Association between emotional intelligence and turnover intention among nurses in Saudi Arabia: a cross-sectional mediation analysis Authors: Bayan Alilyyani Published: 2026 ABSTRACT AbstractAimThis study aimed to examine the association between emotional intelligence and turnover intention among nurses in Saudi Arabia, as well as the indirect effect through job stress.BackgroundNursing shortages remain a critical global concern. Turnover intention among nurses is a key workforce issue associated with negative organizational and patient outcomes.MethodsA cross-sectional, non-experimental survey design was used. Data were collected from nurses across Saudi Arabia using an online questionnaire. Emotional intelligence, job stress, and turnover intention were measured using validated instruments. Data were analyzed using SPSS and the PROCESS macro.ResultsEmotional intelligence was negatively associated with job stress (B = -0.308, p < 0.001) and turnover intention (direct effect: B = -0.572, p < 0.001). Job stress was positively associated with turnover intention (B = 0.357, p = 0.012). A significant indirect effect of emotional intelligence on turnover intention through job stress was observed (B = -0.110, 95% CI [-0.215, -0.006]), indicating partial mediation.ConclusionEmotional intelligence is associated with turnover intention both directly and indirectly through job stress. However, due to the cross-sectional design, causal relationships cannot be inferred.Keywords: Emotional intelligence, Job stress, Turnover intention, Nursing, Saudi Arabia Introduction The shortage of nurses has become a significant global concern affecting the nursing profession and healthcare systems worldwide [1]. According to the World Health Organization (WHO, 2020), the global demand for healthcare workers is expected to reach 40 million by 2030 [2]. This trend highlights the increasing need for nurses to ensure the delivery of adequate patient care. In Saudi Arabia, turnover intention is recognized as a major workforce issue and has been widely discussed in previous studies [3]. The reported rate of turnover intention among experienced nurses is approximately 20% [4], while among newly graduated nurses it ranges from 12% to 25% [5]. Turnover intention is considered a key contributor to the nursing shortage, as it reflects nurses’ intentions to leave their positions [6]. Previous research in Saudi Arabia has identified several factors associated with turnover intention, including low salary, working in critical care areas, poor organizational communication, and lack of recognition [7, 8]. Emotional intelligence has been identified as one of the factors associated with nurses’ turnover intention [9, 10]. It refers to an individual’s ability to perceive, regulate, and manage their own emotions as well as the emotions of others in ways that guide thinking and behavior [11]. In nursing practice, emotional intelligence is important for managing workplace interactions, particularly in emotionally demanding situations involving patients and their families [12]. Higher levels of emotional intelligence have been associated with positive outcomes such as improved job performance and retention [10, 13], while lower levels have been associated with negative outcomes such as burnout, job stress, and turnover intention [9]. Given the emotionally demanding nature of nursing, the ability to regulate emotions effectively is essential for maintaining professional performance and well-being. Job stress is another important issue within the nursing profession and is widely recognized as a common global challenge [14]. Nursing is often considered a highly stressful profession due to factors such as heavy workloads, high responsibility, and limited autonomy [15]. Elevated levels of job stress have been associated with reduced productivity and lower quality of patient care [14]. Additionally, high stress levels may impair nurses’ ability to perform effectively and may contribute to negative patient outcomes [16, 17]. Job stress has also been associated with increased turnover intention, which may negatively affect both healthcare organizations and patient care outcomes. Although eliminating job stress entirely may not be feasible due to the nature of nursing work, reducing stress levels remains an important goal for healthcare institutions. This study addresses an important issue in nursing practice by examining factors associated with turnover intention among nurses in Saudi Arabia. Specifically, the study aims to examine the association between emotional intelligence and turnover intention, as well as the indirect effect through job stress. The findings may provide insights into how emotional intelligence and job stress relate to nurses’ intentions to leave their jobs. These insights may help inform strategies aimed at supporting nurses and improving workforce stability. Additionally, nurse leaders may benefit from understanding the potential role of emotional intelligence and job stress when developing interventions, such as training programs and organizational strategies, to support nurses in managing workplace demands. Research hypotheses H1 Emotional intelligence is negatively associated with turnover