Patterns of Social Network Site Use Among University Students: A Latent Profile Analysis of Academic and Psychosocial Outcomes

Abstract

Social Networking Sites (SNSs) play a central role in university students’ social and academic lives by facilitating relationship maintenance, emotional support, and the exchange of information, especially for those studying away from home. However, it remains unclear how different patterns of SNS use influence academic outcomes and psychosocial well-being. Grounded in social capital and self-determination theory, the present study adopted a person-centered approach using Latent Profile Analysis (LPA) to identify distinct profiles of SNS engagement, academic outcomes and well-being. A sample of 275 Greek undergraduate students completed anonymous self-report questionnaires [SNSs use intensity, bonding and bridging social capital, perceived social support via SNSs, fear of missing out (FoMO), phubbing, nomophobia (NoMo), academic outcomes and well-being]. LPA revealed four user profiles: (1) Low Use-Low Support (poorest well-being, moderate academic outcomes); (2) Active and Supported (high well-being and academic outcomes); (3) At-Risk Heavy Users (intermediate academic outcomes and moderate well-being, comparable to Profile 2) and (4) Low Use-High Support (highest well-being, poorest academic outcomes). These findings indicate that SNS engagement may be associated with both benefits and risks for students, depending on how and why they are used. Adopting a person-centered perspective allowed the identification of meaningful usage patterns, providing critical insights for developing targeted interventions to support student adjustment.

1. Introduction

Social Networking Sites (SNSs) have become central to modern communication and relationship-building and have fundamentally reshaped how people interact [1,2]. The rapid spread of mobile internet has especially impacted emerging adults’ relationships [3], and university students often rely on SNSs to stay connected socially, particularly when they are studying away from home [4]. SNSs can help bridge physical distance by providing avenues for instant communication, peer engagement, and academic information sharing. Nevertheless, it remains unclear how patterns of SNS use (beyond mere frequency) affect academic outcomes and subjective well-being. Findings are mixed [5,6], with some studies highlighting benefits and others pointing to risks. While research has examined specific purposes for using SNSs (e.g., maintaining close ties versus meeting new people, coping with Fear of Missing Out), much of it predominantly employed variable-centered approaches that consider these factors in isolation [7], overlooking how motivations, social resources, and problematic tendencies co-occur.

In this study, we directly address this gap by grounding our work in social capital and self-determination theory and by using a person-centered Latent Profile Analysis (LPA), to capture different combinations of SNS use. Our design intentionally integrates both adaptive indicators (bonding and bridging social capital; SNS-based support) and maladaptive indicators (FoMO, phubbing, nomophobia), as well as students’ academic outcomes (lecture attendance, grades, failed classes) and well-being. Unlike prior variable-centered analyses, this person-centered approach can reveal latent subgroups of students with unique combinations of characteristics, offering insights beyond examining each factor in isolation [8,9]. Given that undergraduates navigate multiple developmental challenges, including relocation to unfamiliar environments [10], understanding diverse SNSs use profiles is critical for developing tailored interventions to promote student adjustment and success.

2. Literature Review

2.1. SNSs, Social Relations, and Online Social Support

SNSs are a key medium for university students seeking social and emotional support, especially while they are far from home [4,11]. By enabling immediate access to friends, family, and broader networks, SNSs can reduce isolation and foster a sense of community [6,12]. Support sources matter; family support fosters stability and unconditional backing, promoting resilience and a sense of belonging [13,14]; friends provide peer-understanding and validation, aiding students in navigating life transitions and shared academic experiences [15]. Support from a significant other (i.e., romantic partner) often involves intimacy and encouragement that contribute to emotional well-being [16]. Yet, reliance on romantic partners via SNSs, particularly in long-distance relationships, can also heighten emotional distress, jealousy, or social comparison [17]. Thus, while SNSs lower many barriers to communication and make it easier to seek support, the quality and authenticity of online social support can vary widely depending on the closeness of the relationship and the context of the interaction.

These dynamics are often interpreted through the lens of social capital theory [18], which views interpersonal networks as resources that individuals can draw upon for support, information, and a sense of belonging. SNS platforms greatly extend the reach and efficiency of these networks by allowing students to both maintain ties with existing friends, family and significant others and to create connections with more distant acquaintances or interest-based communities. Two subtypes of social capital are especially relevant to SNSs use: bonding and bridging social capital [19]. Bonding social capital reinforces strong, pre-existing ties (e.g., close friends and family) and can buffer homesickness or academic stress [4]. However, an exclusive emphasis on bonding ties may limit exposure to new perspectives or external opportunities if one’s network is very insular. In contrast, bridging social capital involves forming newer, weaker ties (such as connections with classmates, faculty, or distant acquaintances) and can broaden access to diverse information, collaboration opportunities, and novel ideas [11]. Bridging ties expand a student’s horizons and adaptability, but they may also lead to more superficial relationships or feelings of social overload [19]. Moreover, maintaining many weak ties requires continuous effort and attention, which can strain cognitive and emotional resources and potentially diminish well-being [19,20]. Thus, both bonding and bridging social capital are valuable for students, but they must be balanced to optimize psychosocial and academic outcomes [21]. Gendered differences also emerge; women tend to use SNSs more for relationship maintenance, implying heavier overall engagement relative to men [22,23]. These considerations motivated our inclusion of bonding, bridging, and SNS-based support, as well as overall SNS intensity, as core indicators feeding into the LPA.

