Study design and sampling

This cross-sectional study included students from three universities located in Maragheh County: Maragheh University of Medical Sciences, Islamic Azad University (Maragheh Branch), and Payame Noor University. Maragheh is a mid-sized urban city in north-western Iran’s East Azerbaijan province with roughly 175,000 inhabitants.

A multi-stage cluster sampling method was employed: each university was considered a cluster, faculties within each university were randomly selected, and students were then proportionally and randomly sampled from each faculty based on enrollment lists. This approach enhanced the representativeness of the sample and minimized selection bias. The minimum required sample size was initially estimated using Cochran’s formula for finite populations, assuming a 95% confidence level, 5% margin of error, and a 50% response distribution, which yielded approximately 384 participants. Given the use of cluster sampling, a design effect (DEFF) was applied; assuming a design effect of about 1.2, the adjusted sample size increased to roughly 460 individuals. This sample size also ensured sufficient statistical power for correlation and multiple regression analyses, in line with standard recommendations of at least 10–20 participants per predictor variable [20,21]. Using this approach allows for the detection of medium-sized effects while maintaining adequate representation across the three universities. A total of 498 eligible students were approached during the recruitment period. Of these, 460 agreed to participate and completed the questionnaires, yielding a response rate of 92.4%. Non-participation was mainly due to a lack of time or unwillingness to complete the full survey. Within each selected faculty, class rosters were obtained from administrative offices, and participants were randomly selected using computer-generated random numbers in proportion to faculty enrollment size. Data collection was conducted in classrooms and student common areas by trained research assistants, who explained the study objectives and obtained written informed consent before questionnaire administration.

Data collection method

Data were collected between 23 September 2023 and 20 February 2024. Data were collected using a four-part structured questionnaire comprising the following components: Demographic Questionnaire: Included items on age, gender, marital status, place of residence, parents’ educational level and occupation, and household income. Psychological Capital Questionnaire (PCQ): Developed by Luthans et al. (2007), (Copyright © 2007 by Fred Luthans, Bruce J. Avolio & James B. Avey. All rights reserved in all media. Published by Mind Garden, Inc., www.mindgarden.com. The use of this instrument was authorized through a research license issued to Arman Latifi by Mind Garden, Inc.), this 24-item scale consists of four subscales: hope, resilience, optimism, and self-efficacy (6 items each) [22]. Each item is rated on a 6-point Likert scale ranging from 1 (“strongly disagree”) to 6 (“strongly agree”), with higher scores indicating stronger psychological capital. The total score ranges from 24 to 144. The Persian version was validated by Bahadori Khosroshahi et al. (2012), reporting a Cronbach’s alpha of 0.85 [23]. Social Capital Questionnaire: Developed by Rafiey et al. (2015), this 20-item instrument measures five subscales: empathy and belonging (6 items), diverse interests (5 items), lifestyle diversity (4 items), trust (3 items), and social participation/cooperation (2 items). Responses are recorded on a 5-point Likert scale (1 = “strongly disagree” to 5 = “strongly agree”), yielding total scores ranging from 20 to 100. Reported internal consistency (Cronbach’s alpha) was 0.82 [24]. social well-being Questionnaire: Originally developed by Keyes (1998), this instrument includes 20 items across five dimensions: social actualization (4 items), social cohesion (3 items), social integration (3 items), social acceptance (5 items), and social contribution (5 items) (6). Each item is rated on a 5-point scale from 1 (“completely disagree”) to 5 (“completely agree”). Total scores range from 20 to 100, with higher scores reflecting better perceived social well-being. Reliability in Iranian samples was confirmed by Babapour (2009; α = 0.78) and Ahari (2012; α = 0.80) (20, 21). All the three instruments were selected based on their wide use in prior research, strong psychometric properties, and availability of validated Persian versions.

