change behavior on social media: an experimental study and mediation model. Comput Human Behav. 2021;124:106940. doi: 10.1016/j.chb.2021.106940. https://doi.org/10.1016/j.chb.2021.106940 . [DOI] [Google Scholar]
- 6.Marino C, Gini G, Vieno A, Spada MM. The associations between problematic Facebook use, psychological distress and well-being among adolescents and young adults: a systematic review and meta-analysis. J Affect Disord. 2018;226:274–281. doi: 10.1016/j.jad.2017.10.007. https://doi.org/10.1016/j.jad.2017.10.007 . [DOI] [PubMed] [Google Scholar]
- 7.Steinsbekk S, Wichstrøm L, Stenseng F, Nesi J, Hygen BW, Skalická V. The impact of social media use on appearance self-esteem from childhood to adolescence: a 3-wave community study. Comput Human Behav. 2021;114:106528. doi: 10.1016/j.chb.2020.106528. https://doi.org/10.1016/j.chb.2020.106528 . [DOI] [Google Scholar]
- 8.Bandura A. In: Self-efficacy beliefs of adolescents. Pajares F, Urdan T, editors. Information Age Publishing; 2006. Guide for constructing self-efficacy scales; pp. 307–337. [Google Scholar]
- 9.Luszczynska A, Scholz U, Schwarzer R. The general self-efficacy scale: multicultural validation studies. J Psychol. 2005;139(5):439–457. doi: 10.3200/JRLP.139.5.439-457. https://doi.org/10.3200/JRLP.139.5.439-457 . [DOI] [PubMed] [Google Scholar]
- 10.Chen S, Chen C, Xu H, Wang L, Xie W, Tao W, et al. The interdependent effects of parental, peer, and self factors on children's screen time: a cross-lagged network panel study. Curr Psychol. 2025;44(7):6315–6325. doi: 10.1007/s12144-025-07509-4. https://doi.org/10.1007/s12144-025-07509-4 . [DOI] [Google Scholar]
- 11.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–1457. doi: 10.1016/S0140-6736(07)61602-X. https://doi.org/10.1016/S0140-6736(07)61602-X . [DOI] [PubMed] [Google Scholar]
- 12.Ogundimu EO, Altman DG, Collins GS. Adequate sample size for developing prediction models is not simply related to events per variable. J Clin Epidemiol. 2016;76:175–182. doi: 10.1016/j.jclinepi.2016.02.031. https://doi.org/10.1016/j.jclinepi.2016.02.031 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schou Andreassen C, Billieux J, Griffiths MD, Kuss DJ, Demetrovics Z, Mazzoni E, et al. The relationship between addictive use of social media and video games and symptoms of psychiatric disorders: a large-scale cross-sectional study. Psychol Addict Behav. 2016;30(2):252–262. doi: 10.1037/adb0000160. https://doi.org/10.1037/adb0000160 . [DOI] [PubMed] [Google Scholar]
- 14.Ma JL, Jin Z, Liu C. The Bergen Facebook addiction scale: a reliability generalization meta-analysis. Front Psychol. 2025;15:1444039. doi: 10.3389/fpsyg.2024.1444039. https://doi.org/10.3389/fpsyg.2024.1444039 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Schwarzer R, Jerusalem M. In: Measures in health psychology: a user’s portfolio: causal and control beliefs. Weinman J, Wright S, Johnston M, editors. NFER-NELSON; 1995. Generalized self-efficacy scale measures in health psychology; pp. 35–37. [Google Scholar]
- 16.Novrianto R, Marettih AK, Wahyudi H. Construct validity of the General Self-Efficacy Scale: Indonesian version. J Psikol. 2019;15(1):1–9. doi: 10.24014/jp.v15i1.6943. https://doi.org/10.24014/jp.v15i1.6943 . [DOI] [Google Scholar]
