Mediating Role of Psychological Resilience between Social Support and Negative Emotions in Nursing Interns in the Post-COVID-19 Era: A Cross-Sectional Survey

Student summary

Why This Research Matters

Nursing internships are stressful even in ordinary times, but the post-COVID-19 period added extra strain: staffing gaps, infection worries, and disrupted clinical placements. This study asked a simple but important question: why do some nursing interns cope with that stress better than others, and does something specific — psychological resilience — explain how having social support around you translates into fewer negative emotions like anxiety and depression?

The researchers surveyed 788 nursing interns using a cross-sectional design, meaning everyone answered the questionnaires at a single point in time rather than being followed over months or years. Four validated tools were used: a Psychological Resilience Scale, a Social Support Scale, the Beck Anxiety Inventory, and the Beck Depression Inventory. With these scores in hand, the team used structural equation modeling (SEM), a statistical technique that lets researchers test whether one variable (resilience) sits in the causal pathway between two others (social support and negative emotions), rather than just checking if two things are correlated.

The numbers were sobering: 24.7% of interns screened positive for anxiety disorder, and 10.5% screened positive for depression. That is a meaningful chunk of the nursing workforce pipeline entering clinical practice while already managing significant mental health symptoms.

The correlational findings lined up with what resilience theory would predict. Social support and psychological resilience were both negatively correlated with negative emotions — in plain terms, interns who felt more supported and more resilient reported less anxiety and depression. Psychological resilience was also positively correlated with social support, suggesting the two protective factors reinforce each other rather than working in isolation.

The centerpiece finding is the mediation result: psychological resilience partially mediated the relationship between social support and negative emotions, accounting for 53.9% of that relationship, with an effect value of -0.1456. In everyday language, roughly half of the protective effect that social support has on reducing anxiety and depression appears to work through building up an intern's psychological resilience — their internal capacity to bounce back from setbacks, regulate emotions, and stay engaged with problems rather than avoid them. The other roughly 46% of social support's protective effect worked directly, without going through resilience at all. That is what "partial" mediation means here: resilience is an important pathway, but not the only one.

Why does this matter for a nursing student? It suggests that support systems (mentors, family, peer groups, clinical instructors) don't just make interns feel better in the moment — they may also be building a more durable internal skill (resilience) that keeps protecting the intern even when a specific support person or resource isn't immediately available. That has practical implications: nursing programs that only offer emotional support (a listening ear) may be missing half the picture if they don't also intentionally build resilience skills — things like reframing setbacks, structured debriefing after difficult clinical experiences, and stress-management training.

A few cautions are worth carrying into how you read this study. It is cross-sectional, so it captures a single snapshot in time; it cannot prove that resilience causes reduced negative emotions, only that the statistical pattern is consistent with that causal story. It relies entirely on self-report questionnaires, which are useful but imperfect measures of anxiety, depression, resilience, and support. And while the abstract itself does not name the setting, confirmed bibliographic records identify this as a Chinese nursing-intern sample published in Psychology in the Schools (2023), so the specific figures may not transfer directly to Canadian nursing programs, where post-pandemic student wellbeing is a live concern but the clinical-education and support context differs. Read the mediation percentage as a description of the statistical model, not a guarantee about any one individual's experience — group-level patterns don't always predict what will help a specific struggling intern.

Faculty notes

Educational Relevance

This cross-sectional survey (N = 788 nursing interns) examines whether psychological resilience mediates the relationship between perceived social support and negative emotions (anxiety, depression) among nursing students completing clinical internships in the post-COVID-19 period. The study is a useful teaching case for structural equation modeling (SEM) and mediation analysis, both increasingly common in nursing education and workforce-wellbeing research.

Methodologically, the design pairs four validated self-report instruments — a Psychological Resilience Scale, a Social Support Scale, the Beck Anxiety Inventory, and the Beck Depression Inventory — with SEM to test a mediation model in which social support is the predictor, negative emotion is the outcome, and psychological resilience is the mediating mechanism. This is a stronger analytic approach than simple correlation because it estimates the proportion of an association that runs through a specified intervening variable, but it still cannot establish temporal precedence or rule out reverse or reciprocal causation — resilience could just as plausibly buffer against symptoms that then make support feel more available, or vice versa, given the single time-point design.

The headline prevalence figures — 24.7% screening positive for anxiety and 10.5% for depression — are useful benchmarks for discussing intern/student mental health burden in seminar, and are broadly consistent with other post-pandemic nursing-student mental health surveys cited in the literature (e.g., compassion fatigue and PTSD studies in similarly aged cohorts). Instructors can use these figures to prompt discussion about whether such rates would meet a threshold for programmatic intervention at their own institution.

The mediation result — a partial mediating effect of 53.9%, effect value -0.1456 — is the study's central contribution. It supports a dual-pathway model: social support protects against negative emotion both directly and indirectly through building resilience. For curriculum planning, this argues against treating peer/mentor support and resilience training as interchangeable interventions; the abstract's own findings suggest programs should invest in both, since roughly half of support's protective value operates independently of resilience.

Discussion prompts for class: (1) Ask students to diagram the mediation pathway and explain in their own words what "partial" versus "full" mediation would mean for intervention design. (2) Have students critique the self-report, single-timepoint design and propose a longitudinal or mixed-methods follow-up that could strengthen causal claims. (3) Use the prevalence figures to discuss screening and referral pathways available to interns at your own clinical sites. (4) Ask what confounders (workload, shift patterns, prior mental health history, program year) might not be controlled for given the abstract does not describe covariates.

Limitations for appraisal: cross-sectional design precludes causal inference; reliance on self-report scales introduces common-method variance and social-desirability bias; the abstract provided does not specify country, recruitment method, response rate, or how missing data were handled, so generalizability claims should be made cautiously; and the abstract does not report effect sizes for the direct paths separately from the total effect, limiting how precisely instructors can quantify the "direct" versus "indirect" contributions beyond the reported percentage. External bibliographic records place this study in Psychology in the Schools (2023, Vol. 60, Issue 8), which instructors may wish to consult in full text via institutional library access, as the full paper was not openly retrievable during this review; discuss with students the value of checking DOI 10.1002/pits.22874 through institutional subscriptions when full appraisal is required.