2.2. Problematic SNS Use and Psychological Needs

Despite the many benefits of SNSs, heavy or unregulated use is not universally advantageous. Excessive engagement relates to poorer mental health, including stress, sleep disruption, and diminished face-to-face relationship quality [6,24]. Several problematic use patterns have been identified as contributing to these negative outcomes. One prominent example is Fear of Missing Out (FoMO), defined as a pervasive anxiety that others might be having rewarding experiences from which one is absent. SNS platforms can exacerbate FoMO by providing real-time windows into peers’ activities and achievements [6,25,26]. Students high in FoMO feel uneasy or anxious when they are offline and often compulsively check SNSs to avoid feeling left out of conversations, events, or trends. FoMO has been associated with increased SNS usage frequency and intensity, which can worsen stress and feelings of loneliness over time [26].

From a self-determination theory (STD) perspective, these maladaptive patterns can be understood as manifestations of unmet psychological needs. SDT posits three fundamental needs- autonomy, competence, and relatedness- and it is suggested that individuals may turn to SNSs when these needs are not adequately fulfilled in offline relationships [27,28]. If SNS interactions succeed in satisfying the need for social relatedness (or provide a sense of competence and autonomy through self-expression), then such use might contribute positively to well-being. However, if a student’s usage is largely driven by external pressures or anxieties- as in the case of FoMO, where the motivation to stay constantly connected is fueled by fear- the result may be poorer. Over time, in SDT terms, the individual’s motivation for using SNSs may shift from an autonomous motivation (using SNSs by genuine choice or enjoyment) to a controlled motivation (using SNSs due to compulsion or fear of social sanctions), which is associated with lower well-being and less optimal functioning [28].

Other maladaptive SNS-related behaviors further illustrate these points. In severe cases, FoMO overlaps with nomophobia, or “no mobile phone phobia,” which is the anxiety experienced when one cannot access or use their mobile phone [29]. For instance, a student high in nomophobia might panic if their phone battery is low or if they are in an area with no network coverage [29]. Another related behavior is phubbing (a portmanteau of “phone snubbing”), which refers to ignoring one’s in-person companions in favor of concentrating on a smartphone. Phubbing is known to damage face-to-face interactions; it increases interpersonal conflict and relationship dissatisfaction, and in group settings (like classes or study groups) it undermines cohesion and participation [30,31]. Notably, recent research suggests that these tendencies show limited gender differences, at least among younger adults who have grown up with smartphones [32].

Overall, the quality and motivations of SNS use are more important determinants of outcomes than sheer frequency of use. This aligns with recent integrative findings that underscore the need to consider combinations of factors: for example, the same amount of SNSs use might enhance a student’s well-being if it is driven by social connection and online support, or undermine well-being if it is driven by social anxiety and compulsion [33]. Therefore, we included FoMO, phubbing, and nomophobia to capture maladaptive facets of engagement that theory and evidence link to poorer outcomes, enabling the LPA to differentiate adaptive versus risk-laden profiles.

2.3. Academic Outcomes and Well-Being

Research has documented both constructive and disruptive impacts of SNSs use on students’ academic trajectories. On one hand, when used strategically, SNSs can foster academic engagement and learning, since students can benefit from increased collaboration, knowledge exchange, and peer support [34]. Indicatively, participation in course-related Facebook groups or other online study communities allows students to share resources, discuss assignments, and coordinate study sessions; activities that have been associated with improved academic performance [34,35]. In addition, the social support and sense of belonging that students gain through SNSs can buffer against academic stressors; knowing that one can reach out to friends or mentors online for advice or encouragement may mitigate the negative effects of academic pressure [36].

Conversely, excessive or non-academic SNSs use may displace study time and attention, with intensive use correlating negatively with performance indicators [37,38,39]. One reason is that SNSs introduce constant temptations for multitasking and distraction (checking notifications or scrolling through feeds during lectures or study sessions), which interrupts student focus and impairs the consolidation of learned material [37,38]. Over-reliance on SNSs can also facilitate procrastination; even otherwise high-performing students may find themselves delaying coursework in favor of browsing SNSs [37].

Well-being shows a similarly nuanced pattern. Past research indicates that high SNS use can contribute to enhanced subjective well-being when it strengthens social ties; for example, maintaining close friendships or supportive communities via SNSs can improve mood and provide outlets for emotional expression [6]. Yet, behaviors like phubbing and FoMO have been linked to stress, sleep disturbances, exacerbated symptoms of anxiety or depression and a persistent dissatisfaction with one’s own social life, which can be particularly taxing among students who already have trouble coping with the challenges of transitioning to university life [5,40,41,42,43].

Importantly, maladaptive SNS use can both stem from and exacerbate academic/psychological difficulties. For example, students with social anxiety or depression may retreat to the safer, more controllable space of online interaction, but this may erode offline skills and relationships [44]; likewise, those burdened by academic stress may increasingly turn to SNSs for escapism, yet the time and focus lost to scrolling or gaming further undermine study habits and academic performance, heightening distress [45]; over time, the combination of reduced face-to-face support and escalating online dependence may be linked to greater loneliness, lower well-being, and declining achievement [46,47].

It is also crucial to recognize that academic success does not always go hand-in-hand with personal well-being in university populations. A comprehensive meta-analysis [48] found only a small positive correlation (around r = 0.16) between academic achievement and subjective well-being in student samples. This indicates that excelling academically is neither a guarantee nor a requisite for being happy or mentally healthy. A student could be earning top grades yet experiencing burnout, anxiety, or low overall life satisfaction. Conversely, a student with mediocre grades might report high levels of happiness and mental well-being if they have strong social support, effective coping mechanisms, or a balanced lifestyle [48]. These discrepancies may extend to patterns of SNS use as well. Recent evidence suggests exactly this sort of divergence: students who perform exceptionally well academically but feel unhappy or isolated tend to show different SNSs habits than students with average grades who are content with their lives [49]. These mixed effects justify examining academic outcomes (attendance, grades, failed classes) and well-being as profile-level correlates, testing whether distinct constellations of SNS engagement map onto different academic and psychosocial patterns.

