A total of 300 nursing interns were invited to participate in the survey, of whom 268 responded (response rate: 89.3%). After excluding 18 invalid questionnaires, 250 valid responses were retained for analysis. Hospital records confirmed that non-respondents (n = 32) did not differ significantly from respondents in terms of age, gender, or education level (p > 0.05). The minimum sample size was estimated using two approaches. First, based on the widely accepted rule of 5–10 participants per scale item in structural equation modeling (Comrey and Lee, 1992), the 12-item ICNE scale required a minimum of 60–120 participants. Second, an a priori power analysis using G*Power 3.1 with α = 0.05, medium effect size (Cohen’s f² = 0.15), and power = 0.95 indicated a required sample size of 200 (Babenko-Mould and Laschinger, 2014). Considering a potential 20% attrition rate due to invalid or incomplete responses, the target sample size was set at 240. A post hoc power analysis confirmed that the final sample (N = 250) achieved 97% statistical power, exceeding the recommended threshold.

Inclusion criteria were nursing interns who (1) had completed at least 7 months of clinical internship; (2) could independently understand and complete the questionnaire, and (3) voluntarily signed the informed consent form. The exclusion criteria were those (1) on sick or personal leave during the survey period; (2) with severe physical or mental health conditions affecting clinical training; and (3) unwilling or unable to complete the questionnaire.

The study was conducted at a tertiary teaching hospital in Hunan Province, China. The institution provides a clinical training environment comparable to other urban hospitals nationwide, although regional cultural and socioeconomic differences may limit the generalizability of findings to rural or non-urban contexts. Ultimately, 268 questionnaires were collected, and after excluding those with more than 20% missing responses or inconsistencies in reverse-scored items, 250 valid responses were analyzed.

All responses were anonymized to ensure confidentiality. A pilot test with 10 nursing interns confirmed questionnaire clarity and reliability using a paper-based format. Formal data collection was conducted online. A QR code linking to the digital questionnaire was distributed via the WeChat management group. Participants scanned the code, provided informed consent, and submitted responses within a 30-min window. The average completion time was 15 ± 3.2 min. The system restricted submissions to one per participant, and researchers monitored the process to prevent collaboration. To minimize common method bias, measures included anonymous participations and reverse-scored items such as “I am full of energy when I study” from the Learning Burnout Scale. Questionnaires with more than 20% missing responses or logically inconsistent answers to reverse-scored items were excluded. Missing data, which represented less than 2% per variable, were handled by listwise deletion.

Data analysis was conducted using SPSS 27.0 and AMOS 26.0. Descriptive statistics were used to summarize demographic characteristics and variable distributions. Harman’s single-factor test indicated no significant common-method bias (first-factor variance = 38.7%, below the 50% threshold). Pearson correlation analysis was performed to examine relationships among variables. Regression analysis assessed the effect of ICNE on learning burnout while controlling for covariates. Structural equation modeling (SEM) with a bias-corrected bootstrap method was employed to evaluate both direct and mediating effects. Model fit indices indicated good fit: χ²/df = 1.696, IFI = 0.996, TLI = 0.992, CFI = 0.996, and RMSEA = 0.053.

The participants in this study were aged 15–28 years, with a mean age of 20.25 ± 2.62 years. Females constituted 88.8% of the sample. Educational attainment was distributed as follows: junior college (associate degree) or below, 50.4%; bachelor’s degree, 35.2%; and master’s degree or above, 14.4%. Most participants were of Han ethnicity (90.8%), while 9.2% belonged to ethnic minority groups. Regarding family structure, 62.4% came from multi-child families. The motivations for choosing nursing varied, with 30% citing personal interest, 25% family influence, and 45% career prospects. The sample was almost evenly divided between urban (49.2%) and rural (50.8%) backgrounds. With respect to career intentions, 79.6% planned to pursue nursing or related fields, while 20.4% intended to pursue non-nursing careers. Nearly half of the participants (48%) had relatives in the nursing profession, and 70.8% were first-time interns. Detailed demographic data are presented in Table 1.