Statistical analysis

Data were analyzed using SPSS version 24. Descriptive statistics (means, standard deviations, frequencies, and percentages) were calculated for all study variables. Pearson correlation coefficients were computed to examine associations among the main continuous variables and their subscales. Given the large number of pairwise correlations, the Bonferroni correction was applied to adjust the significance threshold and reduce the likelihood of Type I error; adjusted p-values are reported in the tables. The normality of continuous variables and regression residuals was evaluated using the Kolmogorov–Smirnov test, supplemented by visual inspection of histograms and Q–Q plots. Minor deviations from normality were deemed acceptable given the large sample size (n = 460), consistent with the Central Limit Theorem. Linearity was assessed through scatterplots between each predictor and the outcome variable. Homoscedasticity was examined using residual-versus-predicted value plots. Independence of errors was evaluated using the Durbin–Watson statistic. Multicollinearity was assessed through Variance Inflation Factor (VIF) and Tolerance values, all of which were within acceptable ranges. Prior to regression modeling, diagnostic analyses confirmed that model assumptions were adequately met. Multicollinearity was not problematic, with VIF values ranging from 1.08 to 1.76 and tolerance values ranging from 0.57 to 0.93. The Durbin–Watson statistic was 1.94, indicating independence of residuals. Visual inspection of residual histograms and Q–Q plots suggested approximate normality. Scatterplots of standardized residuals against predicted values indicated acceptable linearity and homoscedasticity without major outliers or influential observations. For comparisons of mean scores across demographic groups (as presented in Table 4), independent-samples t-tests were employed for binary variables (e.g., gender, residence status), and one-way ANOVA was used for variables with more than two categories (e.g., parental education, parental occupation). When ANOVA results were significant, post-hoc analyses with Bonferroni correction were applied. Multiple linear regression analysis was performed with social capital (total score) as the dependent variable. Based on theoretical frameworks and prior evidence, psychological capital and social well-being (total scores) were entered as the primary explanatory variables. Demographic variables (gender, marital status, residence status, and father’s education level) were included as potential confounders. Categorical predictors were dummy-coded, and the following reference categories were defined: male (gender), single (marital status), living with family (residence), and fathers with non-academic education (education level). A two-sided significance level of p < 0.05 was used for all analyses. To ensure adequate power for the regression analysis, an a priori power analysis was conducted using G*Power 3.1. For a model with six predictors, a medium anticipated effect size (f² = 0.15), α = 0.05, and power = 0.80 required a minimum of 98 participants. Additionally, a correlation-based estimate (anticipated effect size r = 0.15, α = 0.05, power = 0.80) suggested a required sample of at least 346 participants. The final sample of 460 students exceeded both thresholds, ensuring sufficient statistical power for the analyses.

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. The difference of social capital, psychological capital, and social well-being according to demographic variables (n = 460).

https://doi.org/10.1371/journal.pone.0353262.t004

Ethics approval and consent to participate

This study was approved by the Research Ethics Committee of Maragheh University of Medical Sciences, Iran (Approval code: IR.MARAGHEHPHC.REC.1397.019). All procedures were conducted in accordance with relevant national research ethics guidelines and the principles of the Declaration of Helsinki (2013, Informed Consent) [25]. Participants were fully informed about the study’s objectives, procedures, confidentiality, and the voluntary nature of participation. All of the participants completed the written informed consent form.

Discussion

The present study explored the relationship between psychological capital, social well-being, and social capital among university students. The results indicated that all components of psychological capital and social well-being were significantly and positively associated with social capital. These findings are consistent with previous research in Iran that has demonstrated similar associations between these constructs [26–28]. The findings support the theoretical assumption that internal psychological resources—such as hope, self-efficacy, resilience, and optimism—play a critical role in fostering students’ external social connections. Among the subcomponents, hope exhibited the strongest correlation with empathy and belonging, a key dimension of social capital. This result aligns with the studies of Khosroshahi (2012) and Azarabadi (2021), which emphasized the reciprocal influence of social support and hope [26,29]. When students feel emotionally supported and understood within their peer networks, they are more likely to maintain a positive outlook and engage more actively in social settings. Similarly, self-efficacy showed a strong positive relationship with overall social capital. Students who believe in their ability to succeed are more confident in initiating and maintaining social relationships, which reinforces their sense of social connectedness. This is in line with Bandura’s theory of self-efficacy [30], and consistent with prior findings in both Iranian and international contexts [17,29]. A potential mechanism underlying these associations is that higher Psychological Capital enables students to actively build and maintain supportive social networks, while greater social well-being enhances trust, reciprocity, and participation within social groups [31]. Cultural context may also shape these relationships; for example, in Iranian university settings, communal norms and family ties may reinforce social engagement, influencing the interplay between psychological resources and Social Capital.

