Mediating Role of Perceived Social Support in the Relationship between Stress and Resilience among Nursing Students in Saudi Arabia

2.1. Research Design

This research utilized a descriptive cross-sectional design.

2.2. Setting

The study was undertaken at the Hail Region College of Nursing at the University of Hail (UoH) during the 2024 second semester. The university is the first public university established in Ha’il, Northern Saudi Arabia. It is a self-contained public university with its campuses dispersed throughout its outreach region. The University of Ha’il, College of Nursing, offers a full-time Bachelor of Science in Nursing (BSN) program. Theory instruction is housed on the main campus, with major affiliated hospitals and healthcare centers in the region serving as facilities for clinical assessments of students.

2.3. Participants and Sampling

The study population consisted of undergraduate nursing students at the University of Hail. Official enrollment and attrition data were obtained from the university registration office, which identified a total undergraduate population of approximately 1,220 students. The BSN program follows a five-year curriculum, comprising four years of theoretical and clinical coursework followed by a mandatory fifth-year internship. Because interns remain fully enrolled students until the completion of this final year, they were included in the sample alongside students in years 2, 3, and 4. This specific cohort was selected because they possessed both academic and clinical experience, providing a necessary bridge between theoretical knowledge and applied clinical data. To be eligible, students required full-time enrollment and active registration for the term. Conversely, students on academic, medical, or study leave, as well as part-time students, were excluded from the study.

Participants were invited to the study through official university communication channels, including student email accounts and the digital learning management system. The invitation included a description of the study’s purpose, an assurance of confidentiality, and a link to the informed consent form. This study used convenience sampling, as the study location was a single site. To achieve this, students who met the required academic level were selected for the sample. The sample size was determined in advance to ensure adequate statistical power. The total population for the study was 1,220, which allowed the sample size to be calculated based on a 95% confidence level and a 5% margin of error, ultimately requiring 169 participants. Also, to account for non-respondents or respondents who skipped survey questions, 200 students were invited; 190 completed the survey (a 95% response rate), exceeding the minimum required sample of 169. Consent was obtained from students, and the questionnaires distributed to them were used to achieve the required sample size for the study

2.4. Instrument

Data collection for this study utilized four primary components: a demographic profile and three established psychological scales. The initial section of the questionnaire collected essential participant demographic information, including age, sex, marital status (Married/Single/ Other), year of study (Second to Fifth Year/Interns), and GPA from the last semester, categorized into specific ranges (e.g., 4.0 - 3.5, 3.49 - 3.0, etc.).

The three primary instruments used to measure the psychological variables were the Multidimensional Scale of Perceived Social Support (MSPSS), the 10-item Connor-Davidson Resilience Scale (CD-RISC-10), and the 10-item Perceived Stress Scale (PSS-10). Perceived social support was measured using the MSPSS [11], a 12-item scale that assesses perceived support from family, friends, and a significant other. Participants responded on a 7-point Likert scale, ranging from 1 (Very Strongly Disagree) to 7 (Very Strongly Agree).

Resilience was assessed using the 10-item CD-RISC-10 [12]. This concise, unidimensional scale employed a 5-point rating (0 = Not true at all, 4 = True nearly all the time), with higher scores indicating greater resilience. Finally, the level of global perceived stress was evaluated using the PSS-10 [13]. The PSS-10 consists of 10 items rated on a 5-point scale from 0 (never) to 4 (very often), measuring the degree to which life situations are appraised as stressful, with positive items being appropriately reverse-scored to ensure higher scores reflect greater perceived stress.

The questionnaires underwent a validation process by four field experts (university professors and researchers/ psychometricians) who unanimously affirmed their content validity. A pilot study was conducted with 20 student nurses to assess the reliability of the measures. The reliability results for the MSPSS were 0.88, CD-RISC-10 was 0.91, and PSS-10 was 0.92 based on the 20 participants.

2.5. Data Collection Process

Data collection commenced only after receiving formal approval from the Ethics Review Board of the University of Hail. To minimize potential power imbalances or undue influence, recruitment was designed to be indirect and voluntary. While Department Heads served as principal collaborators for survey dissemination, their role was strictly administrative. They provided the Google Form link via automated university channels (email and the Learning Management System) rather than through personal solicitation.

To further protect participants, the process was designed to be entirely anonymous. Neither the researchers nor the Department Heads had access to identify which students chose to participate or decline. The invitation explicitly stated that participation was strictly voluntary and would have no impact on academic standing or grades. This ensures that students could complete the survey privately, following their scheduled classes, without any direct observation or pressure from faculty or administration. Data collection was conducted over a one-month period, from September 30 to October 30, 2025, providing students ample time to complete the survey.