The primary variables analyzed in this study were ICNE, career calling, general self-efficacy, and learning burnout (Table 2). The mean ICNE score was 16.96 ± 7.13, suggesting that nursing interns experienced mild to moderate levels of incivility in clinical education, with considerable variability. ICNE scores differed significantly by education level, reasons for choosing nursing, and internship experience (p < 0.05). The mean learning burnout score was 55.78 ± 7.02, indicating a moderately high level of burnout, with some participants experiencing severe burnout. Learning burnout scores varied significantly across age, education level, only-child status, reasons for choosing nursing, career intentions, and whether family members were in the nursing profession (p < 0.05). The mean career calling score was 31.39 ± 3.56, reflecting a moderate level of career calling among most interns. Significant differences were observed across age, education level, reasons for choosing nursing, and internship experience (p < 0.05). The mean self-efficacy score was 22.84 ± 2.14, indicating a moderate level of self-efficacy with relatively low variability. Self-efficacy scores differed significantly across age, education level, and reasons for choosing nursing (p < 0.05). Results are summarized in Table 2.

As shown in Table 3, all four variables were significantly correlated. ICNE was negatively correlated with both career calling and self-efficacy, and positively correlated with learning burnout. Learning burnout was negatively correlated with career calling and self-efficacy, whereas career calling was positively correlated with self-efficacy.

Regression analysis results are presented in Table 4. ICNE and selected demographic factors significantly influenced learning burnout. The overall regression model was significant (F = 88.02, p < 0.001), explaining 78.6% of the variance in learning burnout. Significant demographic covariates included age, education level, and only-child status, with only-child status remaining significant after adjustment (β = 0.102, p < 0.05). ICNE remained a strong predictor of learning burnout, with coefficients stable before and after adjustment (β = 0.784 vs. β = 0.772). Specifically, ICNE was positively associated with learning burnout (t = 19.876, p < 0.001). These findings indicate that incivility in clinical nursing education is closely linked to higher levels of learning burnout among interns.

AMOS was used to test the mediating effects of career calling and self-efficacy. After a single model adjustment, residual 6 was correlated with residual 8 (Figure 1), and all fit indices met acceptable standards (Table 5). Path analysis (Table 4) showed that all five model paths were statistically significant. The results demonstrated that higher levels of perceived incivility were associated with lower self-efficacy and career calling, and with higher learning burnout. Bootstrap mediation analysis confirmed partial mediation (Table 6). The indirect effect of career calling (indirect effect 1) was significant, accounting for 27.3% of the total effect. The indirect effect of self-efficacy (indirect effect 2) was also significant, accounting for 20.6% of the total effect. Together, the indirect effects explained 47.9% of the variance, while the direct effect of incivility on learning burnout accounted for 52.1%. There was no statistically significant difference between the contributions of the two mediators (p = 0.387).

The mean incivility score reported by nursing interns was 16.96 ± 7.13, reflecting a mild to moderate level of perceived incivility with substantial variability. This finding is consistent with Ahn and Choi (2019), who similarly documented the prevalence of incivility in nursing education. Incivility has been reported across cultural contexts (Kim et al., 2024) and is known to adversely affect students’ learning experiences and psychological health, underscoring the urgency of addressing this issue. In the Chinese context, incivility may be linked not only to traditional hierarchical teaching models but also to stress, heavy workloads, and poor communication in clinical education. The present study also identified differences in incivility scores across demographic groups, particularly by education level and reasons for choosing nursing. Interns with higher education levels may be more aware of dignity and rights, making them more sensitive to incivility, whereas those who entered nursing under family pressure or by default may have weaker professional identities and thus perceive greater incivility. These findings highlight the importance of tailoring support to the individual backgrounds and developmental needs of interns, offering personalized support and assistance.

The total learning burnout score was 55.78 ± 7.02, indicating a moderately high level overall, with some interns experiencing markedly severe burnout. This result supports previous findings (Ghods et al., 2023; Wu et al., 2021) showing that burnout is a common issue among nursing interns, often linked to learning pressure, demanding workloads, and stressful clinical environments. Some interns may feel overwhelmed by academic and clinical demands, leading to disengagement and fatigue. These results emphasize the need for nursing educators to monitor interns’ learning status, implement strategies to reduce pressure, and enhance engagement in clinical education.