intention among nurses. H2 There is a negative association between emotional intelligence and job stress. H3 There is a positive association between job stress and turnover intention. H4 Emotional intelligence is indirectly associated with turnover intention through job stress. These four hypotheses are illustrated in Fig. 1. Conceptual model Materials and methods Design This study employed a non-experimental, cross-sectional survey design to examine the associations among emotional intelligence, job stress, and turnover intention among nurses. This design was appropriate for addressing the study objectives by allowing the assessment of relationships among variables at a single point in time. Settings Data were collected using an online survey targeting registered nurses working in public healthcare institutions across Saudi Arabia. Participants were recruited using a non-probability convenience sampling approach through professional online networks. The survey link was distributed via commonly used platforms among nurses, including WhatsApp groups, LinkedIn professional communities, and institutional email networks, to facilitate broad geographic reach. Although the survey was distributed nationwide, participation was voluntary and dependent on access to and engagement with these networks. Consequently, the final sample was predominantly composed of nurses from the Western region of Saudi Arabia (75.8%), with limited representation from other regions. Therefore, the sample should be interpreted as regionally concentrated rather than nationally representative. Due to the nature of open online distribution, the total number of individuals who received or viewed the survey could not be determined, and a response rate was not calculated. Sample and participants criteria The sampling frame consisted of nurses accessible through online professional networks during the data collection period. A total of 157 participants were included in the study. The sample size was considered adequate based on commonly cited methodological recommendations suggesting 10–20 participants per predictor variable for regression analyses [18]. Mediation analysis using bootstrapping procedures (e.g., PROCESS macro) is considered robust with moderate sample sizes. However, no a priori sample size or power calculation was conducted, which should be acknowledged as a limitation. A non-probability convenience sampling method was used. The study targeted registered nurses working in public healthcare institutions under the Saudi Ministry of Health. Inclusion criteria Registered nurses currently working in public healthcare settings in Saudi Arabia. Registered nurses currently working in public healthcare settings in Saudi Arabia. At least six months of experience in their current clinical position. At least six months of experience in their current clinical position. Ability to read and understand English. Ability to read and understand English. Exclusion criteria Nurses in managerial or administrative roles. Nurses in managerial or administrative roles. Nurses on leave during the data collection period. Nurses on leave during the data collection period. Nurses with less than six months of experience. Nurses with less than six months of experience. Individuals unable to complete the survey in English. Individuals unable to complete the survey in English. These criteria ensured that participants had sufficient clinical experience and were directly involved in patient care, which is relevant to the study variables. Instruments Three standardized self-report instruments were used to measure emotional intelligence, job stress, and turnover intention. Emotional intelligence Emotional intelligence was assessed using the Schutte Self-Report Emotional Intelligence Test (SSEIT) [19]. The scale consists of 33 items rated on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). Total scores are calculated by summing all items, with higher scores indicating higher emotional intelligence. Three items [18, 29, and 32] were reverse-coded. The total score ranges from 33 to 165. The SSEIT has demonstrated strong psychometric properties in prior studies, with Cronbach’s alpha values typically exceeding 0.85 [20]. In the present study, internal consistency was high (Cronbach’s α = 0.93). Job stress Job stress was measured using the 10-item Perceived Stress Scale (PSS-10) [21]. Items are rated on a 5-point Likert scale (0 = never to 4 = very often). Four items [4, 5, 7, and 8] were reverse-coded. Total scores range from 0 to 40, with higher scores indicating greater perceived stress. The PSS-10 has demonstrated good reliability and validity across healthcare populations [22, 23]. In this study, internal consistency was acceptable (Cronbach’s α = 0.80). Turnover intention Turnover intention was measured using the Turnover Intention Scale developed by Kelloway et al. (1999) [24]. The scale includes four items rated on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). Higher scores indicate greater intention to leave the current job. The scale has demonstrated good reliability and validity in previous nursing research [25]. In the present study, internal consistency was good (Cronbach’s α = 0.88). Psychometric