2.6. Ethical Consideration

This protocol received approval from the Ethics Review Board of the University of Hail (H-2024-32). Confidentiality, privacy, and the rights of subjects to withdraw at any time without penalty were all addressed in the study. The study was designed and conducted in accordance with the ethical principles for medical research involving human subjects as set forth in the World Medical Association's Declaration of Helsinki.

2.7. Statistical Analysis

Descriptive statistics, including means, standard deviations, and frequencies, were calculated to summarize the sample's demographic characteristics and baseline levels of the primary variables: The PSS-10, CD-RISC-10, and MSPSS. Independent sample t-tests were used to assess mean differences across binary demographic groups (e.g., sex), while One-Way ANOVA was employed for groups with three or more categories (e.g., year level). Additionally, Spearman rank-order correlations were used to assess the strength and direction of relationships between all study variables. All inferential analyses were performed using total scores from the overall scale to ensure statistical accuracy. Lastly, path analysis was conducted to evaluate the hypothesized role of MSPSS as an intervening variable between stress and resilience. Statistical evaluation was conducted according to the criteria for mediation [14], with the indirect effect assessed via the Sobel test.

4.1. Perceived Social Support, Resilience, Stress, and Differences between Demographics

The cohort demonstrated a moderately high level of perceived social support on the MSPSS. This result is generally consistent with many other studies that have used the MSPSS to measure social support across a variety of populations. For example, one study reported an average MSPSS score in a group of older adults in Greece, which is very close to the average score in the present study [15]. Therefore, moderately high social support appears to be a stable and common experience across diverse populations when measured with this instrument. Additionally, researchers assessing medical students in Nigeria reported average item scores, indicating high levels of social support similar to the present findings [15]. The MSPSS has been shown to be a valid and reliable method of assessing perceived social support from family, friends, and significant others, providing a strong rationale for its use in this context [15, 16]. These levels may be indicative of the supportive nature of medical school, where students often complete difficult academic tasks together and form close peer relationships [17]. It has been shown that peer support is a key contributor to the psychological well-being and academic self-concept of medical students, and that a lack of such support can synergistically increase the likelihood of negative academic self-perception [17].

The present study found statistically significantly higher perceived social support levels in female participants than in male participants. This finding is consistent with a large body of research showing gender differences in support. For instance, female inmates were found to report statistically significantly higher levels of social support than male inmates on the MSPSS total score and all three subscales [18]. Similar patterns have been observed in samples of Korean adolescents [19]. However, there is some variability in the literature. One study of Lebanese adults did not find any statistically significant gender differences across any MSPSS dimensions, suggesting that discrepancies may be due to differing societal and cultural expectations regarding gender roles [ 20]. Other research found that while scores were largely equivalent between men and women among survivors of intimate partner violence, men reported statistically significantly greater levels of familial support [21]. The present study's statistically significant (though potentially small) gender difference appears consistent with the specific culture of the medical student population studied.

Statistically significantly higher perceived social support levels were reported by married participants than by unmarried participants. Previous studies support the idea that marital or partnered status is a strong predictor of perceived social support [22]. Furthermore, research has shown that single women report lower global support scores on the MSPSS than women who are romantically involved, which closely mirrors the current results [22]. This is likely due to a spouse or partner being the primary source of emotional support.

There was a statistically significant positive correlation between the level of perceived social support and the year of study. This may be because as students progress, they form more extensive professional and social networks of peers and mentors. Peer social support is a key predictor of academic success and mental health in this population; students with low levels of support are more likely to report academic difficulties [17]. Because the MSPSS can identify students at risk of mental health problems and guide interventions, monitoring these levels over time is highly useful for educational and clinical support [15]. While the present study found a statistically significant correlation, it is worth noting that year of study is only one of many factors contributing to a student's social network quality.

In this current study, participants reported high levels of resilience. Male participants reported statistically significantly higher levels of resilience than female participants. This is consistent with the literature, which suggests that social support and resilience are interdependent constructs [16]. It is possible that males in this study relied more on internal resilience, whereas females relied more on social connections to cope with challenges. Resilience also shows protective effects against stress; lower levels of parental resilience have been associated with higher levels of moderate-to-severe stress [23]. While students in the highest GPA group reported higher levels of descriptive resilience, this difference was not statistically significant. This may be due to the homogeneity of resilience levels in the cohort.

Participants reported moderate levels of perceived stress on the PSS-10. Fourth-year students and those with lower GPAs reported the highest levels of stress, with statistically significant correlations found for both years of study and GPA. This relationship between GPA and stress suggests a reciprocal link between academic performance and well-being [18]. Supportive friends and family have been found to be protective against moderate to severe stress [24]. The moderate stress levels found here, despite high social support, suggest that support mitigates some, but not all, of the stressors inherent in medical education.