2.4. Aims of the Present Study

Most studies linking SNSs to student outcomes have taken a variable-centered approach, testing one predictor (e.g., time online, FoMO) at a time and estimating a single average effect for the whole sample. Such designs miss the possibility that different subgroups of students combine the same characteristics in very different ways [8]. A person-centered method, Latent Profile Analysis (LPA), can overcome this limitation by clustering students who show similar patterns across several indicators, thereby revealing “hidden” usage profiles and their unique outcome patterns [9].

Our study applies LPA to facets of SNSs use, academic outcomes and subjective well-being. The rationale blends social-capital theory, which views online bonding and bridging ties as potential resources for information and emotional aid [19], with self-determination theory, which predicts that need-satisfying use (e.g., supportive interaction) fosters thriving, whereas need-frustrated, compulsive use (e.g., FoMO-driven checking) undermines functioning [28]. To justify the selection of variables for the LPA, we drew on established theoretical and empirical work. SNSs use intensity captures the overall integration of these platforms into students’ lives [19], while bonding and bridging social capital reflect the two main types of relational resources that SNSs provide [11,19]. Perceived social support via SNSs has been shown to buffer stress and promote adjustment, particularly for students living away from home [4]. At the same time, problematic patterns of use (FoMO, phubbing, and nomophobia) have been consistently linked to negative psychosocial outcomes such as anxiety, loneliness, and impaired academic functioning [6,26,29,30]. Including both adaptive (social capital, online support) and maladaptive (FoMO, phubbing, nomophobia) indicators allows us to capture the dual nature of SNS engagement and to examine how these co-occur within distinct student profiles. This balanced set of variables ensures that the profiles reflect both the supportive and the potentially harmful dimensions of SNSs use, which is essential for assessing their significance for well-being and academic outcomes. We therefore anticipated identifying adaptive profiles characterized by rich online support and goal-directed use, and maladaptive profiles marked by anxious, externally driven behavior.

Specifically, the following Research Questions (RQ) and Hypotheses (H) were formulated;

RQ1. 

Are SNS use, SNS-based support, bridging, bonding, phubbing and FoMO associated with lecture attendance and well-being among university students?

H1. 

SNS use, SNS-based support, bridging and bonding, will correlate positively with higher lecture attendance and greater well-being, while phubbing and FoMO will correlate negatively.

RQ2. 

Do distinct groups defined by gender, living arrangement, mean grades, and number of failed courses differ in terms of SNSs use, SNS-based support, bridging, bonding, phubbing and FoMO and well-being?

H2. 

Participants who are female, reside away from their permanent home, or report lower academic indicators will show more frequent SNS use, higher levels of phubbing and FoMO, and lower well-being.

RQ3. 

What latent profiles of SNSs use, SNS-based support, bridging, bonding, phubbing, FoMO, and nomophobia emerge among university students?

H3. 

Distinct profiles will emerge, differentiated by adaptive (supportive, bonding-bridging focused) versus maladaptive (FoMO, phubbing, nomophobia) SNSs use patterns.

RQ4. 

Are there demographic differences (gender, living arrangement) in profile membership?

H4. 

Female students will be more likely to belong to higher-intensity, support-rich user profiles, whereas male students will be overrepresented in lower-support or less engaged profiles (reflecting known gender differences in SNSs use). Students living away from home may also be inclined toward profiles characterized by greater bonding social capital and online support (due to their reliance on SNSs to maintain distant ties), although we examine this exploratively.

3. Materials and Methods

3.1. Participants

The sample consisted of 275 undergraduate students attending public universities in Greece. Of these participants, 79% identified as women, and most were in the late adolescent age range (19-20 years old; M = 19.4, SD = 0.6). Nearly half of our participants (≈50%) were enrolled in the Department of Primary Education at the University of Ioannina. The remainder were drawn from a variety of fields such as Economics, Mathematics, Chemistry, Psychology, Philology, Nursing, and several other departments across Greek universities. Notably, a large majority (84%) of the students were studying at an institution located away from their permanent hometown or family residence. Participation was voluntary and anonymous.

3.2. Procedure

Data was collected via an online survey, administered after obtaining ethical approval from the university’s research ethics committee. A questionnaire link was distributed to students through faculty mailing lists and digital campus announcements. Participants first read a brief overview of the study’s aims and provided informed consent electronically before proceeding. The survey took approximately 15 min to complete. Respondents were assured of the confidentiality of their answers.

3.3. Measures

3.3.1. Demographic Information

Participants reported basic demographic information, namely age, gender, and whether they were studying in their hometown or living away from home for university.

3.3.2. SNSs Usage Intensity

The Social Media Intensity Scale [19], as modified [50], was used to assess participants’ overall intensity of SNS use. It consists of six Likert-type items (rated from 1 = Strongly Disagree to 5 = Strongly Agree) that measure the extent to which SNSs are integrated into one’s daily life (e.g., “SNS have become part of my daily routine”). Two supplemental questions regarding daily SNSs time and the number of friends or followers were not utilized in the present study. In this sample, the SNS Intensity scale showed good internal consistency (Cronbach’s α = 0.81), and a confirmatory factor analysis (CFA) supported its one-factor structure; χ²(5) = 6.12, p = 0.294, CFI = 0.998, TLI = 0.994, RMSEA = 0.029 [90% CI: 0.000, 0.093], SRMR = 0.022].