- 17.Cui J, Wang Y, Liu D, Yang H. Depression and stress are associated with latent profiles of problematic social media use among college students. Front Psychiatry. 2023;14:1306152. doi: 10.3389/fpsyt.2023.1306152. https://doi.org/10.3389/fpsyt.2023.1306152 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ma W, Zhen R, Tan X, He Y, Zhang J, Tang S, et al. Self-control and social support in the link between academic pressure anxiety depression and social media addiction in college students. Sci Rep. 2026;16(1):7444. doi: 10.1038/s41598-026-37112-x. https://doi.org/10.1038/s41598-026-37112-x . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yang SY. Effects of self-efficacy and self-control on internet addiction in middle school students: a social cognitive theory-driven focus on the mediating influence of social support. Child Health Nurs Res. 2020;26(3):357–365. doi: 10.4094/chnr.2020.26.3.357. https://doi.org/10.4094/chnr.2020.26.3.357 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pramukti I, Nurmala I, Nadhiroh SR, Tung SE, Gan WY, Siaw YL, et al. Problematic use of internet among Indonesia University students: psychometric evaluation of Bergen social media addiction scale and internet gaming disorder scale-short form. Psychiatry Investig. 2023;20(12):1103–1111. doi: 10.30773/pi.2022.0304. https://doi.org/10.30773/pi.2022.0304 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Berdida DJ. Nursing students' personality traits, sleep quality, social media addiction, and academic performance: a multi-site structural equation model analysis. J Prof Nurs. 2025;56:26–35. doi: 10.1016/j.profnurs.2024.11.004. https://doi.org/10.1016/j.profnurs.2024.11.004 . [DOI] [PubMed] [Google Scholar]
- 22.Berte DZ, Mahamid FA, Affouneh S. Internet addiction and perceived self-efficacy among university students. Int J Ment Health Addiction. 2021;19(1):162–176. doi: 10.1007/s11469-019-00160-8. https://doi.org/10.1007/s11469-019-00160-8 . [DOI] [Google Scholar]
- 23.Fu SC, Pang NT, Wider W. Relationship between internet addiction, personality factors, and emotional distress among adolescents in Malaysia. Children (Basel) 2022;9(12):1883. doi: 10.3390/children9121883. https://doi.org/10.3390/children9121883 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lee AY, Hancock JT. Developing digital resilience: an educational intervention improves elementary students’ response to digital challenges. Comput Educ Open. 2023;5:100144. doi: 10.1016/j.caeo.2023.100144. https://doi.org/10.1016/j.caeo.2023.100144 . [DOI] [Google Scholar]
- 25.Yang YY, Delgado MR. The integration of self-efficacy and response-efficacy in decision making. Sci Rep. 2025;15(1):1789. doi: 10.1038/s41598-025-85577-z. https://doi.org/10.1038/s41598-025-85577-z . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Moningka C. In: Recent advances in digital media impacts on identity, sexuality, and relationships. Wright MF, editor. IGI Global Scientific Publishing; 2020. Self-comparison in the digital world; pp. 231–240. [DOI] [Google Scholar]
- 27.Nyman J, Tornivuori A, Salanterä S, Barroso T, Parisod H. Systematic review of digital interventions to support refusal self-efficacy in child and adolescent health promotion. Health Promot Int. 2022;37(5):daac085. doi: 10.1093/heapro/daac085. https://doi.org/10.1093/heapro/daac085 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Throuvala MA, Griffiths MD, Rennoldson M, Kuss DJ. Mind over matter: testing the efficacy of an online randomized controlled trial to reduce distraction from smartphone use. Int J Environ Res Public Health. 2020;17(13):4842. doi: 10.3390/ijerph17134842. https://doi.org/10.3390/ijerph17134842 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tag B, van Berkel N, Vargo AW, Sarsenbayeva Z, Colasante T, Wadley G, et al. Impact of the global pandemic upon young people’s use of technology for emotion regulation. Comput Hum Behav Rep. 2022;6:100192. doi: 10.1016/j.chbr.2022.100192. https://doi.org/10.1016/j.chbr.2022.100192 . [DOI] [Google Scholar]
- 30.Fereidooni GJ, Ghofranipour F, Zarei F. Interplay of self-care, self-efficacy, and health deviation self-care requisites: a study on type 2 diabetes patients through the lens of Orem’s self-care theory. BMC Prim Care. 2024;25(1):48. doi: 10.1186/s12875-024-02276-w. https://doi.org/10.1186/s12875-024-02276-w . [DOI] [PMC free article] [PubMed] [Google Scholar]
