1Department of General Surgery, The Second People’s Hospital of Yibin, Yibin, China
2Department of Emergency Medicine, West China Hospital, Sichuan University, Chengdu, China
3Department of Intensive Care Unit, The Second People’s Hospital of Yibin, Yibin, China
4Department of Nursing, The Second People’s Hospital of Yibin, Yibin, China
Objective: This study examines the mediating role of regulatory emotional self-efficacy (RESE) in female nurses on the relationship between work-family conflict (WFC) and turnover intention (TI), focusing specifically on female nurses with two children in Tertiary Grade-A hospitals in Sichuan Province. By exploring how RESE influences the impact of WFC on nurses’ intentions to leave, the study aims to offer valuable insights for strategies to reduce nurse turnover and enhance the stability of the nursing workforce.
Methods: A convenience sampling method was employed to recruit 1,370 female nurses with two children from 65 tertiary hospitals across 21 cities and prefectures in Sichuan Province. Participants completed a general information questionnaire along with the Work-Family Behavioral Role Conflict Scale (WFBRCS), the Regulatory Emotional Self-Efficacy Scale (RESES), and the Turnover Intention Scale (TIS) to assess WFC, RESE and TI.
Results: Female nurses with two children reported a mean TIS score of 13.11 ± 3.93, with an average item score of 2.18 ± 0.66, reflecting a high level of TI. WFC was positively associated with TI (r = 0.485, p < 0.01), whereas RESE was negatively associated with TI (r = −0.382, p < 0.01). Furthermore, RESE was found to partially mediate the relationship between WFC and TI, suggesting that higher emotional self-efficacy can buffer the impact of WFC on nurses’ intentions to leave.
Conclusion: It is recommended that hospital management implement a dual-track intervention system emphasizing both resource optimization and psychological empowerment. Practically, this could involve adopting flexible scheduling and increasing nursing staff to alleviate the strain of heavy workloads on nurses’ family responsibilities. Simultaneously, providing mindfulness training, regular psychological counseling, and emotional management workshops can strengthen nurses’ RESE, enabling them to better balance work and family demands, reduce TI, and promote the stability and sustainable development of the nursing workforce.
1 Introduction
As the population ages and global public health emergencies become increasingly frequent, the demand for nursing personnel continues to grow (Zhang et al., 2023), making nurse shortages a universal challenge for healthcare systems (Fasbender et al., 2019). In this context, reducing turnover rates is pivotal to stabilizing the nursing workforce. Global nurse turnover rates vary widely, ranging from 8 to 36.6% (Ren et al., 2024), while Chinese tertiary hospitals face an attrition rate of 13.1% (National Health Commission of the People’s Republic of China, 2023), which represents a high-risk threshold. Nurse turnover not only results in wasted training resources but also compromises the quality of patient care (Wu et al., 2014). Turnover intention (TI), as a key antecedent of actual turnover behavior (Hann et al., 2011), averages 28% globally (Xu et al., 2023) and reaches 49.58% among Chinese nurses (Wang et al., 2022), underscoring the urgent need for effective human resource management interventions.
China’s universal two-child policy, fully enacted in 2016, has substantially restructured China’s labor force demographic, disproportionately large impact on female employees, as cumulative empirical evidence shows each additional child is associated with a 10-percentage-point reduction in women’s labor force participation rates (Yang and He, 2022). This policy-driven trend creates acute organizational challenges for healthcare systems, particularly pronounced for nursing given that women constitute 98% of China’s nursing workforces (National Health Commission of the People’s Republic of China, 2023). Among these female nurses, 77% are of childbearing age (≤35 years) (National Health Commission of the People’s Republic of China, 2023), accounting for 40.79% of this group (Wu et al., 2022) have two children, and their turnover rate (63.73%) significantly exceeds that of the general nursing population (Li, 2018). Evidence indicates that the number of children is positively associated with turnover risk (Rasheed, 2019; Yao et al., 2024), particularly among nurses with two children (Huang et al., 2020), with work-family conflict arising from childcare explicitly identified as the top reason for female healthcare workers leaving their jobs (Lin and Li, 2025). Grounded in the conservation of resources (COR) theory (Hobfoll, 1989), these nurses face a resource loss spiral, including temporal conflicts between shift work and childcare, as well as emotional labor arising from both professional caregiving and family responsibilities. When losses (e.g., missed family events due to night shifts) outweigh gains (e.g., social support), TI emerges as a protective response to halt further depletion.
Work-family conflict (WFC) exemplifies this resource depletion. Under the two-child policy, a resource loss spiral emerges: night shifts drain temporal resources, while the dual demands of professional and family caregiving exhaust emotional resources, collectively exceeding nurses’ available resource thresholds; Emotional stress from balancing professional care and family responsibilities, compounded by disrupted career progression due to childbirth, contributes significantly to WFC. WFC refers to the tensions that arise when role demands in the work and family domains are incompatible (Beutell, 1985), encompassing both work-to-family interference (WIF) and family-to-work interference (FIW) (Netemeyer et al., 1996), and represents a critical manifestation of resource depletion. According to role conflict theory (Kahn et al., 1964), women face dual expectations from the workplace and family, intensifying WFCs, a phenomenon particularly pronounced in nursing. Empirical research indicates that the high-pressure, high-intensity nature of nursing work exacerbates WFC among nurses (Zhang et al., 2017; Grzywacz et al., 2006), which not only leads to emotional exhaustion but also serves as a significant predictor of TI (Hanoum et al., 2024), with WFC exerting a particularly strong negative impact (Zhou et al., 2020).