3.3.3. Bridging and Bonding Social Capital

Informed by prior research [19], we constructed six Greek-language items to assess bonding and bridging social capital via SNSs: (a) maintaining close ties (e.g., “How often do you use SNS to keep in touch with your family?”) and (b) forming new connections (e.g., “How often do you use SNS to meet new people or acquaintances?”). Participants responded on a 5-point frequency scale (0 = Never, 4 = Very Frequently). A two-factor CFA confirmed the distinction between bonding and bridging uses of SNSs; χ²(8) = 23.50, p = 0.003, CFI = 0.958, TLI = 0.920, RMSEA = 0.084 [90% CI: 0.046, 0.125], SRMR = 0.061, and internal reliabilities were acceptable (Cronbach’s α = 0.66 for bonding, 0.72 for bridging).

3.3.4. SNS-Based Social Support

Perceived social support obtained via SNSs was measured using an adapted version of the Multidimensional Scale of Perceived Social Support [51], a 12-item scale that assesses support from family (e.g., “My family really tries to help me”), friends (e.g., “I have friends with whom I can share my joys and sorrows”), and significant others (e.g., “There is a special person who is around when I am in need”). For this study, the MSPSS instructions were slightly modified to emphasize support received through or facilitated by SNSs communication. Participants rated each item on a 7-point agreement scale (1 = Strongly Disagree, 7 = Strongly Agree). In the present study, the scale showed excellent reliability (Cronbach’s α = 0.92 for Family support, 0.95 for Friend support, 0.92 for Significant Other support), and CFA supported the expected three-factor structure; χ²(51) = 127.80, p < 0.001, CFI = 0.943, TLI = 0.927, RMSEA = 0.116 [90% CI: 0.091, 0.142], SRMR = 0.057.

3.3.5. Fear of Missing out (FoMO)

We assessed FoMO using the 10-item Fear of Missing Out scale [26], a widely used unidimensional scale measuring a person’s anxiety about missing out on rewarding experiences that others might be having (e.g., “I fear others have more rewarding experiences than me”). Participants indicated how true each statement was for them on a 5-point Likert scale (1 = Not at all true of me, 5 = Extremely true of me). In our sample, the FoMO scale exhibited a Cronbach’s α of 0.81, and CFA indicated an acceptable fit for a single-factor model [χ²(32) = 81.46, p < 0.001, CFI = 0.941, RMSEA = 0.075, SRMR = 0.058].

3.3.6. Phubbing

Phubbing (phone snubbing) was measured with the Generic Scale of Phubbing [52], which consists of 15 items and covers four dimensions of phubbing: Nomophobia (e.g., “I feel anxious if my phone is not nearby”), Interpersonal Conflict (e.g., “People complain about my phone use when we are talking”), Self-Isolation (e.g., “I would rather interact with my phone than with the people around me”), and Problem Acknowledgement (e.g., “I realize that my phone use sometimes interferes with conversations”). Participants rated how often each behavior described in the items applied to them on a 7-point scale (1 = Never, 7 = Always). The GSP has been validated as a reliable instrument [52], which was also confirmed in the present study (α ranging from 0.68 to 0.82 and a CFA confirmed 4-factor structure; χ²(80) = 193.71, p < 0.001, CFI = 0.937, TLI = 0.918, RMSEA = 0.072 [90% CI: 0.059, 0.085], SRMR = 0.062.

3.3.7. Academic Achievement

Academic achievement was measured via self-report using three indicators. (a) Grades (GPA) from the most recent university semester was recorded using four ordinal categories reflecting the Greek grading system: 0–4.9 (fail to poor performance), 5–6.49 (adequate), 6.5–8.49 (good), and 8.5–10 (very good to excellent). (b) Number of failed courses (response options: None, 1–2 courses, 3–4 courses, 5 or more courses). (c) Lecture attendance is rated on a 5-point scale from 1 (Never) to 5 (Always).

3.3.8. Well-Being

Students’ subjective well-being was evaluated using the WHO-5 Well-Being Index [53], a widely used, brief, unidimensional measure of general emotional well-being and positive mood states. It includes 5 items (e.g., “I have felt cheerful and in good spirits,”) that respondents rate in terms of how often they felt that way over the past two weeks, on a scale from 0 (At no time) to 5 (All of the time). In this study, the WHO-5 demonstrated good internal reliability (Cronbach’s α = 0.86), and CFA indicated an excellent fit for a one-factor model after allowing one pair of item residuals to correlate [χ²(1) = 0.17, p = 0.680, CFI = 1.000, TLI = 1.013, RMSEA = 0.000] [90% CI: 0.000, 0.119], SRMR = 0.003.

3.4. Data Analysis

All data analyses were conducted in IBM SPSS 27 and R 4.5.0. The analysis proceeded in four stages to address the study’s research questions. After validating every multi-item instrument with Confirmatory Factor Analyses (in R’s lavaan package), we calculated descriptive statistics, Cronbach’s α, and bivariate correlations to address RQ1 and to establish baseline links between each SNS indicator (use intensity, bonding, bridging, perceived online support, FoMO, phubbing) and student outcomes (lecture attendance and well-being). We then examined whether these associations differed across major subgroups, contrasting gender and living-arrangement groups with t-tests and χ² tests, and comparing grades and failed-course categories with one-way ANOVAs, to identify any demographic or academic variability in SNS impacts (RQ2). Shifting to a person-centered perspective, we used the mclust package to run a Latent Profile Analysis that combined overall use intensity, bonding and bridging social capital, SNS-based support, FoMO, phubbing/nomophobia, academic and well-being outcomes (RQ3). Chi-square and one-way ANOVA tests were conducted to examine the associations between latent profile membership and grade category, and between latent profile membership and well-being, respectively. Finally, to address RQ4, we compared the emergent profiles on gender and living arrangement with χ².

4. Results

4.1. Descriptive Statistics

Descriptive statistics (means and standard deviations), as well as correlation coefficients and reliability scores are summarized in

Table 1. Descriptive Statistics, Reliability Scores and Correlations (Cronbach Alpha Scores in Parenthesis).