The relationships observed are consistent with a theoretical framework, suggesting that social support and resilience are protective against psychological stress [15]. Even in high-stress environments like medical school, high levels of social support and resilience can exist concurrently. The statistically significant differences observed, with females reporting higher social support, males reporting higher resilience, and fourth-year students reporting higher stress, indicate that interventions designed to promote psychological well-being must take into account the unique needs of each demographic subgroup.

4.2. Mediation Analysis of Perceived Social Support on the Relationship between Perceived Stress and Resilience

The aim of this study was to determine if perceived social support acts as a mediator between stress and resilience. The path analysis and Sobel test yielded non-significant results in this study. This finding is further underscored by the final model R² of 0.007, indicating that the combination of perceived stress and social support accounts for less than 1% of the total variance in resilience scores. This finding posits that, despite stress, individuals may experience a sense of social support while recognizing that this may not positively influence their resilience, particularly in fraught situations such as schooling or clinical work. The buffering hypothesis suggests that social support acts as a barrier, protecting individuals and fostering resilience against the negative effects of stress. However, research has shown that social support and stress are related in more complex ways. Some research studies state that the role of social support in stress management is significant, and that social support is more effective when it is provided to a greater rather than a lesser extent [25, 26]. This result stands in opposition to previous studies that viewed the relation of social support to resilience as being more direct and quantitative [27].

Numerous studies emphasize that social support and social connections must be of high quality rather than simply being available. Earlier findings stressed that, for psychological adjustment, an individual needs support derived from quality social interactions [28]. Furthermore, the lack of perception regarding quality social support despite the presence of strong support networks when examining emergency service workers with post-traumatic stress [29]. Research on specific groups, such as cancer patients or emergency service workers, has produced varying results when examining social support and resilience. Researchers have shown that perceived social support does not always lead to increased resilience [30, 31]. Conversely, it was noted that the presence of social support and its impact on resilience among people with substance use disorders [32] highlight the diversity of impact across different demographic categories. Moreover, the value of social support may erode, particularly in high-stakes clinical settings, in the presence of excessive stressors.

Scholars have demonstrated that, although stress levels remain high, caregivers may have social support that does not alleviate their mental health problems [31, 33]. This aligns with the results of the current study, which suggest that during stressful times, merely feeling that social support is available does not necessarily increase resilience. The Sobel test results, which were not significant, suggest that social support is not a simple “bridge” that enhances resilience when stress is high. Instead, it indicates that the MSPSS may have limitations in measuring the complexity of social support in stressful situations. Earlier research suggests that the complexities of personal relationships, the details of social interaction, and the qualitative value of support may require the use of additional, more sophisticated instruments [32]. This study illustrates the need for further work to understand the interplay among social support, stress, and resilience, especially regarding the specifics of the context and high-pressure situations. There are a number of possibilities that may help define the qualitative aspects of support, as well as the interactions between certain coping mechanisms and particular demographic groups, which may further explain the relationships between these key components in the field of psychology.

4.3. Study Implications

The findings of the current study indicate that nursing education needs to move away from a broad “open-door” support policy toward a more specific, operational support model. Given that perceived social support does not seem to play a mediating role in the relationship between stress and resilience, it implies that a student needs more than the feeling of support to recover from clinical or academic trauma. Thus, nursing programs should focus on the provision of clinical support, in the form of direct clinical teaching and structured peer debriefing, rather than general social support. These clinical and educational support strategies may be more beneficial if they are proactive rather than merely responsive; that is, they should be designed to counteract the “support-stress paradox” in which students seek support only when they are already stressed and overwhelmed.

The pivotal place of the “Year of Study” emphasizes that resilience in nursing is experiential rather than a static attribute. This suggests that the nursing curriculum could be restructured to include incremental clinical exposure, with more robust scaffolding for first- and second-year students to mitigate the “attrition of resilience” before they reach high-stakes clinical internships. Furthermore, the specific stressors for married and older students indicate that support must go beyond a “one-size-fits-all” model. To promote professional socialization and optimize academic productivity, the complexities of stress and adult role theory suggest that enhanced stress-resilience training, focusing on self-compassion and clinical self-efficacy, is needed.

4.4. Study Limitations

The primary limitations of this study are rooted in its design. A longitudinal study would be designed to address the temporal causation addressed by this study, but a cross-sectional design does not allow for that. Convenience sampling in only one site of the University of Hail also means the findings are not generalizable to other university cohorts. The sole use of self-report data (MSPSS, CD-RISC-10, PSS-10) means these findings are likely subject to response bias, too. Future research should account for objective data such as physiological measures or GPAs. Use of the Sobel test for mediation also reflects the standard of work in this research. More sophisticated methods exist, such as nonparametric bootstrap methods, which are standard for most work in the field of social science research. These limitations, combined with the likely exclusion of other relevant cohorts of the student population, dilute the findings and their wider relevance.