Another important finding was that social coherence, a component of social well-being, had the strongest correlation with both empathy/belonging and overall social capital. This supports Hezarjaribi’s (2012) conclusion that social cohesion fosters resilience and trust within a community [32]. Individuals who perceive their social environment as predictable and meaningful tend to develop stronger interpersonal ties and coping mechanisms. The current findings are further reinforced by international studies. Recent international evidence further supports our findings. For example, Zhao et al. (2022) reported that psychological and social capital jointly contributed to better mental health and well-being among Chinese adults during the COVID-19 pandemic [19]. Similarly, Slåtten et al. (2023) found that psychological capital and supportive social environments significantly predicted academic outcomes among Norwegian university students [18]. More recently, Dong et al. (2025) demonstrated that psychological capital mediated the relationship between interpersonal relationships and mental health among Chinese medical students [17]. Together, these findings support the broader international relevance of the present results. Likewise, Selvaraj and Bhat (2018) demonstrated the predictive value of optimism and resilience for mental health among Indian and American students [33]. These findings, in combination with our own, highlight the cross-cultural relevance of psychological capital and social well-being as foundational resources for enhancing social capital, particularly in academic environments. Such evidence supports the broader applicability of interventions aimed at strengthening psychological strengths as a means of fostering social integration and support. The observed roles of optimism and resilience in this study are also conceptually supported. Optimism fosters trust and expectation of positive social outcomes, while resilience enables individuals to maintain social ties despite adversity [30,34]. These psychological strengths may facilitate students’ social integration and participation. It is important to note that the cross-sectional design limits causal inference; the observed associations do not imply temporal or causal relationships. Reliance on self-reported questionnaires may introduce response bias, and exclusion of individuals with chronic conditions or recent stressful events could underestimate variability in psychological capital and social well-being.

Demographic analysis revealed that marital status, residency (dormitory vs. living with family), and father’s education were significantly associated with social capital. Single students, dormitory residents, and those with academically educated fathers had higher social capital scores. These patterns may reflect greater opportunities for social interaction in communal settings. Similar trends were observed in previous studies conducted by Moradian et al. (2011), Rezaei et al. (2014), Irandoost et al. (2021), and Solhi et al. (2020) [34–38]. However, findings on marital status and parental education were not entirely consistent across studies, which may be attributed to cultural differences, research tools, or contextual variations [35,36,39,40]. No significant associations were found between social capital and age, gender, or income status in this study,results that partially align with or contrast prior findings [35–37,41]. While several Iranian studies have examined similar constructs, this study contributes by simultaneously analyzing psychological capital, social well-being, and Social Capital within a single model, specifically in Maragheh universities. From a theoretical perspective, Bourdieu’s Social Capital Theory helps interpret how individuals’ psychological capacities can shape resources within social networks, and Luthans’ Psychological Capital Theory explains why hope, resilience, and self-efficacy promote social engagement. These frameworks, combined with cultural considerations, provide insight into the observed relationships. The present study extends existing evidence in several important ways. First, although previous studies have examined associations between psychological resources and social outcomes, few studies, particularly in Iran, have simultaneously modeled psychological capital, social well-being, and social capital within a single multivariable framework. This allowed us to estimate the independent contribution of each construct while adjusting for demographic covariates. Second, our findings suggest that psychological capital may be a stronger predictor of social capital than social well-being, as reflected by the larger standardized regression coefficient. This indicates that internal psychological resources such as hope, resilience, and self-efficacy may play a particularly influential role in facilitating trust, participation, and social connectedness among university students. Third, this study contributes evidence from a Middle Eastern cultural context, where collectivist norms, stronger family bonds, and residential arrangements (e.g., dormitory versus family living) may shape the development of social capital differently from Western settings. Together, these findings provide incremental theoretical and contextual evidence for understanding the mechanisms linking psychological and social resources in higher education.

