To determine the levels of stress among nursing students and to examine the associations among demographic characteristics.

This study was guided by Lazarus and Folkman’s Transactional Model of Stress and Coping (1984), which conceptualises stress as a dynamic process arising when individuals perceive that the demands of a situation exceed their available coping resources. Stress is determined not solely by external demands but by the interaction between individuals and their environments (Kromoser 2025; Labrague 2024). This includes their cognitive appraisal of the situation and their available coping strategies. Primary appraisal involves evaluating whether a situation is threatening, challenging or benign; secondary appraisal assesses perceived ability to manage the stressor (Lazarus & Folkman 1984). Coping strategies are then employed to mediate stress, either through problem-focused methods, addressing the source of stress or emotion-focused approaches, regulating emotional responses.

In the context of nursing education, the Transactional Model is relevant, as students at different levels of training face distinct stressors (Lazarus & Folkman 1984). First-year students primarily contend with basic clinical skills and adjustment to the academic environment, reflected in physical and academic stressors (Langtree et al. 2018; Zondi 2018). Second-year students experience heightened academic stressors because of more complex theoretical modules, such as physiology, and the introduction of community-based clinical placements (Mhlongo & Masango 2020; Mudaly 2023). Third- and fourth-year students face advanced clinical responsibilities, preparation for professional integration and cumulative academic pressures, which may increase exposure to environmental and interpersonal stressors.

The Student Stress Inventory (SSI) aligns with this multidimensional view of stress, assessing four interrelated domains (Arip et al. 2020). These domains include physical, academic, interpersonal and environmental stressors, capturing the broad spectrum of stress experiences across all levels of nursing students (Arip et al. 2020). Notably, physical stressors encompass symptoms such as fatigue, headaches and sleep disturbances, reflecting the physiological manifestation of perceived academic and clinical demands (Arip et al. 2020). Academic stressors involve challenges in mastering lessons, meeting deadlines and managing study-related pressures (Arip et al. 2020). Interpersonal stressors include relational challenges with family, peers and faculty, while environmental stressors reflect contextual difficulties such as transportation, campus facilities and residential conditions (Arip et al. 2020). By mapping these sub-scales onto the Transactional Model of Stress and Coping, the study interprets stress as arising from the interaction of perceived demands and coping resources, highlighting why students at different levels may experience different types of stress of varying intensity (Lazarus & Folkman 1984).

Overall, the integration of the Transactional Model with the SSI provides a theoretical lens to understand the multidimensional and cumulative nature of stress in nursing education (Arip et al. 2020; Lazarus & Folkman 1984). It underscores the importance of interventions that target both the demands of the educational environment and the enhancement of students’ coping resources to promote resilience, well-being and academic success across all levels of training.

A quantitative, descriptive, cross-sectional study was conducted among students at a selected NEI in Limpopo province, South Africa (Botma et al. 2022). A cross-sectional design is appropriate for this study as it allows the determination of stress levels and the examination of associations between stress and selected demographic characteristics at a single point in time, providing a snapshot of the prevalence and distribution of stress among nursing students. While this design permits the identification of statistical associations, it does not allow for causal inference. Given that perceived stress is inherently subjective, a self-report approach was appropriate for capturing students’ experiences. The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (see Appendix) (Von Elm et al. 2014).

The study was conducted at a selected NEI in the Limpopo province of South Africa. The NEI is situated in the Capricorn District, approximately 32 km east of Polokwane, the provincial capital. It is classified among the eight rural and historically disadvantaged institutions in the country and is accredited by the South African Nursing Council (SANC) to train learner nurses under Regulation 174 (R174). The institution offers both theoretical instruction in classrooms and work-integrated learning at accredited clinical practice sites. The study population comprised all undergraduate students enrolled in the Bachelor of Nursing (R174) programme during the 2025 academic year, with a total of 320 students registered across all four levels of study. All registered students who provided informed consent were eligible for inclusion. Students who were not formally registered, those on academic leave of absence or those who declined or withdrew consent were excluded.

The study included all undergraduate students registered in the Bachelor of Nursing (R174) programme during the 2025 academic year, totalling 320 students across the four levels of study. Stratified random sampling was employed to select participants proportionally from each level, ensuring representativeness and minimising sampling bias. The sample size for each level was determined using the Raosoft online calculator, applying a 95% confidence level, a 5% margin of error, and an assumed response distribution of 50%, with a 10% adjustment added to account for potential non-response or incomplete questionnaires. A total of 262 students participated in the study, with each level achieving a high proportional representation of the students selected. Stress was assessed using the SSI, a validated self-report instrument, which captures students’ perceived stress levels at a single point in time.