Regulatory emotional self-efficacy (RESE) may serve as a crucial internal resource that mitigates the impact of WFC on TI. RESE reflects an individual’s perceived ability to regulate and manage emotions (Bandura et al., 2003) and encompasses three key dimensions: These dimensions include perceived positive self-efficacy (POS), despondency/distress emotional self-efficacy (DES), and anger emotional self-efficacy (ANG) (Carletto et al., 2016). Managing despair from overtime and anger from patient conflicts align with COR’s tenet that individuals prioritize protecting vulnerable resources. Previous research indicates that WFC is negatively associated with RESE (Wang et al., 2023), whereas nurses with higher RESE tend to exhibit lower TI (Ling et al., 2024). These individuals are more likely to employ adaptive coping strategies, which help buffer stress and mitigate the adverse effects of WFC (Yaghoobi et al., 2019). However, most existing studies have focused primarily on external resources, such as workplace conditions (Jiang and Kira, 2024) and organizational support (Galanis et al., 2024), while giving insufficient attention to internal psychological resources like RESE. Additionally, international research has largely concentrated on nursing workforces in Western contexts, providing limited analysis of gender-role dynamics within the framework of China’s two-child policy.
Building on COR theory, this study frames WFC as a resource threat spiral for female nurses with two children, adapted to China’s contexts. The dual-domain nature of WFC (WIF and FIW) directly reflects COR’s principle that stress occurs when key resources, such as time, energy, and emotional capacity, are depleted across competing domains. Under the influence of China’s traditional culture of “men do external things and women do internal things,” WFC’s influence on TI has been amplified. RESE is proposed as a resource caravan passageway that buffers this spiral, with its dimensions, DES, ANG, and perceived POS, aligning with COR’s idea that individuals prioritize protecting the resources most susceptible to loss. Notably, COR theory underemphasizes how China’s cultural role expectations worsen resource depletion, a factor rarely addressed in Western COR nursing studies. To fill this gap, our study integrates this element into COR: it links cultural norms to amplified WFC, notes scheduling constraints reduce resource protection, positions RESE as a buffer, and clarifies the interplay of WFC, RESE and TI in China’s context. Based on this, we present the conceptual model in Figure 1.
Figure 1
Figure 1. The theoretical model.
This study aims to examine the impact mechanisms of WFC and RESE on TI among female nurses with two children, while validating a culturally adapted COR framework. Theoretically, it contributes by refining COR’s application in non-Western healthcare settings. Practically, it provides a foundation for hospital administrators to design targeted psychological interventions and human resource management strategies. By mitigating TI associated with WFC in this nursing subgroup, the study supports nursing team stability and healthcare service quality improvement, while enhancing COR’s cultural sensitivity in the context of Chinese nursing research.
We propose the following hypotheses:
H1: WFC is positively associated with TI (path c).
H2: WFC is negatively correlated with RESE (path a).
H3: RESE is negatively associated with TI (path b).
H4: RESE mediates the effect of WFC on TI (path c’).
2 Materials and methods
2.1 Participants
This study employed a multi-center convenience sampling approach, recruiting 1,370 female nurses with two children from 65 tertiary hospitals across Sichuan Province between September and December 2023. Sichuan Province is one of the most populous provinces in China’s southwest region, administering 21 prefecture-level cities and autonomous prefectures. Tertiary hospitals are the core of China’s medical system, and the 65 tertiary hospitals involved in the study are distributed across all 21 prefecture-level administrative regions of Sichuan Province, ensuring the practical accessibility of sampling and the diversity of the sample. Inclusion criteria required participants to be registered female nurses employed by the hospital, with at least 1 year of clinical experience, two healthy children, and provision of informed consent. Exclusion criteria encompassed nurses undergoing standardized training programs, those with multiple pregnancies, divorced or widowed nurses, and individuals not currently engaged in clinical nursing or absent from such roles for over 6 months.
The sample size was determined through a rigorous, multi-step calculation to address both descriptive study requirements and the statistical demands of structural equation modeling (SEM). First, the sample size was calculated using the formula for descriptive studies, number of research variables multiplied by 5–10, totaling 26 variables: 18 demographic items, 2 dimensions from the WFC Scale, 3 from the RESE Scale (RESES), and 3 from the TI scale (TIS). Accordingly, the estimated sample size ranged from 130 to 260, which, after adjusting for a 20% attrition rate, increased to 312-468. Third, combine the sample size calculation requirements of structural equation model, to ensure the robustness of the model, the sample size should not be less than 200 cases (Wu, 2010). The study further expanded enrollment to ensure methodological rigor, the actual sample size reached 1,370 nurses, well above the minimum requirement. The study protocol received approval from the Medical Ethics Committee of The Second People’s Hospital of Yibin (Ethical Approval No.: 2023-225-01), and all participants provided written informed consent.
2.2 Research tools
2.2.1 General information questionnaire
Drawing on a comprehensive literature review, the researchers developed a survey tailored to the study’s objectives. The questionnaire first ensured respondents met the core inclusion criterion of “being female nurses with two children” and then gathered detailed information on their demographic, professional, and family characteristics to explore potential influencing factors of core variables. The specific items included: hospital affiliation, age, professional title, educational background, department, position, personnel relationships, spouse’s occupation, years of work experience, monthly income, number of night shifts per month, the age and gender of both the first and second child, whether the second child was planned (designed to investigate the impact of fertility decision-making background and core study variables), the family’s primary caregiver, and total household income.
2.2.2 Work-family behavior role conflict scale(WFBRCS)
This study utilized the Chinese version of the WFBRCS, translated by Sun et al. (2023) from the original 2019 version developed by Clark (2019). The scale includes two core subdimensions, 30 items in total, to measure bidirectional work-family conflict (WFC). The first is Work-to-Family Interference (WIF), which contains 15 items focusing on how work roles disrupt family life, with scores reflecting the frequency of work interfering with family responsibilities; the second is Family-to-Work Interference (FIW), which includes 15 items assessing how family roles hinder work performance, with scores capturing the frequency of family interfering with work duties. By assessing specific behavioral role conflicts, the scale offers a comprehensive understanding of the WFC process and informs the design of targeted interventions. All items are rated on a 5-point Likert scale (1 = “never,” 5 = “always”), with a theoretical mean score of 3 serving as the threshold—scores below 3 indicate low conflict, while scores of 3 or higher indicate high levels of WFC. In the present study, the scale demonstrated excellent reliability, with a Cronbach’s α coefficient of 0.952.