properties in the present study Internal consistency reliability for all instruments was assessed using Cronbach’s alpha. Construct validity was examined through correlation analysis among study variables, which showed relationships consistent with theoretical expectations, providing preliminary support for validity in the current sample. Data collection Following ethical approval, data were collected using an online survey distributed via LinkedIn, WhatsApp, and email networks. Study information was provided on the first page of the survey. Participants indicated their voluntary consent before proceeding. Only individuals who met the inclusion criteria were able to participate. Data analysis Data were analyzed using SPSS version 28.0.1.1 (SPSS Inc., Chicago, IL, USA). Descriptive statistics (frequencies and percentages) were used to summarize demographic characteristics. Means and standard deviations were calculated for the main study variables. Internal consistency reliability was assessed using Cronbach’s alpha. Pearson correlation analysis was conducted to examine bivariate relationships among emotional intelligence, job stress, and turnover intention. Simple linear regression analyses were performed to test the hypothesized relationships. Mediation analysis was conducted using the PROCESS macro (Model 4) developed by Andrew F. Hayes [26], with 5,000 bootstrap resamples to estimate the indirect effect. Unless otherwise specified, results are reported as unstandardized regression coefficients (B), with corresponding standard errors, 95% confidence intervals, and p-values. Standardized coefficients (β) are presented in Fig. 2 and in the β column of Table 4 to facilitate comparison across variables. For clarity, unstandardized coefficients (B) are interpreted in the text, whereas standardized coefficients (β) are provided for comparative purposes. Standardized path coefficients (β) for the mediation model. Note Values represent standardized regression coefficients (β). *p < 0.05 Table 4Mediation analysis resultsPathBSE95% CI (Lower, Upper)βp-value Indirect Effect EI → Job Stress → Turnover Intention-0.1100.053[-0.215, -0.006]-0.0540.039 Direct Effects EI → Job Stress-0.3080.084[-0.474, -0.143]-0.28< 0.001Job Stress → Turnover Intention0.3570.143[0.078, 0.637]0.190.012EI → Turnover Intention-0.5720.157[-0.880, -0.264]-0.28< 0.001 Total Effect EI → Turnover Intention-0.6820.154[-0.985, -0.380]-0.33< 0.001Note: B = unstandardized regression coefficient; SE = standard error; CI = confidence interval; β = standardized coefficient. Indirect effects were estimated using 5,000 bootstrap resamples. EI = emotional intelligence Mediation analysis results Note: B = unstandardized regression coefficient; SE = standard error; CI = confidence interval; β = standardized coefficient. Indirect effects were estimated using 5,000 bootstrap resamples. EI = emotional intelligence Regression assumptions, including linearity, normality of residuals, independence of errors, and absence of multicollinearity, were assessed prior to analysis, and no serious violations were detected. No missing data were present, as all survey items were mandatory. Although nationality and level of education were significantly associated with turnover intention, they were not included as covariates in the mediation model to avoid over-adjustment and to preserve statistical power given the sample size. Future studies with larger samples are recommended to include these variables. Ethical considerations This study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Scientific Research Ethics Committee at Taif University (IRB No: 46–285; Date: 2025-04-17). Participants were provided with detailed information about the study and gave informed consent prior to participation. Participation was voluntary, and confidentiality and anonymity were ensured. No identifying information (e.g., names, hospital affiliations, or contact details) was collected. Participants were informed of their right to withdraw at any time without consequences. All data were stored securely and were accessible only to the researcher. Results The demographic characteristics of the participants are presented in Table 1. The largest proportion of participants was aged 24–29 years (32.5%), followed by 30–35 years (31.2%). Participants aged 36–41 years represented 17.8%, while 18.5% were older than 41 years. The majority of participants were female (72.6%), and most were non-Saudi (76.4%). Table 1Characteristics of participants (N = 157)DemographicsNumberPercent Age 24–295132.5% 30–354931.2% 36–412817.8% Above 412918.5% Gender Female11472.6% Male4327.4% Nationality Saudi3723.5% Non-Saudi12076.4% Level of Education Bachelor of Nursing10063.7% Master Degree in Nursing74.5% Nursing Diploma4931.2% PhD Degree in Nursing10.6% Years of Experience 1–3 years2314.6% 4–5 years2214.0% Less than a year2314.6% More than 5 years8956.7% Region Central region74.5% Eastern region159.6% Northern region95.7% Southern region74.5% Western region11975.8% Characteristics of participants (N = 157) In terms of education, 63.7% of participants held a bachelor’s degree in nursing, 31.2% held a diploma, 4.5% held a master’s degree, and 0.6% held a PhD. Regarding work