Interns generally reported moderate levels of career calling and self-efficacy, consistent with findings from Jiang et al. (2023) and Bulfone et al. (2022). Both constructs were shown to buffer burnout (Chen and Wang, 2019). In the Chinese cultural context, career calling is reinforced by Confucian values that emphasize professional ethics and social responsibility, thereby strengthening professional identity. Meanwhile, self-efficacy enhances confidence in clinical competence and resilience in stressful environments. Thus, clinical nursing education should incorporate individualized career guidance and structured support to cultivate career calling and self-efficacy.

The results revealed a significant negative correlation between incivility and both career calling and self-efficacy, and a significant positive correlation between incivility and learning burnout. These findings are consistent with Jiang et al. (2023) and Pitacho and Cordeiro (2023), who reported that incivility undermines career calling and self-efficacy, thereby exacerbating burnout. Previous research has also shown that interventions promoting self-efficacy and professional identity buffer the effects of workplace incivility (Lee et al., 2019). However, studies such as Chen et al. (2019) suggest that in highly authoritarian educational contexts, these interventions may be less effective, emphasizing the need for context-specific approaches. Additionally, this study found a significant negative correlation between learning burnout and both career calling and self-efficacy and a significant positive correlation between career calling and self-efficacy. This finding indicates that interns with stronger professional confidence and calling are more motivated and self-disciplined in their professional studies.

Regression analysis further demonstrated that incivility positively predicted learning burnout (β = 0.784, p < 0.001), accounting for 78.6% of the variance. This aligns with Babenko-Mould and Laschinger (2014), who also identified incivility as a strong predictor of burnout. Additionally, being an only child significantly influenced burnout levels, possibly due to the absence of sibling support and heightened family expectations under China’s one-child policy (Watson, 2023). Interns from single-child families may rely more heavily on individual coping mechanisms, making them more vulnerable to stress and burnout. These findings highlight the importance of providing additional psychological and educational support for this demographic group.

Structural equation modeling revealed that career calling and self-efficacy partially mediated the relationship between incivility and learning burnout. This finding offers a new perspective for understanding the mechanism by which incivility in nursing education affects learning burnout. Specifically, incivility directly contributed to 52.1% of the total effect on burnout, with indirect effects mediated by career calling (27.3%) and self-efficacy (20.6%). These findings are consistent with studies emphasizing the buffering role of intrinsic motivation in adverse environments (Watson, 2023; Wu et al., 2021). Specifically, the mediating effects of career calling and self-efficacy highlight the crucial role of intrinsic motivation and confidence when interns face incivility. Career calling allows interns to sustain motivation and a sense of professional responsibility toward their profession when facing challenges, while self-efficacy enables them to cope more effectively with stressors. Together, these mediators highlight the protective role of internal resources in mitigating negative educational experiences. Therefore, enhancing career calling and self-efficacy through career values education, clinical practice opportunities, timely feedback and support, and improved teamwork and communication should be prioritized in nursing education to strengthen interns’ resilience and career development.

Consistent with global literature, this study confirms that incivility significantly predicts learning burnout (β = 0.784, p < 0.001), echoing findings from Western contexts (Babenko-Mould and Laschinger, 2014). However, the present study extends prior knowledge by quantifying the mediating effects of career calling and self-efficacy, which together accounted for 47.9% of the incivility–burnout relationship. Although Lopes and Nihei (2020) emphasized the buffering role of self-efficacy, they did not specify its proportion in the overall mechanism.

Cultural context also plays a critical role. Whereas Western studies have highlighted external support systems (Kim et al., 2024), Chinese interns appear to rely more on internal motivators such as career calling, reflecting Confucian values that prioritize duty and perseverance. This cultural influence may explain the comparatively larger mediating role of career calling (27.3%) observed in our model compared with reports from more individualistic cultures (Kim et al., 2021).