The RESES, originally developed by Caprara et al. (2008) and later localized and revised by Sun et al. (2023), comprises three dimensions: perceived POS, DES, and ANG. The scale consists of 12 items, each rated on a 5-point scale, where 1 represents “very inconsistent” and 5 represents “very consistent,” with higher total scores reflecting greater RESE. Based on average item scores, RESE levels can be classified as low (≤2.5), medium (2.6–4.0), or high (≥4.1). In this study, the scale demonstrated excellent reliability, with a Cronbach’s α coefficient of 0.967.
2.2.4 Turnover intention scale (TIS)
The TIS, originally developed by Michaels and Spector (1982) in 1982 and later translated and revised by Li and Li (2000), is designed to assess an individual’s level of TI. The scale consists of six items across three dimensions: TI I (Intention to Search for Alternative Jobs) includes 2 items measuring the willingness to actively seek other positions; TI II (Intention to Resign) contains 2 items assessing the specific intention to leave the current job; TI III (Evaluation of Job Mobility) includes 2 items reflecting the attitude toward job switching. All items are rated on a 4-point Likert scale (1 = never, 2 = rarely, 3 = occasionally, 4 = frequently), yielding a total score range of 6 to 24. Higher scores indicate stronger TI. Based on average item scores, TI is categorized as very low (≤1), low (>1 to ≤2), high (>2 to ≤3), and very high (>3). In the present study, the scale demonstrated good reliability, with a Cronbach’s α coefficient of 0.838.
2.3 Data collection methods
Data were collected using the Questionnaire Star electronic platform. Prior to the formal survey, a pilot study was conducted in August 2023 to refine questionnaire wording and structure based on feedback. The research team held online meetings with nursing departments from all 65 participating hospitals to explain the study objectives (Ethics Approval No. 2023-225-01), anonymity safeguards, and data usage. After obtaining institutional authorization, standardized training was provided to nursing liaison officers at each hospital, including demonstrations of questionnaire completion, answers to frequently asked questions, and guidance on optimal administration timing, such as avoiding shift-change periods. During this training, we explicitly required liaison officers to only distribute the questionnaire link to nurses confirmed to meet the “having two children” criterion and to refuse access to ineligible nurses, ensuring the initial participant pool targeted the core study population. The platform incorporated quality control measures, including mandatory responses and restriction to one submission per IP address. All nurses voluntarily completed the questionnaires anonymously after providing informed consent. During the data collection period (September–December 2023), two researchers monitored submissions in real-time and tracked response rates for each hospital. A total of 1,392 questionnaires were collected, and to further ensure eligibility, we conducted post-collection screening: cross-checking with mandatory “age of the first child” and “age of the second child” items in the general information questionnaire (excluding those with blank, contradictory, or single-child-implying responses); ultimately, 22 invalid responses were excluded (9 for not meeting the “having two children” criterion, 13 for patterned responses or logical inconsistencies), resulting in 1,370 valid questionnaires and a 98.4% validity rate.
2.4 Statistical methods
After double-checking the data, questionnaire results were imported into SPSS 27.0 for statistical analysis, with p < 0.05 considered statistically significant. The normality of variables was assessed using Q-Q plots, and scale scores along with other quantitative data were reported as mean ± standard deviation. Descriptive statistics, including frequencies and percentages, were used to summarize demographic information. Pearson correlation coefficients were calculated to examine relationships among WFC, RESE, and TI. A structural equation model (SEM) was constructed in AMOS 24.0 to investigate the mediating role of RESE in the relationship between WFC and TI among female nurses with two children, and the significance of the mediation effect was tested using the Bootstrap method.
3 Results
3.1 General information about the subjects
The participants’ general information are presented in Table 1.
Table 1
Table 1. General data of female nurses with two children (n = 1,370).
3.2 Common method bias
To evaluate potential common method bias, Harman’s single-factor test was conducted. Factor analysis using SPSS identified eight factors with eigenvalues greater than 1, which together accounted for 68.02% of the total variance. The first factor explained 31.00% of the variance, below the critical 40% threshold, indicating that no significant common method bias was present in the questionnaire data.
3.3 Female nurses with two children WFBRC, RESE, and TI scores
The total scores and dimension scores of WFBRC, RESE, and TI are shown in Table 2
Table 2
Table 2. WFBRC, RESE, and TI scores (n = 1,370).
3.4 Correlation among WFC, RESE, and TI in female nurses with two children (aligned with H1, H2, H3)
The study found that WFC among female nurses with two children was positively correlated with TI (r = 0.485, p < 0.01), while RESE was negatively correlated with TI (r = −0.382, p < 0.01). Additionally, WFC was negatively correlated with RESE (r = −0.418, p < 0.01). Detailed correlation coefficients are presented in Table 3. The results of the study verified the validity of hypotheses 1–3.
Table 3
Table 3. Correlation analysis of WFC, RESE and TI of female nurses with two children (n = 1,370, r value).
3.5 The mediating effect of RESE of female nurses with two children on WFC and TI (aligned with H4)
In this study, WFC served as the independent variable, TI as the dependent variable, and RESE as the mediating variable. Stepwise regression analysis in SPSS indicated that RESE partially mediates the relationship between WFC and TI, with the explained variance increasing from 23.6 to 27.4% (p < 0.001). Detailed results are presented in Table 4.
Table 4
Table 4. Analysis of the mediating effect of RESE between WFC and TI.
Based on the aforementioned findings, we employed AMOS 24.0 to construct a structural equation model (SEM) following a two-stage analytical procedure. First, confirmatory factor analysis (CFA) was performed to validate the measurement model, demonstrating adequate psychometric properties with factor loadings ranging from 0.579 to 0.92 (all p < 0.01), satisfactory composite reliability (CR > 0.7), and acceptable convergent validity (AVE > 0.5) (Bagozzi et al., 1981). Discriminant validity was confirmed as the square roots of AVEs for all latent variables exceeded their inter-construct correlations in accordance with the Fornell-Larcker criterion. Subsequently, maximum likelihood estimation was applied to test and refine the hypothesized SEM, yielding final model fit indices that met recommended standards (Kang and Ahn, 2021) (see Table 5) with the structural relationships illustrated in Figure 2.