Data was collected from 18 August 2025 to 05 September 2025. Recruitment was conducted through the distribution of leaflets that outlined the study’s objectives, eligibility criteria and essential contact information. The leaflets invited students to take part voluntarily. Permission to use the SSI was obtained from Arip et al. (2020). For data collection, the SSI questionnaire was uploaded as a Google Form for convenient access, and the link was shared with students via their respective WhatsApp groups. The first page of the form contained an informed consent statement, requiring participants to indicate their consent before accessing the questionnaire. Although Google Forms offers an option to require completion of all items before submission, this feature was not enabled. Given the sensitive nature of stress, participants were allowed to skip any items they felt uncomfortable answering. This approach respected participants’ autonomy and ensured that they could provide responses voluntarily; incomplete responses were monitored for data quality.

The SSI consists of 40 items assessing four domains of stress: Physical Stressors (10 items), Academic Stressors (10 items), Interpersonal Relationships (10 items and Environmental Stressors (10 items). Items were rated on a 4-point Likert scale as follows: Never (1), Somewhat Frequent (2), Frequent (3) and Always (4). The average completion time was 15–20 min. The instrument has demonstrated strong psychometric properties, with a reported content validity of 80.5% and a reliability coefficient of 0.85. Responses were summed to generate both domain-specific and overall stress scores. Sub-scale scores between 10 and 18 indicated low stress, 19–29 moderate stress and 30–40 severe stress. Overall stress levels were classified as follows: 40–80 (mild stress), 81–121 (moderate stress) and 122–160 (severe stress).

Data analysis for this study involved systematic coding, editing, tabulation and statistical evaluation to summarise and interpret the information collected. The Statistical Package for the Social Sciences (IBM SPSS Statistics, Version 30, IBM Corp., Armonk, New York, United States) was used to perform all statistical analyses. At the same time, Microsoft Excel was employed to organise data and present findings through graphical representations, such as bar charts and pie charts. Descriptive statistics, including frequencies and percentages, were used to provide an overview of stress levels among nursing students, categorising participants into mild, moderate or severe stress groups. Associations between demographic characteristics and the four stressor domains of the SSI (physical, interpersonal, academic and environmental) were assessed using Fisher’s Exact Test. Pearson’s correlation coefficients were computed to examine linear relationships between age and each stress domain, as well as interrelationships among the stress domains themselves. To evaluate the influence of gender on overall stress levels, an ordinal logistic regression analysis was conducted. Finally, the Kruskal–Wallis H test was performed to determine whether stress levels differed significantly across students at different levels of study.

The Turfloop Research Ethics Committee granted ethical approval for the study (TREC/654/2025:UG). Permission to conduct the study was also obtained from the Director of the School of Health Care Sciences and the Head of the Department of Nursing Sciences. Participation was entirely voluntary, and individuals provided electronic informed consent prior to completing the survey. All responses were anonymous, and data were stored securely to ensure confidentiality. While the questionnaire inquired about stress levels, which may have prompted participants to reflect on personal stressors, measures were taken to minimise any potential distress. Participants were provided with contact information for the institutional student counselling departments, which are responsible for providing mental health services to students. They were encouraged to seek support from these services if completion of the questionnaire triggered any emotional discomfort.

As shown in Table 2, nursing students were assessed for stress across four domains: physical, interpersonal relationships, academic and environmental. For physical stress, nearly half of the participants reported moderate stress (50.4%), followed closely by mild stress (46.6%), with only 3.1% experiencing severe stress. In the interpersonal relationship domain, most students reported mild stress (55.7%), 42.2% reported moderate stress, and 1.9% reported severe stress. Regarding academic stress, more than half of the participants (55.0%) experienced moderate stress, 25.6% reported severe stress, and 19.5% reported mild stress. Similarly, in the environmental domain, stress was predominantly moderate (54.6%), followed by severe (24.0%) and mild (21.4%). Overall, while moderate and severe stress levels were most prevalent in the academic and environmental domains, mild stress was observed in all domains, particularly physical and interpersonal stressors.

The findings of this study revealed several statistically significant associations, as shown in Table 3. Gender is significantly associated with physical stressors (p < 0.001) and academic stressors (p = 0.007), suggesting that male and female students experience these domains differently. No significant associations were found between gender and interpersonal relationship stressors (p = 0.950) or environmental stressors (p = 0.315). The level of study is significantly associated with physical stressors (p = 0.011) and academic stressors (p = 0.001), indicating that stress experiences vary depending on the stage of study. No significant associations were observed with interpersonal (p = 0.166) or environmental stressors (p = 0.190). Financial Status is also significantly associated with academic stressors (p = 0.015). However, no significant relationships were found with physical (p = 0.195), interpersonal (p = 0.502) or environmental stressors (p = 0.313). In addition, race and relationship status showed no significant associations with any of the stressor domains (all p > 0.05).