experience, more than half of the participants (56.7%) had more than five years of experience, while smaller proportions had 1–3 years (14.6%), less than one year (14.6%), or 4–5 years (14.0%) of experience. The majority of participants were from the Western region of Saudi Arabia (75.8%), with limited representation from other regions. Descriptive statistics and reliability estimates for the main study variables are presented in Table 2. Emotional intelligence had the highest mean score (M = 3.92, SD = 0.55), indicating relatively higher levels among participants. Job stress showed a moderate mean score (M = 2.47, SD = 0.67), while turnover intention had the lowest mean score (M = 2.28, SD = 1.13). Table 2Descriptive statistics and reliability of study variablesVariablesNumber of ItemsMeanStandard DeviationCronbach’s AlphaEmotional Intelligence333.920.550.932Job Stress102.470.670.803Turnover Intention42.281.130.883 Descriptive statistics and reliability of study variables All scales demonstrated acceptable to excellent internal consistency. Emotional intelligence showed excellent reliability (Cronbach’s α = 0.93), while job stress (α = 0.80) and turnover intention (α = 0.88) demonstrated good reliability. The associations between demographic variables and the main study variables are presented in Table 3. Most demographic variables were not significantly associated with emotional intelligence, job stress, or turnover intention. Table 3Associations between demographic variables and study variablesDemographicsStudy VariablesEmotional IntelligenceJob StressTurnover Intention Gender t = 0.513, p = 0.60t = 0.86, p = 0.39t =-0.37, p = 0.71 Age F = 0.73, p = 0.53F = 1.18, p = 0.32F = 0.50, p = 0.67 Nationality t = 0.78, p = 0.86t =-0.56, p = 0.44t =-0.46, p = 0.01 Level of Education F = 0.85, p = 0.46F = 0.95, p = 0.41F = 4.08, p = 0.008 Years of Experience F = 1.61, p = 0.83F = 0.29, p = 0.82F = 1.65, p = 0.18 Region F = 0.25, p = 0.90F = 2.01, p = 0.09F = 0.87, p = 0.48Note. Independent samples t-tests were used for dichotomous variables (e.g., gender, nationality), and one-way ANOVA (F-test) was used for variables with more than two categories (e.g., age, education, experience, and region). Statistical significance was set at p < 0.05 Associations between demographic variables and study variables Note. Independent samples t-tests were used for dichotomous variables (e.g., gender, nationality), and one-way ANOVA (F-test) was used for variables with more than two categories (e.g., age, education, experience, and region). Statistical significance was set at p < 0.05 However, nationality was significantly associated with turnover intention (t = -0.46, p = 0.01), and level of education was also significantly associated with turnover intention (F = 4.08, p = 0.008). No other statistically significant associations were observed. The results of the hypothesized model The results of the mediation model are presented in Fig. 2; Table 4. Emotional intelligence was significantly and negatively associated with job stress (B = -0.308, SE = 0.084, 95% CI [-0.474, -0.143], p < 0.001). Job stress was significantly and positively associated with turnover intention (B = 0.357, SE = 0.143, 95% CI [0.078, 0.637], p = 0.012). Emotional intelligence also showed a significant negative direct association with turnover intention (B = -0.572, SE = 0.157, 95% CI [-0.880, -0.264], p < 0.001). The indirect effect between emotional intelligence and turnover intention through job stress was significant (B = -0.110, SE = 0.053, 95% bootstrap CI [-0.215, -0.006]). As the confidence interval did not include zero, this indicates a statistically significant indirect effect. The total effect of emotional intelligence on turnover intention was also significant (B = -0.682, SE = 0.154, 95% CI [-0.985, -0.380], p < 0.001). These findings indicate that job stress partially explains the association between emotional intelligence and turnover intention. Discussion This study examined the associations among emotional intelligence, job stress, and turnover intention among nurses in Saudi Arabia, with particular attention to the indirect effect through job stress. The findings indicate that emotional intelligence was significantly associated with both job stress and turnover intention, while job stress was positively associated with turnover intention. In addition, the results suggest that job stress may partially account for the relationship between emotional intelligence and turnover intention. From a conceptual perspective, these findings are consistent with the view that emotional intelligence may function as an individual resource associated with how nurses perceive and respond to workplace demands. Nurses with higher emotional intelligence may be better able to regulate their emotional responses, interpret challenging situations more effectively, and apply adaptive coping strategies. In turn, this may be associated with lower perceived job stress, which in turn is linked to reduced turnover intention. However, given the cross-sectional design, these interpretations should be considered theoretical and not indicative of directional or causal relationships. The observed negative association between emotional