Table 5
Table 5. Standards and results for fitting indices in SEM.
Figure 2
Figure 2. Model diagram of the mediation effect of RESE between WFC and TI in female nurses with two children.
The Bootstrap method was used to examine the mediating effect of RESE on the relationship between WFC and TI. Bootstrap sampling was conducted with 5,000 iterations and a 95% confidence interval, with mediation considered significant if the interval did not include zero. The 95% confidence interval for the mediating effect was 0.050–0.157, excluding zero, indicating that RESE partially mediates the relationship between WFC and TI among female nurses with two children, accounting for 19.48% of the mediating effect (Table 6). This means that the validity of the H4 hypothesis is verified.
Table 6
Table 6. Analysis of the mediation effect of RESE between WFC and TI.
4 Discussion
4.1 Current status of WFC, RESE, and TI among female nurses with two children
4.1.1 Status of WFC
Our study found that the level of WFC among female nurses with two children was higher than that of Gan et al. regarding the general nurse population in China (Gan et al., 2025), consistent with COR theory: rotating night shifts compete directly with childcare needs, such as disrupted sleep versus school routines, while emotional labor is divided between patient care and parenting responsibilities (Hwang and Yu, 2021). Recent research indicates that nurses endure dual sources of emotional exhaustion arising from professional empathy at work (Huaman et al., 2023). The data reveal a significant imbalance between WIF (50.63 ± 10.81) and FIW (32.38 ± 11.17, p < 0.001). This discrepancy stems from the unique nature of nursing work: Female nurses with two children face rigid work requirements such as fixed hospital schedules and unexpected overtime (Chen et al., 2022). These demands make it difficult to flexibly allocate working hours and energy, thereby exerting greater pressure on family care responsibilities. Moreover, in China’s public hospitals, where nurses face expanding responsibilities including clinical care, teaching, and research (Cooper et al., 2019), these tasks take up a lot of downtime. In contrast, FIW had a weaker impact, which is associated with the “work-first” professional culture in China’s healthcare settings (Pajakoski et al., 2025). Most nurses proactively adjust family matters to minimize interference with work. Although this approach may temporarily enhance productivity, chronic role overload elevates turnover risk by straining family relationships (Gehri et al., 2023).
4.1.2 Current situation of RESE
This study found that the score of RESE among female nurses with two children exhibited above-average (Sun et al., 2023), this result is consistent with the findings of Wang et al., who conducted a study on infectious disease nurses in China (Wang et al., 2023), suggesting an adaptive capacity developed through prolonged dual-role exposure. This prolonged dual training may passively enhance their emotional regulation capabilities, which aligns with Bandura’s “practical reinforcement efficacy” theory (Bandura, 1977). However, their lowest scores were observed in anger management ANG, the possible reasons for this result are: First, in professional settings, nurses continuously cope with emotional events such as patient anxiety and conflicts in family communication (Pace et al., 2023). Long-term passive exposure to negative emotions easily depletes their emotional resources for anger regulation. Second, the complex task of childcare has left nurses extremely exhausted, physical fatigue further impairs their ability to perceive and manage anger (Getie et al., 2025). From the perspective of COR theory, these two factors essentially reflect the continuous depletion of emotional and physical resources among this group. Thus, it is suggested that healthcare institutions prioritize targeted interventions, such as providing regular anger regulation training to replenish emotional resources.
4.1.3 Current situation of TI
This study revealed elevated TI among female nurses with two children, with a mean item score of 2.18 ± 0.66. This score exceeds both the established benchmarks (Clark, 2019) and the TI score of nurses in a Chinese hospital reported by Zhang et al. (2023). Several factors likely contribute to this finding. First, 64.7% of participants held mid-level professional titles, reflecting high professional competence and career maturity; according to human capital theory (Yakusheva et al., 2024), such individuals may perceive resignation as a strategic career decision rather than a passive response to dissatisfaction. Second, 38.5% of nurses worked night shifts at least five times per month, interfering with family responsibilities (He et al., 2024). Within the context of traditional Chinese gender norms, where men are expected to work outside the home and women manage domestic duties, female nurses may be more inclined to leave their jobs to prioritize family obligations. Third, limited career advancement opportunities elevate turnover risk (Şahin et al., 2025). Specifically, among the three dimensions of TI, the subdimension “motivation for seeking other jobs” (TI II) had the highest average score (2.33 ± 0.71). This pattern is likely driven by the fact that 92.7% of participants held mid-level professional titles, which creates a bottleneck for further career progression. Compounding this, 78.8% of the nurses had a second child under 6 years old, a stage marked by particularly intensive parenting demands that may further amplify their inclination to seek other childcare support jobs. These findings suggest that nursing managers should closely address the unique needs of female nurses with two children and implement proactive measures to enhance job satisfaction and retention.
4.2 Correlation among WFC, RESE, and TI in female nurses with two children
4.2.1 The relationship between WFC and TI
Grounded in COR theory, this study demonstrates that WFC significantly depletes the personal and emotional resources of female nurses with two children, ultimately elevating their TI (β = 0.42, p < 0.01), consistent with existing evidence on nursing stress and attrition (He et al., 2024). Nursing, inherently demanding due to high work intensity and disruptive shift schedules (Yao et al., 2024), conflicts with the substantial resource demands of raising two children, including time, energy, and emotional labor. When these competing strains exceed nurses’ resource thresholds, COR theory’s “resource loss spiral” is triggered: the chronic depletion of critical resources, such as work-life balance and psychological stamina, diminishes job satisfaction and amplifies withdrawal behaviors like TI, as nurses seek to halt further resource erosion. Notably, the asymmetrical impact of FIW (r = 0.440) versus WIF (r = 0.412) reflects the defensive prioritization of family resources within Chinese cultural contexts. According to COR theory, individuals safeguard non-replenishable resources, such as maternal roles and family harmony, when threatened; consequently, unavoidable family crises, such as a child’s illness, often lead nurses to compromise career resources, including job retention, to preserve familial stability rather than risk irreversible damage to valued kinship roles (Yildiz et al., 2021). The administrator can develop a flexible shift system for the group of nurses, such as avoiding spouse shift overlaps, requesting daytime shifts during children’s exams. This measure does not require additional technical input, and the current intelligent shift system can complete it, which directly reducing turnover due to work-family conflict.