From a theoretical perspective, the present findings reinforce both Bourdieu’s and Luthans’ frameworks. Bourdieu conceptualized social capital as resources embedded within social networks that facilitate social mobility and access to opportunities. Our findings suggest that students with stronger psychological resources may be better positioned to develop and utilize these network-based resources. Likewise, Luthans’ Psychological Capital Theory proposes that hope, resilience, self-efficacy, and optimism function as positive psychological assets that enhance adaptive functioning. The current results support this proposition by showing that these internal resources are positively associated with stronger social connectedness and social capital.

Strengths of the study

This study possesses several methodological strengths. First, it was conducted on a relatively large and diverse sample of university students from multiple institutions, which enhances the representativeness of the findings. Second, the use of validated and culturally adapted Persian instruments for assessing psychological capital, social well-being, and social capital supports the reliability and cultural appropriateness of the measures. Third, the application of multivariate statistical analyses enables a comprehensive examination of the relationships among variables. While certain limitations remain,such as the cross-sectional design, reliance on self-reported data, and cluster sampling without adjustment for design effect,these methodological strengths contribute to the overall rigor and interpretability of the study, providing valuable insights into the interplay between psychological resources, social well-being, and social capital among university students.

Limitation

One key limitation of this study is its geographical scope, as data were collected exclusively from universities located in Maragheh County. Therefore, caution is warranted when generalizing the findings to broader student populations within Iran or to other countries, where sociocultural, economic, and institutional contexts may differ in meaningful ways that affect psychological capital, social well-being, and social capital. Although one participating institution was a medical university, the sample also included students from non-medical universities, which improves the broader relevance of the findings. Therefore, the observed relationships may be partially generalizable to students in other academic disciplines and regions of Iran. However, regional sociocultural differences, institutional environments, and student demographics may influence these associations; thus, generalization should be made cautiously. Moreover, the cross-sectional design inherently restricts the interpretation of the results to associations rather than causal relationships. This design does not allow determination of the temporal ordering among psychological capital, social well-being, and social capital; therefore, no conclusions can be drawn regarding whether these variables influence one another over time or whether reciprocal or reverse causal pathways exist. Although statistically significant associations were identified, future longitudinal or experimental studies are required to clarify the directionality and potential mechanisms underlying these relationships. The reliance on self-report questionnaires represents an additional limitation, as responses may be subject to social desirability, recall bias, or measurement error. Furthermore, students with chronic illnesses or recent major stressful life events were excluded, which may limit the representativeness of the sample and reduce generalizability to the wider student population. While several demographic variables were included in the regression model, other potential confounders or effect modifiers may not have been fully accounted for. Finally, although cluster sampling was used, no adjustment was applied for design effect, which may have resulted in underestimation of standard errors and inflated statistical significance. Despite the appropriateness of Pearson correlations and multiple linear regression for the research objectives, future studies may benefit from employing more advanced analytical approaches, such as mediation analysis or structural equation modeling (SEM),particularly within longitudinal or quasi-experimental frameworks. Such designs would provide deeper insight into the direct and indirect pathways linking psychological capital and social well-being to social capital.

Suggestions

The present findings may have practical implications for university counseling services and student support programs. Counseling centers may consider implementing interventions aimed at strengthening psychological capital, such as resilience training, self-efficacy enhancement, and hope-based counseling. Universities may also promote peer-support initiatives, group mentoring, and community-building programs to improve students’ social well-being and facilitate stronger social capital. In light of the associative nature of the findings, future research is encouraged to employ longitudinal, multi-wave, or intervention-based designs to clarify the directionality and underlying mechanisms of the observed relationships. Additionally, qualitative or mixed-method approaches may provide deeper insights into how students perceive and construct Social Capital within their educational and cultural contexts. Further studies should also examine the influence of demographic factors, such as marital status, residential status, and parental education, across diverse universities and sociocultural settings. Moreover, researchers could consider employing structural equation modeling (SEM) to explore potential mediation, moderation, and interaction effects among psychological capital, social well-being, and social capital. Comparative analyses across fields of study (e.g., medical, humanities, engineering) and public versus private universities could further illuminate contextual differences. Finally, while the current findings suggest potential areas of interest for higher-education policymakers, any practical or programmatic initiatives aimed at enhancing students’ psychological or social well-being should be informed by evidence from studies capable of establishing causal relationships.