The finding that career calling and self-efficacy independently mediated the incivility–burnout relationship aligns with Shang et al. (2022), who observed dual pathways in preservice teachers. However, unlike Shang et al., who reported sequential mediation (career calling leading to self-efficacy), our model supports parallel effects, suggesting that in nursing interns, calling and self-efficacy operate independently. This difference may be linked to Confucian collectivism, which emphasizes professional responsibility as a standalone value.

These findings underscore the need for cross-cultural replications. For example, whereas our results suggest that Chinese interns rely primarily on internal regulation, studies from other collectivist contexts, such as South Korea, highlight the importance of peer networks (Riley et al., 2021). Future research should investigate how cultural norms and educational policies interact to shape responses to incivility across different settings.

This study has several limitations. First, the sample comprised nursing interns from a single region in China, limiting the generalizability of the findings to broader national or cross-cultural contexts. Although the hospital’s clinical training environment was consistent with that of urban teaching hospitals, regional and cultural differences may restrict applicability to rural or less resourced settings. Second, the cross-sectional design prevents causal inference, and reverse causality (e.g., burnout influencing perceptions of incivility) cannot be excluded. Third, the exclusive use of self-reported data may have introduced response bias. Fourth, confirmatory factor analysis (CFA) was not performed to validate the measurement structures, thereby limiting verification of convergent and discriminant validity. The decision not to conduct CFA was methodologically justified by three considerations: (1) sample size limitations, as robust CFA for multidimensional scales requires larger samples than mediation models. For instance, Wolf et al. (2013) recommend at least 10 participants per item (e.g., 120 for the 12-item ICNE), whereas our sample (N = 250) was allocated to ensure adequate power for path analysis rather than psychometric testing (MacCallum et al., 1999); (2) established alternatives, given that validated scales with high internal consistency (α > 0.85) and theoretically consistent correlation patterns (e.g., ICNE–career calling r = −0.847) provide sufficient evidence for model testing without CFA (Podsakoff et al., 2012); and (3) cross-cultural precedent, since the Chinese adaptations of these measures have shown strong metric invariance in prior studies (CFI = 0.98; Lian et al., 2005: RMSEA = 0.04), reducing the immediate need for revalidation (Cheung and Rensvold, 2002). Future research should therefore extend sampling to diverse geographic and healthcare settings, employ longitudinal or experimental designs to clarify causality, and integrate multi-source or objective data to reduce bias. Additionally, CFA should be incorporated in future studies to strengthen psychometric evaluations, particularly when adapting measures across culturally distinct populations.

This study advances theoretical understanding of the mechanisms linking incivility with learning burnout in nursing education. By identifying career calling and self-efficacy as significant mediators, accounting for 27.3 and 20.6% of the indirect effects respectively, our findings validate and extend the JD-R model (Bakker and Demerouti, 2017), emphasizing the role of intrinsic motivators in buffering workplace stressors. Furthermore, the integration of career calling within the Chinese cultural context, shaped by Confucian values that prioritize professional ethics and social responsibility, offers a culturally nuanced perspective on vocational identity (Dik and Duffy, 2009). This contribution addresses a gap in the literature, which has primarily focused on Western, individualistic frameworks.

The findings have important implications for clinical nursing education in China. First, enhancing career calling—identified as the stronger mediator (27.3%)—requires embedding Confucian virtues such as ren (benevolence) and professional ethics into mentorship programs. Clinical instructors should model compassionate care and employ narrative pedagogy to illustrate how preserving patient dignity can overcome workplace adversity. Second, interventions to strengthen self-efficacy should be context-specific. Structured simulation training should replicate high-stress clinical situations, particularly those in emergency department rotations where incivility was most frequently reported. Third, institutional responses should move beyond punitive measures by embedding principles of mutual respect into teacher certification standards and implementing anonymous, real-time feedback systems to systematically reduce incivility. Finally, targeted support for single-child interns—who reported 12.3% higher levels of burnout—could be developed through family-school collaboration models rooted in Chinese collectivist traditions, encouraging parental engagement in resilience-building efforts.

In summary, this study demonstrates that incivility significantly exacerbates learning burnout among Chinese nursing interns and identifies career calling and self-efficacy as key protective mechanisms. These findings contribute to both theoretical frameworks and practical strategies for fostering resilience in clinical education.