4.2.2 The relationship between RESE and TI
This study demonstrates that RESE negatively predicts TI, consistent with prior research in the Chinese context showing that emotional regulation self-efficacy (RESE) significantly influences nurse burnout (Moloney et al., 2018). From a COR theory perspective, RESE functions as a key “personal resource caravan,” a modifiable internal resource that buffers against resource loss spirals triggered by WFC. Dimensional analysis further highlights the DES subscale as the most influential protective factor, suggesting that nurses adept at managing distress, for example, through cognitive reappraisal, are less likely to resign despite the dual-role pressures of parenting and shift work. Self-determination theory (Song et al., 2023) supports this connection, positing that chronic unregulated emotions foster negative cognitions that elevate turnover risk. This aligns with COR’s proposition that individuals prioritize protecting resources most susceptible to depletion; for female nurses with two childrens, distress from unmet family demands or work overload is a primary vulnerable resource, which makes DES a critical buffer. Accordingly, interventions such as mindfulness training and emotion-focused workshops (Karahan Kaplan et al., 2024) may enhance RESE, particularly DES skills, helping nurses navigate the tensions between childcare and professional responsibilities while reducing attrition.
4.2.3 The relationship between WFC and RESE
This study confirms a negative correlation between WFC and RESE, demonstrating that persistent conflicts erode nurses’ confidence in managing emotions, a pattern consistent with Wang et al. (2023) and French studies on stress-induced self-efficacy reduction (Murtaza et al., 2021). From a COR theory lens, chronic WFC acts as a “resource drain” that weakens RESE: scenarios like guilt from missing a child’s parent-teacher conference or anxiety over declining work performance due to family demands deplete emotional resources, in turn reducing nurses’ perceived ability to manage future emotional challenges, as supported by Australian interventions showing that resilience training can restore coping capacity (Foster et al., 2024). By concurrently mitigating external stressors and strengthening internal resources, healthcare systems can disrupt the WFC-RESE depletion cycle, transforming it into a positive reinforcement loop whereby improved conflict management enhances emotional competence and reduces vulnerability to future stress.
4.3 The mediating effect of RESE of female nurses with two children on WFC and TI
The mediation analysis indicates that RESE partially mediates the relationship between WFC and TI, consistent with COR theory, which posits that stress arises when individuals’ critical resources are threatened or depleted (Hobfoll, 1989). In this context, WFC not only directly predicts TI but also indirectly exacerbates it by reducing RESE, a key psychological resource that buffers against emotional exhaustion. For nurses managing dual responsibilities, such as childcare and shift work, prolonged WFC depletes finite resources, including time and emotional energy, and as RESE declines, their ability to manage the resulting distress weakens, further amplifying TI as a protective response to halt resource loss.
Research shows that when balancing family responsibilities and professional commitments, individuals inevitably experience negative emotions such as sadness, frustration, anxiety, anger, and irritability (Feng et al., 2024; Guo et al., 2025). Importantly, RESE moderates this cycle: individuals with high RESE demonstrate greater competence in cognitive reappraisal and negative emotion regulation (Meng et al., 2024; Spitzen et al., 2022), particularly in managing depression (DES, λ = 0.68) and anger (ANG, λ = 0.69), which facilitates faster recovery from distress (Yu et al., 2024). SEM further highlights this asymmetry, showing that the regulation of negative emotions exerts an influence 1.7 times stronger than that of positive emotion expression (POS, λ = 0.39). From a COR theory perspective, this asymmetry reflects the fact that WFC primarily induces negative affect, making DES and ANG the most relevant resources for mitigating resource loss. This finding aligns with prior evidence that the consequences of WFC are driven by unmanaged negative affect (French and Allen, 2020), and it extends COR theory by identifying specific RESE dimensions that act as targeted buffers for this policy-specific subgroup.
Notably, while RESE attenuates the WFC → TI pathway (β = 0.426), its partial mediation underscores the need for concurrent structural and psychological interventions, which is consistent with COR’s “resource protection + resource building” framework (Hobfoll, 1989). Structural measures (flexible scheduling, on-site childcare support) reduce WFC at its source and curb initial resource depletion; targeted RESE training (integrated into in-service education) builds internal resources to buffer residual WFC. To complement these policies, institutional reforms should integrate targeted RESE-training programs into in-service education. For instance, they can organize emotional regulation training workshops twice a month, which are led by clinical psychologists. These seminars will cover modules including mindfulness practice, cognitive-behavioral techniques for anger management during high-stress shifts, emotional decompression after balancing dual work-family demands, and positive self-talk to alleviate turnover anxiety. This approach leverages existing hospital conference rooms, directly reinforces RESE’s mediating role in offsetting WFC-induced resource depletion, and aligns with national policies on nurse support and professional development. By extending COR theory to gendered occupational stress, this study identifies RESE as a trainable resilience factor. Healthcare systems seeking to reduce attrition must therefore address both systemic drivers of WFC and individual psychological resources, promoting sustainable nursing careers and higher standards of patient care.
5 Limitations
This study has several limitations. First, the sample comprised female nurses with two children from tertiary hospitals in Sichuan Province, China; due to regional and cultural factors, the findings may have limited generalizability internationally. Future studies should include more diverse regions and hospital levels to enhance representativeness. Second, although the cross-sectional design allowed for statistical validation of relationships among variables, it cannot establish causality. Future research employing longitudinal tracking or experimental designs is recommended to clarify temporal sequences and strengthen causal inferences in this field. Third, this study did not examine the influence of spousal support, such as participation in co-parenting, on resource compensation. Future research should incorporate marital interaction variables to better understand how family dynamics contribute to nurses’ WFC, RESE, and TI. Fourth, our study did not statistically control for potential confounders such as age, work experience, workplace type, marital status, professional titles, or night shift frequency. For example, nurses with longer tenure may have developed better coping strategies for WFC, potentially attenuating the observed effect sizes. Similarly, unmeasured differences in workplace support could bias the relationship between self-efficacy and turnover intention. Future studies should incorporate these variables into hierarchical regression or multi-group SEM analyses to validate our findings under more rigorous conditions.