	Scale	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Attendance	1–5	3.79	10.07	-
2. SNS Intensity Use	1–4	2.76	0.61	0.07	(0.81)
3. Bridging Social Capital	0–4	1.01	0.74	−0.12 *	0.25 **	(0.72)
4. Bonding Social Capital	0–4	2.39	0.93	0.11	0.52 **	0.25 **	(0.66)
5. Phubbing	1–7	2.62	0.92	−0.01	0.51 **	0.30 **	0.33 **	(0.89)
6. NoMo	1–7	3.58	1.38	0.04	0.57 **	0.21 **	0.42 **	0.82 **	(0.82)
7. Interpersonal Conflict	1–7	1.92	1.00	−0.11	0.32 **	0.31 **	0.20 **	0.80 **	0.48 **	(0.81)
8. Self-Isolation	1–7	1.80	0.93	−0.09	0.24 **	0.31 **	0.14 *	0.75 **	0.39 **	0.67 **	(0.81)
9. Problem Recognition	1–7	3.39	1.40	0.09	0.42 **	0.12 *	0.24 **	0.79 **	0.58 **	0.48 **	0.42 **	(0.68)
10. FoMO	1–5	2.70	0.73	0.03	0.45 **	0.22 **	0.33 **	0.56 **	0.50 **	0.45 **	0.33 **	0.47 **	(0.81)
11. Online Social Support	1–7	5.60	1.22	0.06	0.24 **	0.04	0.32 **	0.11	0.19 **	0.05	−0.01	0.10	0.10	(0.93)
12. Online Family Support	1–7	5.48	1.56	0.11	0.16 **	0.06	0.24 **	0.03	0.05	0.02	0.01	−0.01	0.02	0.81 **	(0.92)
13. Online Friend Support	1–7	5.58	1.48	−0.01	0.24 **	0.051	0.30 **	0.15 *	0.23 **	0.08	−0.01	0.12 *	0.12 *	0.85 **	0.50 **	(0.95)
14. Online Significant Other Support	1–7	5.78	1.46	0.06	0.20 **	−0.04	0.25 **	0.12	0.21 **	0.01	−0.05	0.15 *	0.10	0.78 **	0.42 **	0.58 **	(0.92)
15. Well-being	0–5	2.96	0.96	−0.02	−0.03	0.12 *	0.01	−0.10	−0.07	−0.06	−0.08	−0.13 *	−0.17 **	0.28 **	0.24 **	0.28 **	0.15 *	(0.86)

Note. * p < 0.05 ** p < 0.01.

.

In terms of grades, most students (66%) reported an average between 6.5 and 8.49. Fewer (25%) reported an average between 8.5 and 10, while 9% had an average between 5 and 6.49. Regarding failed courses, 38% of students reported having no failed courses, while 29% had failed one or two courses. Additionally, 14% reported failing three or four courses, and 19% indicated failing more than five courses.

Table 1: Significant negative correlations were observed between class attendance and bridging social capital. Well-being showed significant positive correlations with bridging social capital, overall online social support, online family support, online friend support, online significant other support, and negative with problem recognition regarding phone use and FoMO.

4.2. Group Comparisons

Chi-square tests and independent-samples t-tests were conducted to examine gender differences in SNSs usage (intensity, bridging, bonding, online social support), SNS-related problems (FoMO and phubbing) and well-being. Results showed a significant association between gender and number of failed classes, χ²(3, n = 273) = 13.73, p = 0.003, with males having more failed classes than females. Female students reported significantly higher scores than male students on intensity of SNSs usage [t(272) = −3.73, p < 0.001], bonding [t(272) = −3.82, p < 0.001], online significant-other support [t(272) = −2.26, p = 0.025], FoMO [t(272) = −2.29, p = 0.023] and NoMO [t(272) = −2.08, p = 0.038].

Similarly, Chi-square tests and t-tests were used to compare students studying in their hometown versus those studying away. The only statistically significant difference was in bonding social capital, with students studying away reporting significantly higher bonding via SNSs than those studying in their hometown [t(273) = 3.66, p < 0.001].

One-way ANOVAs were conducted to examine differences among grade categories (5–6.49, 6.5–8.49, 8.5–10). A significant difference was found for overall online social support [F(2, 262) = 4.49, p = 0.012]. Post hoc comparisons (LSD) showed that students with grades 6.5–8.49 reported higher overall online social support than those with grades 5–6.49.

Similarly, one-way ANOVAs were conducted for groups with different numbers of failed courses. Significant differences were found for SNSs usage intensity [F(3, 270) = 5.83, p = 0.001], bonding [F(3, 270) = 3.02, p = 0.030] and NoMO [F(3, 270) = 2.77, p = 0.042]. Post hoc comparisons (LSD) showed that students who failed more than five courses reported higher SNSs usage intensity than those who failed none, 1–2, or 3–4 courses. For bonding, students who failed 1–2 courses reported higher scores than those who failed 3–4 or more than five courses. For NoMO, students who failed 1–2 courses reported higher scores than those who failed none or more than five courses.

4.3. Latent Profile Analysis

To address RQ3, we used Latent Profile Analysis (LPA) with standardized indicators of SNSs use intensity, bonding, bridging, online support, and problematic tendencies to uncover distinct user profiles. Competing models with 2 through 5 profiles were estimated, and a four-profile solution provided the best fit to the data. This solution was favored based on model fit indices (e.g., substantially lower BIC compared to the three- or five-profile solutions) and theoretical interpretability, and it achieved satisfactory classification quality (entropy ≈ 0.85). We therefore retained four latent profiles, which together accounted for 265 students included in the LPA.

Table 2. Means and Sample Sizes for Key Study Variables Across Latent Profiles (LPA).