A correlation analysis was conducted to examine the relationships between age and stress across the different domains (physical, interpersonal, academic and environmental) as depicted in Table 4. Age showed no significant association with any of the stress domains, indicating that age was not related to stress levels in this sample. In contrast, significant positive correlations were observed among the stress domains. Physical stress was moderately correlated with interpersonal (r = 0.408, p < 0.001), academic (r = 0.337, p < 0.001) and environmental stress (r = 0.367, p < 0.001). Similarly, interpersonal stress was positively associated with academic (r = 0.319, p < 0.001) and environmental stress (r = 0.342, p < 0.001), while academic stress was moderately correlated with environmental stress (r = 0.368, p < 0.001).

The final model demonstrated a significantly better fit than the intercept-only model, χ²(1) = 17.45, p < 0.001 as indicated in Table 5. Model fit indices suggest a good fit to the data, Pearson χ²(1) = 0.86, p = 0.353; Deviance χ²(1) = 0.65, p = 0.419. The assumption of proportional odds was met, χ²(1) = 0.65, p = 0.419. Therefore, gender significantly predicts stress levels. Specifically, males have significantly lower odds of reporting higher levels of stress than females (B = –1.28, s.e. = 0.31, Wald χ²(1) = 16.83, p < 0.001, 95% CI [–1.90, –0.67]). The odds ratio indicates that males were about 72% less likely to report higher stress levels than females (OR = 0.28, 95% CI [0.15, 0.51]). Observed and expected counts were closely aligned across categories, with Pearson residuals all < 1, further supporting model adequacy.

The results indicate a statistically significant difference in stress among study levels, H(3) = 12.71, p = 0.005. As shown in Table 6, Level 2 students reported the highest mean rank of stress (MR = 152.47), while Level 3 students had the lowest mean rank (MR = 114.25). Level 1 (MR = 127.34) and Level 4 (MR = 130.63) reported intermediate stress levels. Post-hoc pairwise comparisons using Bonferroni adjustment reveal that Level 2 students experience significantly higher stress than Level 3 students (p = 0.003). No other pairwise comparisons reached statistical significance. These findings suggest that stress levels vary across study years, with second-year students reporting the greatest stress.

This study relies on self-reported data, which may be subject to social desirability and recall bias. As a result, the reported stress levels may underestimate or overestimate the true prevalence of stress among nursing students, potentially affecting the accuracy of associations with demographic factors. The cross-sectional design limits causal inferences, providing only a snapshot of stress at a single point in time. Consequently, observed relationships between stress and demographic characteristics are correlational and do not reflect temporal changes or causality. Data were collected from a single NEI, which may limit generalisability to other settings. Thus, findings should be interpreted cautiously when applying them to nursing students in other provinces or NEIs. Finally, while the SSI demonstrated strong reliability and content validity, future studies could consider using tools specifically validated for South African nursing students, such as the Perceived Stress Scale adapted by Engelbrecht (2021), to enhance construct validity and ensure findings are culturally and contextually relevant. Future research could also employ mixed-methods approach, longitudinal designs and multi-site sampling to provide a more comprehensive understanding of stress and its determinants in this population.

Based on the findings of this study, several recommendations are proposed for nursing students, NEIs, policymakers, curriculum developers and future research. Individual students should adopt effective coping strategies, including time management, mindfulness, peer support and other stress reduction techniques, particularly female and second-year students who reported higher stress levels. Students should also be encouraged to seek professional mental health support and make use of available counselling and mentorship services to manage both academic and clinical stressors.

For NEIs, it is recommended that gender-sensitive mentorship and support programmes should be developed, acknowledging that female students may experience higher levels of physical and academic stress. Furthermore, NEIs should monitor stress levels regularly to identify high-risk groups and intervene early.

Policymakers are encouraged to ensure adequate resources and funding for student mental health support within nursing education, including the implementation of stress management programmes and peer mentorship initiatives. Policies should also address financial constraints, which are associated with increased academic stress, to reduce barriers to student success.

Curriculum development should incorporate mental health literacy, stress management, and resilience training into the nursing programme, while structuring academic modules and clinical placements in a manner that balances rigour with student well-being. Special attention should be given to first- and second-year students who are introduced to complex theoretical and clinical modules, as these cohorts were observed to experience higher stress levels.