6 Conclusion
This study found that female nurses with two children working in public hospitals in Sichuan Province exhibit elevated TI, with WFC positively correlated and RESE negatively correlated with TI. Moreover, WFC was negatively associated with RESE, which partially mediated its effect on TI. In response, hospital management should establish a dual-track intervention system emphasizing both resource optimization and psychological empowerment. Flexible scheduling and increased staffing can mitigate the negative impact of excessive workloads on nurses’ family lives, while mindfulness training, regular psychological counseling, and emotion management workshops can enhance nurses’ emotional regulation skills, enabling them to better navigate WFC, reduce TI, and support workforce stability and sustainable development. This study advances COR theory by empirically demonstrating that RESE functions as a modifiable energy resource that mitigates the resource loss spiral among female nurses with two children. Academically, this fills two gaps: prior COR studies in nursing focused on fixed personal resources (not modifiable emotional resources like RESE) and overlooked the “female nurses with two children” subgroup, while theoretically, it expands COR’s “resource types” (including modifiable state resources), refines the bidirectional “resource loss spiral” mechanism for dual-role nurses, and enhances the theory’s cultural adaptability to Chinese contexts; practically, it guides nursing management, policy formulation, and nursing education, addressing current gaps in each domain.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by Second People’s Hospital of Yibin, Sichuan, China. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
The author(s) declare that no financial support was received for the research and/or publication of this article.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The authors declare that no Gen AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Bagozzi, R. P., Fornell, C., and Larcker, D. F. (1981). Canonical correlation analysis as a special case of a structural relations model. Multivar. Behav. Res. 16, 437–454. doi: 10.1207/s15327906mbr1604_2
Bandura, A., Caprara, G. V., Barbaranelli, C., Gerbino, M., and Pastorelli, C. (2003). Role of affective self-regulatory efficacy in diverse spheres of psychosocial. Child Dev. 74, 769–782. doi: 10.1111/1467-8624.00567
Caprara, G. V., Di Giunta, L., Eisenberg, N., Gerbino, M., Pastorelli, C., and Tramontano, C. (2008). Assessing regulatory emotional self-efficacy in three countries. Psychol. Assess. 20, 227–237. doi: 10.1037/1040-3590.20.3.227
Carletto, S., Borghi, M., Francone, D., Scavelli, F., Bertino, G., Cavallo, M., et al. (2016). The efficacy of a mindfulness based intervention for depressive symptoms in patients with multiple sclerosis and their caregivers: study protocol for a randomized controlled clinical trial. BMC Neurol. 16:7. doi: 10.1186/s12883-016-0528-0
Chen, K., Wei, L., Zhang, Y., Jiang, W., Wang, J., and Pan, Y. (2022). Work stress in nurses returning to tertiary a general hospitals in China after the delivery of their second child: a cross-sectional study. BMC Health Serv. Res. 22:492. doi: 10.1186/s12913-022-07912-8
Clark, M. A. E. R. (2019). Work-family behavioral role conflict: scale development and validation. J. Bus. Psychol. 34, 39–53. doi: 10.1007/s10869-017-9529-2
Cooper, M. A., McDowell, J., and Raeside, L. (2019). The similarities and differences between advanced nurse practitioners and clinical nurse specialists. Br. J. Nurs. 28, 1308–1314. doi: 10.12968/bjon.2019.28.20.1308
Fasbender, U., Van der Heijden, B. I. J. M., and Grimshaw, S. (2019). Job satisfaction, job stress and nurses' turnover intentions: the moderating roles of on-the-job and off-the-job embeddedness. J. Adv. Nurs. 75, 327–337. doi: 10.1111/jan.13842
Feng, H., Zhang, M., Li, X., Shen, Y., and Li, X. (2024). The level and outcomes of emotional labor in nurses: a scoping review. J. Nurs. Manag. 2024:5317359. doi: 10.1155/2024/5317359
Foster, K., Shochet, I., Shakespeare-Finch, J., Maybery, D., Bui, M. V., Gordon, I., et al. (2024). Promoting resilience in mental health nurses: a partially clustered randomised controlled trial. Int. J. Nurs. Stud. 159:104865. doi: 10.1016/j.ijnurstu.2024.104865
French, K. A., and Allen, T. D. (2020). Episodic work-family conflict and strain: a dynamic perspective. J. Appl. Psychol. 105, 863–888. doi: 10.1037/apl0000470
Galanis, P., Moisoglou, I., Papathanasiou, I. V., Malliarou, M., Katsiroumpa, A., Vraka, I., et al. (2024). Association between organizational support and turnover intention in nurses: a systematic review and meta-analysis. Healthcare (Basel) 12:291. doi: 10.3390/healthcare12030291
Gan, Q., Yuan, Y., Li, Y., Gan, Q.-W., Yuan, Y.-L., Li, Y.-P., et al. (2025). Work-family behavioral role conflict and daytime sleepiness on suicide risk among Chinese nurses: a cross-sectional study. World J. Psychiatry 15:100859. doi: 10.5498/wjp.v15.i2.100859
Gehri, B., Bachnick, S., Schwendimann, R., and Simon, M. (2023). Work-schedule management in psychiatric hospitals and its associations with nurses' emotional exhaustion and intention to leave: a cross-sectional multicenter study. Int. J. Nurs. Stud. 146:104583. doi: 10.1016/j.ijnurstu.2023.104583
Getie, A., Ayenew, T., Amlak, B. T., Gedfew, M., Edmealem, A., and Kebede, W. M. (2025). Global prevalence and contributing factors of nurse burnout: an umbrella review of systematic review and meta-analysis. BMC Nurs. 24:596. doi: 10.1186/s12912-025-03266-8