Indicator	Scale	Profile 1 Low Use, Low Support (n = 46)	Profile 2 Active and Supported (n = 113)	Profile 3 At-Risk Heavy Users (n = 34)	Profile 4 Low Use, High Support (n = 72)
Subjective Well-Being	0–5	2.42	2.96	2.88	3.25
Academic Grade	1–4	3.28	3.20	3.12	3.01
Failed Courses 1	0–3	1.20	0.83	1.29	1.38
Class Attendance	1–5	3.96	3.99	3.53	3.64
SNSs Use Intensity	1–5	2.43	3.09	3.20	2.27
Bridging Social Capital	0–4	0.83	1.06	1.59	0.76
Bonding Social Capital	0–4	1.84	2.85	2.93	1.78
Fear of Missing Out (FoMO)	1–5	2.48	2.98	3.42	2.10
Phubbing-NoMo	1–7	2.85	4.19	5.28	2.36
Phubbing-Interpersonal Conflict	1–7	1.52	1.87	3.67	1.38
Phubbing-Self-Isolation	1–7	1.52	1.59	3.60	1.45
Phubbing-Problem Recognition	1–7	2.94	3.73	4.90	2.38
SNS Support-Family	1–7	3.65	5.91	5.58	6.02
SNS Support-Friends	1–7	3.16	6.32	5.48	5.96
SNS Support-Significant Other	1–7	3.95	6.37	5.60	6.18

Note. ¹ Categorical indicator for number of failed courses (0 = none, 1 = 1–2 courses, 2 = 3–4 courses, 3 = more than 5 courses).

displays the mean standardized scores for each profile on all indicators.

Profile 1 (“Low Use, Low Support”) included students with the second lowest SNS intensity and limited bonding and bridging capital, accompanied by the lowest online support. They also showed low FoMO and phubbing, though some phone-use problems were present. This group reported the poorest well-being scores, while academic outcomes were relatively strong, with mid-to-high class attendance and the highest grade distribution.

Profile 2 (“Active and Supported”) was the largest group and stood out for its strong resources: it showed markedly higher bridging and bonding than Profiles 1 and 4, and the strongest online support from friends and significant others. While FoMO and NoMo were moderate, phubbing remained low to moderate. Students in this profile showed the highest class attendance, the lowest number of failed courses, and middle-to-high grades. Their well-being scores were high- second only to Profile 4.

Profile 3 (“At-Risk Heavy Users”) differed most sharply from Profile 2: although both groups had high bonding, Profile 3 reported substantially higher FoMO and the highest phubbing across all profiles. They had the lowest class attendance, intermediate grades, and more failed courses than Profiles 1 and 2. Well-being scores were moderate, and statistically comparable to those of Profile 2.

Finally, Profile 4 (“Low Use, High Support”) was marked by the lowest SNS intensity and the lowest bridging and bonding capital, but the highest online family support and high support from friends and significant others. This group had the highest well-being scores, though their academic outcomes were average to slightly lower compared with other groups (the lowest grades, the most failed courses, and medium class attendance).

A chi-square test of independence was conducted to examine the relationship between grade category and latent profile membership. Results showed a significant association, χ²(9) = 12.64, p = 0.049, indicating that grade distributions varied across profiles. Profile 1 had the highest proportion of students in the top grade category, while Profile 4 had the highest proportion in the lowest. Profiles 2 and 3 had predominantly middle-range grades.

Similarly, an one-way ANOVA showed significant differences in well-being among profiles, F(3, 261) = 7.55, p < 0.001, η² = 0.080. Levene’s test was nonsignificant (p = 0.121), so equal variances were assumed. Post hoc LSD comparisons revealed that Profile 4 (M = 3.25) reported higher well-being than Profile 1 (M = 2.42; p < 0.001) and Profile 2 (M = 2.96; p = 0.041), and marginally higher than Profile 3 (M = 2.88; p = 0.059). Profile 1 had significantly lower well-being than all other profiles (p ≤ 0.028). The difference between Profiles 2 and 3 was nonsignificant (p = 0.666).

4.4. Demographic Differences in Latent Profile Membership

Chi-square tests were conducted to examine whether latent profile membership varied by gender or living arrangement. A significant association emerged between gender and profile membership, χ²(3, n = 264) = 15.52, p = 0.001, indicating that female students were disproportionately represented in Profiles 1 (“Low Use, Low Support”), 2 (“Active and Supported”), and 3 (“At-Risk Heavy Users”), whereas male students were more frequently found in Profile 4 (“Low Use, High Support”). No significant differences were observed in profile membership based on students’ living arrangements.

5. Discussion

The present study adopted a person-centered perspective to examine how university students’ SNS usage patterns relate to their academic outcomes and psychosocial well-being.

Guided by the progressive logic of our analyses, we first examined bivariate correlations to establish a baseline picture of how individual SNS indicators relate to student outcomes. We found that class attendance correlated negatively with bridging social capital, while well-being was positively correlated with bridging social capital and with perceived online social support from family, friends, and significant others. Additionally, phone use problem recognition and FoMO correlated negatively with well-being. These results are partially consistent with H1, confirming that not all SNS connections are equally beneficial. Previous studies have suggested that bridging-oriented SNS ties may be broad but relatively superficial and cognitively demanding, potentially diverting time and energy from academics [19]. In SDT terms, bridging-focused use might not fulfill the need for relatedness deeply, which could help explain why greater bridging capital did not have high correlations with well-being or class attendance [19,28]. Likewise, the correlation of perceiving support via SNSs with well-being, reinforced the view that satisfying students’ relatedness needs through close online interactions can enhance their psychological adjustment [54]. Meanwhile, the negative correlations of FoMO and phone-use problems with well-being are consistent with research showing that maladaptive digital habits undermine mental health by disrupting real-life engagement and amplifying social comparison anxiety [26,55].