intelligence and turnover intention aligns with previous studies conducted in different contexts, including Greece and China [9, 27]. These findings suggest that emotional intelligence may be an important factor associated with workforce stability. At the same time, variability in findings across studies, including those reporting different patterns of association [28], indicates that this relationship may be influenced by contextual factors such as organizational culture, leadership practices, and work environment. Therefore, emotional intelligence should be understood as one component within a broader set of individual and organizational factors associated with turnover intention. The negative association between emotional intelligence and job stress further highlights the potential relevance of emotional competencies in shaping how nurses experience workplace demands. Nurses with higher emotional intelligence may be more capable of identifying stressors, managing emotional responses, and engaging in effective coping strategies. This interpretation is consistent with prior research linking emotional intelligence with lower perceived stress and improved psychological well-being [29–33]. However, it is also possible that nurses experiencing lower stress levels may report higher emotional intelligence, reflecting the complex and potentially reciprocal nature of these variables. Consistent with existing literature, job stress was positively associated with turnover intention, reinforcing the importance of workplace conditions in shaping nurses’ intentions to remain in or leave their positions. High levels of job stress have been associated with emotional exhaustion, reduced job satisfaction, and decreased organizational commitment, all of which are factors linked to turnover intention. These findings are consistent with studies conducted both within Saudi Arabia and internationally [34–36]. The mediation analysis suggests that job stress may partially explain the association between emotional intelligence and turnover intention. This finding contributes to the literature by providing a more integrated perspective on how these variables are interrelated. Rather than operating independently, emotional intelligence and job stress appear to be linked within a pathway associated with turnover intention. However, this should be interpreted as an indirect statistical association rather than evidence of a causal mechanism. Longitudinal and experimental studies are required to determine whether changes in emotional intelligence or job stress are followed by changes in turnover intention over time. Within the Saudi Arabian context, these findings provide additional insight into factors associated with nursing workforce challenges. Given ongoing concerns regarding nurse retention in the region, understanding both individual and organizational factors is essential. Previous research in Saudi Arabia has emphasized the role of workplace stress and organizational conditions in relation to turnover intention [34]. The present findings suggest that emotional and psychological dimensions of work may also be relevant. However, the predominance of participants from the Western region limits the extent to which these findings can be generalized to other regions. Overall, this study extends existing research by examining emotional intelligence, job stress, and turnover intention within a single analytical framework. By highlighting the associations among these variables, the findings contribute to a more nuanced understanding of factors related to nurses’ turnover intention. Nevertheless, the results should be interpreted with caution due to methodological limitations, particularly the cross-sectional design and the use of non-probability convenience sampling. Implications for nursing practice The findings of this study indicate that emotional intelligence and job stress are significantly associated with nurses’ turnover intention. These results suggest that both individual and work-related factors may be relevant when addressing workforce stability in healthcare settings. From a practical perspective, healthcare organizations may benefit from supporting nurses’ emotional competencies while also addressing workplace stressors. For example, initiatives such as emotional intelligence training programs, stress management interventions, and supportive leadership practices may help nurses better manage workplace demands. In addition, organizational strategies aimed at reducing job stress—such as workload management, adequate staffing, and the promotion of supportive work environments—may be associated with lower turnover intention. Although the present findings do not establish causal relationships, they highlight areas that may be considered when developing strategies to support nurse retention. Implications for future research The findings of this study suggest several directions for future research. First, further studies are needed to examine factors associated with nurses’ turnover intention, including additional psychological, organizational, and environmental variables. Second, future research should explore potential mediating and moderating variables that may help explain the