Grzywacz, J. G., Frone, M. R., Brewer, C. S., and Kovner, C. T. (2006). Quantifying work-family conflict among registered nurses. Res. Nurs. Health 29, 414–426. doi: 10.1002/nur.20133
Guo, J., Abu Talib, M., Guo, B., Ren, J., and Liu, J. (2025). The mediating role of satisfaction with life and social interaction anxiety in the relationship between loneliness and regulatory emotional self-efficacy. Behav. Sci. 15:392. doi: 10.3390/bs15030392
Hann, M., Reeves, D., and Sibbald, B. (2011). Relationships between job satisfaction, intentions to leave family practice and actually leaving among family physicians in England. Eur. J. Pub. Health 21, 499–503. doi: 10.1093/eurpub/ckq005
Hanoum, F. C., Rizan, M., and Handaru, A. W. (2024). The role of work-life balance, work stress, and work-family conflict in creating job satisfaction and their impact on turnover intention in nurses at level I health facilities Hospital in Karawang. Crit. Care Nurs. Q. 47, 346–369. doi: 10.1097/CNQ.0000000000000527
He, F., Zhang, C., Zhao, D., Zhang, C., Zeng, L., Liu, W., et al. (2024). A questionnaire on the current situation of resignation intention of nurses in intensive care units in China - a descriptive quantitative study. Nurs. Open 11:e2231. doi: 10.1002/nop2.2231
Huaman, N., Morales-García, W. C., Castillo-Blanco, R., Saintila, J., Huancahuire-Vega, S., Morales-García, S. B., et al. (2023). An explanatory model of work-family conflict and resilience as predictors of job satisfaction in nurses: the mediating role of work engagement and communication skills. J. Prim. Care Community Health 14:17195756. doi: 10.1177/21501319231151380
Huang, X., Liao, Y., and Sun, Y. (2020). The impact of work-family conflict on the professional identity and intention to leave for a second child nurse. Shanxi Med. J. 49, 2761–2764. doi: 10.3969/j.issn.0253-9926.2020.20.013
Hwang, E., and Yu, Y. (2021). Effect of Sleep Quality and Depression on Married Female Nurses’ Work-Family Conflict. Int. J. Environ. Res. Public Health. 18:7838. doi: 10.3390/ijerph18157838
Jiang, D., and Kira, J. (2024). New nurse turnover intention and related factors in Japan and China: focusing on nursing practice environment and burnout. JNR 32:e346. doi: 10.1097/jnr.0000000000000636
Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., Rosenthal, R. A., and John, W. (1964). Organizational stress: Studies in role conflict and ambiguity. John Wiley & Sons.
Kang, H., and Ahn, J. (2021). Model setting and interpretation of results in research using structural equation modeling: a checklist with guiding questions for reporting. Asian Nurs. Res. 15, 157–162. doi: 10.1016/j.anr.2021.06.001
Karahan Kaplan, M., Bozkurt, G., Aksu, B. Ç., Bozkurt, S., Günsel, A., and Gencer Çelik, G. (2024). Do emotional demands and exhaustion affect work engagement? The mediating role of mindfulness. Front. Psychol. 15:1432328. doi: 10.3389/fpsyg.2024.1432328
Li, Y. (2018). A survey on occupational burnout and turnover tendency of female emergency medical staff after the two-child policy. Chinese Health Stand. Manage. 9, 6–8. doi: 10.3969/j.issn.1674-9316.2018.16.004
Li, J., and Li, D. (2000). A study on the correlation between role conflict, organizational commitment, and turnover intention in matrix-structured organizations —— Using employees of the industrial technology research institute as a case study [D]. Hsinchu: Institute of Management Science, National Chiao Tung University.
Lin, Y., and Li, J. (2025). Association patterns of work-family balance, perceived stress and turnover intentions among female medical workers in China (2022-2023): integration network and pathway analysis. Archives of public health = Archives belges de sante publique 83:149. doi: 10.1186/s13690-025-01631-2
Ling, C., Tao, L., Wang, X., Wu, Y., Chai, Y., and Zhang, L. (2024). Analysis of turnover intention and influencing factors among female nurses with two children in grade a tertiary public hospitals in Sichuan province: a cross-sectional study. Front. Public Health 12:1416215. doi: 10.3389/fpubh.2024.1416215
Meng, S., Bai, B., Bai, C., Shrestha, S., and Ren, Y. Z. (2024). Invalidating environment and meaning in life: the chain mediating effects of regulatory emotional self-efficacy and basic psychological needs satisfaction. Child Abuse Negl. 151:106736. doi: 10.1016/j.chiabu.2024.106736
Michaels, C. E., and Spector, P. E. (1982). Causes of employee turnover: a test of the Mobley, Griffeth, hand, and Meglino model. J. Appl. Psychol. 67, 53–59. doi: 10.1037/0021-9010.67.1.53
Moloney, W., Boxall, P., Parsons, M., and Cheung, G. (2018). Factors predicting registered nurses' intentions to leave their organization and profession: a job demands-resources framework. J. Adv. Nurs. 74, 864–875. doi: 10.1111/jan.13497
Murtaza, G., Mousa, M., and Roques, O. (2021). Understanding the role of self-efficacy in moderating work-family interface and emotional exhaustion. Hum. Syst. Manag. 40, 353–364. doi: 10.3233/HSM-190785
Netemeyer, R. G., Boles, J. S., and Mcmurrian, R. (1996). Development and validation of work-family conflict and family-work conflict scales. J. Appl. Psychol. 81, 400–410. doi: 10.1037/0021-9010.81.4.400
Pace, F., Sciotto, G., Randazzo, N. A., and Russo, L. (2023). The moderating role of age and seniority on nurses' emotional dissonance and perceived health. Nurs. Health Sci. 25, 619–627. doi: 10.1111/nhs.13053