Next, we compared major student subgroups to determine whether variables varied by gender, living arrangement, or academic standing (RQ2). In terms of gender differences, female students reported more intensive SNS use, higher bonding social capital, and greater FoMO, NoMo, and online social support (especially from significant others) than male students, whereas male students were more likely to have poorer academic outcomes (e.g., a higher number of failed courses). These results align with prior evidence that women tend to engage more actively with SNSs and may experience more anxiety about disconnection [23,56]. Men’s comparatively weaker academic performance is well documented and may stem from factors such as lower self-regulation or less effective use of support resources [57]. Students studying away from their hometown reported higher bonding, consistent with the idea that those studying away use SNSs to maintain close ties [4]. In terms of academic standing, we found that students with mid-range performance (neither failing nor top of the class) reported the highest levels of SNS-based social support. This suggests that moderately achieving students might be leveraging online networks to cope with academic and emotional challenges, a finding in line with the buffering hypothesis that social support (even online) can mitigate stress [54]. In contrast, students with the lowest academic performance (multiple failed courses) did show higher SNS intensity and greater nomophobia, perhaps indicating a compensatory or stress-driven use of SNSs under academic pressure [58].

While these variable-centered results highlight significant trends, they cannot capture how multiple facets of SNS engagement cluster within individuals. We therefore shifted to a person-centered LPA, which supported H3 by revealing four unique profiles of SNS engagement, highlighting that specific combinations of SNSs use intensity, social capital, online social support and problematic use, co-occur in ways that yield either adaptive or maladaptive outcomes for students.

Profile 2, “Active and Supported,” was characterized by high SNS use, strong online social support, relatively high levels of both bonding and bridging social capital and good overall functioning. These students had the highest class attendance, middle-range grades comparable to Profile 1 and better than Profile 4, and well-being levels similar to Profile 3 but below Profile 4. Importantly, although they were frequent users of SNSs, they showed only moderate FoMO and phubbing/NoMo, suggesting their usage was intentional and well-regulated. Their strong support network appears to help balance their high engagement, reinforcing the view that SNSs can be used adaptively when embedded in supportive social environments [6,11].

Profile 3, “At-Risk Heavy Users,” also had high SNS use but showed a more stress-driven pattern. Although they reported the highest FoMO, NoMo, and phubbing scores, their grades were intermediate and their well-being was similar to Profile 2. In this sense, they seem to be coping relatively well despite problematic phone-use tendencies. However, their heavy connectivity may reflect stress-driven habits rather than socially enriching engagement, consistent with evidence that excessive online engagement can lead to superficial interactions and digital overload [19,21]. This pattern is consistent with the possibility that unmet psychological needs contribute to less optimal SNS use [27].

The other two profiles involved relatively low SNS use coupled with very different online support contexts. Profile 4, “Low Use, High Support,” had the lowest SNS use and the highest online support from family, friends, and significant others. They also had the highest well-being scores of all groups. Paradoxically, they showed the highest proportion of students in the lowest grade category, the most failed courses, and one of the lowest class attendance levels. These findings replicate that academic success does not always coincide with personal well-being [48]. Nevertheless, the minimal SNS engagement of students in Profile 4 might have been only for social and not academic purposes, which could translate into missed opportunities for collaborative learning or resource-sharing online [6,35]. Therefore, although strong online support may protect well-being, it does not necessarily translate to academic success in the absence of engagement or study-related behaviors.

Profile 1 “Low Use, Low Support” further underscores that social support via SNSs is mainly related to well-being and not academic achievement, since it was characterized by minimal SNS activity, the lowest perceived online support, and the lowest well-being, despite having the highest proportion of students in the top grade category and among the highest class attendance. Students in Profile 1 may prioritize independent, task-focused coping strategies to succeed academically [59], but they may lack the emotional buffers provided by online social support [4]. Their very limited SNS use likely keeps digital stress low [60], yet this digital minimalism does not inherently confer psychological benefits in the absence of sufficient support.

Together, these profiles demonstrate that the relationship between SNS use, grades, and well-being is nuanced. Low use can be associated with both high (Profile 1) and low (Profile 4) grades, while high use can also be adaptive or risky depending on the context of social support (Profiles 2 vs. 3). These patterns support the idea that it is not SNSs use itself, but how and with whom it is used, that shapes academic and psychological outcomes. These findings challenge the assumption that avoiding SNSs is not automatically protective: well-being may depend less on “screen time” and more on whether SNS experiences meet or frustrate autonomy/relatedness needs, an individualized pattern highlighted by our person-centered approach [8,28,61].

With respect to RQ4, we observed gender-based differences among latent profiles, supporting H4, but no significant differences based on living arrangement. Specifically, female students were overrepresented in Profiles 1, 2, and 3 (the low-online support group, the active-online support group, and the at-risk heavy users, respectively), whereas males were comparatively few in those profiles. This pattern suggests that female students are disproportionately found at both extremes, among both the most well-supported SNS users and the most at-risk users, consistent with our expectations and with research showing that women’s intensive SNS engagement can expose them to both its benefits and its pitfalls [23]. By contrast, profile distributions did not differ for students living at home versus those living away. Modern digital connectivity might equalize students’ experiences, as even those residing with family can use SNSs to engage with distant friends and university peers; thus, physical distance did not appear to shape how students fell into the usage profiles [13].