relationships among emotional intelligence, job stress, and turnover intention. Third, given the limitations of cross-sectional designs, longitudinal and experimental studies are recommended to better understand the temporal relationships among these variables. In particular, intervention studies evaluating the effectiveness of emotional intelligence training programs may provide more robust evidence. Additionally, qualitative research may offer deeper insights into how nurses perceive emotional intelligence and how it relates to their work experiences and intentions to leave. Finally, comparative studies across public and private healthcare settings, as well as across different regions, may enhance the generalizability of findings. Strengths and limitations This study has several strengths. It contributes to the existing literature by examining the associations among emotional intelligence, job stress, and turnover intention within a single analytical framework. It also provides context-specific insights into the Saudi Arabian healthcare system. Furthermore, the study utilized standardized instruments with established reliability, and internal consistency in the present sample ranged from acceptable to high. However, several limitations should be acknowledged. First, the cross-sectional design limits the ability to infer causal or temporal relationships among variables. Second, the use of non-probability convenience sampling and online recruitment may introduce self-selection bias and limit representativeness. Third, the sample was predominantly drawn from the Western region of Saudi Arabia, which may limit the generalizability of the findings to other regions. Fourth, data were collected using self-report measures, which may be subject to common method bias. Additionally, restricting participation to nurses who were able to complete the survey in English may have excluded some potential participants, further affecting generalizability. Finally, the study was limited to nurses working in public healthcare institutions, and the findings may not be fully generalizable to private sector settings. Conclusions This study examined the associations among emotional intelligence, job stress, and turnover intention among nurses working in Saudi Arabia. The findings indicate that emotional intelligence was negatively associated with both job stress and turnover intention, while job stress was positively associated with turnover intention. In addition, job stress showed a significant indirect effect in the relationship between emotional intelligence and turnover intention. These findings contribute to the existing literature by providing a more integrated understanding of how emotional intelligence and job stress are related to turnover intention. However, due to the cross-sectional design, the results should be interpreted as associations rather than causal relationships. Overall, the study highlights the potential relevance of both emotional and work-related factors in understanding nurses’ turnover intention. Addressing these factors may be useful for healthcare organizations aiming to improve workforce stability, although further research—particularly longitudinal and experimental studies—is needed to establish directionality and inform intervention development. Acknowledgements The authors would like to acknowledge the Deanship of Graduate Studies and Scientific Research, Taif University for funding this work. Abbreviations Emotional Intelligence 10–Perceived Stress Scale (10–item version) Schutte Self–Report Emotional Intelligence Test Statistical Package for the Social Sciences Unstandardized regression coefficient Standardized regression coefficient Standard error Confidence interval Standard deviation Author contributions BA conceptualized and deigned the study. BA collected and analyzed data. BA reviewed and edited the manuscript. Author read and approved the final manuscript. Funding The author would like to acknowledge the Deanship of Graduate Studies and Scientific Research, Taif University for funding this work. Data availability The data supporting the findings of this study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate Ethical approval for this study was obtained from the Scientific Research Ethics Committee at Taif University (IRB No: 46–285; Date: 2025-04-17). The study was conducted in accordance with the Declaration of Helsinki and the regulations of the National Committee of Bioethics in Saudi Arabia. All participants were provided with detailed information about the study and gave informed consent prior to participation. Participation was voluntary, and confidentiality and anonymity were ensured. Participants were informed of their right to withdraw from the study at any time without any consequences. No patient or public contribution This study examined the association between emotional intelligence and turnover intention among nurses working in Saudi Arabia. Therefore, no patients or members of the public were involved in the design, conduct, reporting, or dissemination of this study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data supporting the findings of this study are available from the corresponding author upon reasonable request.