Pajakoski, E., Leino-Kilpi, H., Stolt, M., Čartolovni, A., and Suhonen, R. (2025). Nurses' justifications for morally courageous acts in ethical conflicts: a narrative inquiry. Nurs. Ethics 32, 752–766. doi: 10.1177/09697330241284357
Rasheed, M. I. (2019). Work-family conflict and turnover intentions among Chinese nurses: the combined role of job and life satisfaction and perceived supervisor support. Pers. Rev. doi: 10.1108/PR-01-2019-0017
Ren, H., Li, P., Xue, Y., Xin, W., Yin, X., and Li, H. (2024). Global prevalence of nurse turnover rates: a meta-analysis of 21 studies from 14 countries. J. Nurs. Manag. 2024:5063998. doi: 10.1155/2024/5063998
Şahin, M., Bektaş, G., Nal, M., and Küçükkurt, A. C. (2025). The mediating role of job satisfaction related to nurse-nurse collaboration and turnover intention. BMC Nurs. 24:235. doi: 10.1186/s12912-025-02869-5
Song, S., Deng, Y., and Dong, L. (2023). Research progress on the application of self-determination theory in nursing. Psychological Monthly 18, 224–226. doi: 10.19738/j.cnki.psy.2023.19.069
Spitzen, T. L., Tull, M. T., and Gratz, K. L. (2022). The roles of emotion regulation self-efficacy and emotional avoidance in self-injurious thoughts and behaviors. Arch. Suicide Res. 26, 595–613. doi: 10.1080/13811118.2020.1818654
Sun, W., Wu, Y., Gao, J., and Guo, C. (2023). Chinese adaptation and validation of the work-home behavioral role conflict scale. Chin. J. Nurs. 58, 1787–1792. doi: 10.3761/j.issn.0254-1769.2023.14.017
Wang, T., Chen, Y., He, Y., Shen, L., Wang, P., Zhu, M., et al. (2022). Stressors, coping strategies and intention to leave the nursing profession: a cross-sectional nationwide study in China. J. Nurs. Manag. 30, 4398–4408. doi: 10.1111/jonm.13865
Wang, H., Ding, P., Dai, M., Xu, M., Li, Y., Zhang, W., et al. (2023). The mediating role of work-family conflict in the relationship between emotional regulation self-efficacy and stress perception among infectious disease nurses. Guangxi Med. J. 45, 2535–2540. doi: 10.11675/j.issn.0253-4304.2023.20.22
Wu, D., Wang, Y., Lam, K. F., and Hesketh, T. (2014). Health system reforms, violence against doctors and job satisfaction in the medical profession: a cross-sectional survey in Zhejiang Province, eastern China. BMJ Open 4:e6431. doi: 10.1136/bmjopen-2014-006431
Wu, F., Wu, Y., Ma, Y., Dou, J., Wang, X., Wang, Q., et al. (2022). Analysis of the current situation and influencing factors of childbirth in female nurses. J. Changzhi Med. Coll. 36, 87–91. doi: 10.3969/j.issn.1006-0588.2022.02.003
Xu, G., Zeng, X., and Wu, X. (2023). Global prevalence of turnover intention among intensive care nurses: a meta-analysis. Nurs. Crit. Care 28, 159–166. doi: 10.1111/nicc.12679
Yakusheva, O., Lee, K. A., and Weiss, M. (2024). The nursing human capital value model. Int. J. Nurs. Stud. 160:104890. doi: 10.1016/j.ijnurstu.2024.104890
Yaghoobi, H., Seyed Hashemi, S. G., Bedayat, E., and Babaloo, A. (2019). Mediator role of emotion regulation difficulties in the relationship between behavioral activation system and exercise addiction in bodybuilding athletes. Neuropsychol. 5, 55–74. Available online at: https://journals.pnu.ac.ir/article_57835962dafa906393bb09df96a3f45ecf5c.pdf
Yao, X., Wen, S., Song, Z., Wang, J., Shen, Y., and Huang, X. (2024). Work-family conflict categories and support strategies for married female nurses: a latent profile analysis. Front. Public Health 12:1324147. doi: 10.3389/fpubh.2024.1324147
Yildiz, B., Yildiz, H., and Ayaz Arda, O. (2021). Relationship between work-family conflict and turnover intention in nurses: a meta-analytic review. J. Adv. Nurs. 77, 3317–3330. doi: 10.1111/jan.14846
Yu, X., Kong, X., Qian, M., Zhang, X., Li, L., and Yang, Y. (2024). The relationship between nurses' professional identity and their intention to leave in public hospitals in Jilin Province. Medical & Social Sciences 37, 114–120. doi: 10.13723/j.yxysh.2024.03.018
Zhang, Y., Duffy, J. F., and De Castillero, E. R. (2017). Do sleep disturbances mediate the association between work-family conflict and depressive symptoms among nurses? A cross-sectional study. J. Psychiatr. Ment. Health Nurs. 24, 620–628. doi: 10.1111/jpm.12409
Zhang, F., Lin, C., Li, X., Li, M., Jia, R., Guo, X., et al. (2023). The relationships between burnout, general wellbeing, and psychological detachment with turnover intention in Chinese nurses: a cross-sectional study. Front. Public Health 11:1216810. doi: 10.3389/fpubh.2023.1216810
Zhang, W., Tang, W., Hu, F., Zhang, W. Q., Hu, F. H., Jia, Y. J., et al. (2023). Impact of the National Nursing Development Plan on nursing human resources in China: an interrupted time series analysis for 1978-2021. Int. J. Nurs. Stud. 148:104612. doi: 10.1016/j.ijnurstu.2023.104612
Zhou, Y. Y., Zhang, L. G., and Jin, T. T. (2020). The role of job satisfaction and life satisfaction on the relationship between work-family conflict and nurse turnover intention. Chinese J. Ind. Hyg. Occup. Dis. 38, 778–781. doi: 10.3760/cma.j.cn121094-20191011-00486
Citation: Ling C, Li X, Zhao L, Zhao D, Qiu M and Li Y (2025) The mediating effect of regulatory emotional self-efficacy between work-family conflict and turnover intention in female nurses with two children. Front. Psychol. 16:1603872. doi: 10.3389/fpsyg.2025.1603872
Received: 07 April 2025; Accepted: 28 October 2025; Published: 19 November 2025.
†These authors have contributed equally to this work
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
Chengrong Ling
Xuemei Li
Ling Zhao3
Meng Qiu