6. Practical Implications

The findings of this study carry important implications for how universities and student support services may address well-being and academic engagement in the digital age. Our results suggest that SNS use is not inherently beneficial or harmful; rather, its effects on student outcomes such as well-being and academic performance are nuanced and shaped by the context of social support. For instance, some students with low SNS use achieved high grades but reported poor well-being, while others with high use had average academic outcomes but better psychological functioning. Consistent with the buffering hypothesis [54], students who successfully leverage SNSs to receive emotional support, especially from close family and friends, tend to experience better well-being. In our profiles, the “Low Use, High Support” group (Profile 4) reported the highest well-being despite low academic performance, while the “Low Use, Low Support” group (Profile 1) showed the opposite pattern, strong grades but the lowest well-being. This difference highlights the value of helping students maintain strong supportive relationships online; universities could facilitate this by encouraging students to stay connected with family through SNSs [4,62]. At the same time, students should be encouraged not to rely on a single source of support. Broadening one’s support network, through mentorship programs, student clubs, or peer support initiatives, can provide a more robust safety net [63] and ensure that no one relationship bears all the weight of a student’s emotional needs.

Importantly, efforts to improve student well-being via SNSs should focus on guiding constructive engagement rather than simply limiting screen time. For students like those in Profile 1, who perform well academically but lack support and experience poor well-being, the goal should be to foster SNS experiences that build social connection. Interventions might include introducing moderated online communities, informal class group chats, or low-pressure discussion forums where more reserved students can gradually engage with peers. By framing SNSs as a tool for meaningful connection and collaboration, rather than as a distraction, educators and counselors can help these students use SNSs to fulfill their relatedness needs in academically beneficial ways [11].

Students in Profile 3 (“At-Risk Heavy Users”) reported the highest FoMO and phubbing, which may warrant supportive, not prescriptive, strategies to help manage stress-driven use. Despite these challenges, their academic and well-being outcomes suggest that many students in this group are functioning adequately overall. Prior research indicates that excessive SNS involvement, especially when driven by FoMO or nomophobia, can be linked to poorer well-being and interpersonal difficulties [29,40]. For instance, high FoMO can heighten emotional distress and distraction, while phubbing may undermine relationships and social participation [52]. Even students in the “Active and Supported” group showed moderate levels of FoMO and NoMo, suggesting that digital attachment is common, but may not always interfere with functioning. The key distinction is that in more extreme cases, these tendencies may interfere with daily functioning. To better support students, universities may consider exploratory initiatives such as early screening for signs of problematic SNS use (e.g., administering brief FoMO/NoMo questionnaires during orientation or routine counseling sessions), digital wellness workshops, or resources focused on mindfulness, self-regulation, and intentional engagement [64,65]. The aim of such efforts should be to encourage balanced and purposeful SNS use, rather than to restrict usage outright. Indeed, evidence suggests that strategies fostering mindful control of social media habits yield more sustainable benefits for well-being than blanket restrictions [66].

Taken together, these findings reinforce that the intensity of SNS use is not a reliable indicator of student functioning on its own. Low SNS use may reflect either beneficial digital boundaries or problematic isolation, while high use can offer support or signal emotional overdependence, depending on accompanying factors such as support and coping styles. Therefore, universities and practitioners should avoid uniformly discouraging or vilifying SNS use among students. Instead, the emphasis should be on helping students find a balanced, purpose-driven way to incorporate SNSs into their lives. In designing guidance and policies, it is crucial to consider each student’s context: for example, a high-achieving but isolated student may need encouragement to engage more socially online, whereas an over-connected but academically struggling student may need support in setting boundaries and refocusing on offline responsibilities. Our findings also show that good grades do not guarantee well-being, and that high SNSs use does not always produce poor outcomes. Rather, the presence of supportive online relationships appears to facilitate these effects [48,67]. These differences underscore that it is the quality and purpose of SNSs use, not the sheer quantity, that most strongly influences student outcomes. Universities should thus strive to cultivate an environment in which students are empowered to use social networking in intentional, supportive, and enriching ways that align with their academic goals and psychological needs. In line with this, our findings do not support restricting SNSs use broadly; rather, they highlight the importance of guiding students toward balanced, meaningful engagement. By tailoring interventions and resources to the diverse profiles of SNSs use, institutions can better promote both the educational success and the holistic well-being of their students.

7. Limitations

Despite the contributions of this study, several limitations must be acknowledged. First, the research design was cross-sectional, which prevents any causal conclusions. While we observed associations (for example, between maladaptive SNS use and lower well-being), we cannot determine the direction of effects. It is equally plausible that poor well-being drives some students to engage in problematic SNS behaviors as it is that those behaviors cause declines in well-being. Longitudinal studies are needed to untangle these temporal dynamics.

Second, all measures relied on self-report questionnaires, introducing the possibility of response biases. Participants may have under- or over-reported their SNS habits and well-being due to memory limitations or social desirability. Future studies would benefit from incorporating more objective metrics (e.g., actual usage logs or academic records) or employing mixed-method approaches to validate and complement self-reported data [68].

The majority of respondents were first-year, female undergraduate students in Greece, and nearly half were enrolled in the Department of Primary Education at the University of Ioannina. While the remaining participants came from other fields, including Economics, Mathematics, Chemistry, Psychology, Philology, Nursing, and several additional departments, these groups were represented in much smaller numbers. Thus, the profiles identified may primarily reflect the experiences of education majors and might not capture patterns that prevail in more diverse or specialized academic contexts. Replication with larger and more heterogeneous samples, both within Greece and internationally, is needed to assess whether similar profiles emerge elsewhere. In addition, some latent profiles (notably the “At-Risk Heavy Users”) had relatively few members, which may have reduced statistical power to detect subtle differences.

Finally, we did not measure certain potentially important variables, such as academic use of SNSs, qualitative nature of students’ SNS interactions (e.g., emotional tone or authenticity of online interactions), basic psychological needs (autonomy, competence, relatedness), personality traits or mental health history, that could moderate or further explain the patterns observed [69]. Including such variables in future person-centered research could provide a more complete picture of why students gravitate toward particular SNS use profiles.