Mediating Effect Of Stress Response and Grit On the Relationship Between Stress and Depression In Nursing Students: Emperical Research Quantitative Authors: Jaehee Jeon, Seunghye Choi Published: 2024 ABSTRACT ABSTRACTAimsTo explore the prevalence of depression amongst Korean nursing students and to identify the mediating effect of stress response and grit in the relationship between stress and depression in the same population.DesignA total of 184 nursing students from Korea were included. This study was cross‐sectional with a structured questionnaire that included stress, stress response, grit, and depression.MethodsThe study participants were university students from four nursing colleges located in the metropolitan area and Gangwon‐do in South Korea. The data collection period spanned from February 24 to March 30, 2023, using an online questionnaire following the G University's Institutional Review Board's (IRB's) deliberation.ResultsAmongst the participants, the prevalence of depression was 23.8% (95% CI: [17.7%, 29.9%]), and 13.7% had moderate depression, whereas 10.1% had severe depression. The direct effects of stress on stress response (t = 9.26, p < 0.001) and from stress to grit (t = 22.82, p < 0.001) were statistically significant. The direct effects from stress to depression (t = 2.61, p = 0.009), from stress response to depression (t = 9.68, p < 0.001), and from grit to depression (t = −2.09, p = 0.039) were also statistically significant. The association between stress and depression was mediated by stress response but not grit.ConclusionsThe prevalence of depression amongst Korean nursing students was high. Preparing strategies against depression amongst nursing students is an urgent matter. The mediating effect of stress response in the relationship between stress and depression was statistically significant, but that for grit was not. Therefore, it is necessary to investigate students' stress responses and provide personalised interventions, as well as reduce established stressors.Implications for the ProfessionStrategies to manage stress responses can be used to reduce depression caused by stress in nursing students.ImpactTo reduce depression in nursing students, it is necessary to investigate strategies to understand stress response, provide personalised interventions, and reduce already‐known stressors.Reporting MethodThe authors have adhered to relevant EQUATOR guidelines using the STROBE checklist: cross‐sectional studies.Patient or Public ContributionNursing students were the subjects of data collection for this study and participated in the survey upon voluntary consent to participation.Keywords: depression, grit, nursing, stress, students Introduction As a broad psychological state, depression is a state of mind that can adversely affect a person's response, attitude, and thoughts towards loss or negative experience, as well as personal relationships and social activities, encompassing simple sadness to continuous feelings of loss and helplessness (Beck and Beck 1972). In other words, depression is an emotion that all humans can commonly experience (Beck, Steer, and Brown 1987). Depression can cause physical, psychological, and intellectual dysfunction beyond mere experiences of mood and can lead to difficulties in interpersonal relationships or social life (Beck and Beck 1972). According to the World Health Organisation, it has been reported that approximately 4.4% of the world's population suffers from depression (WHO 2017). However, a previous study found that 19.2% of nursing students had moderate or severe levels of depression (Facioli et al. 2020). University students are in a transition stage from adolescence to adulthood, during which ego‐identity is developed, and they may experience depression due to confusion and conflict as autonomy and responsibility for such autonomy increase following a disciplined way of life (Sung and Kim 2018). Nursing students are known to experience higher levels of stress than students in other majors due to the relatively more rigorous curriculum, an excessive amount of learning, the burden of clinical practice with real patients in unfamiliar clinical settings, and the pressure of national nursing examination preparation (Koh and Jeon 2022). According to a recent study, nursing students have a higher incidence rate of depression compared to university students in general and, in particular, higher than that before the COVID‐19 pandemic (Koh and Jeon 2022). Depression is known to affect academic performance and burnout in nursing students (Abu Ruz, Al‐Akash, and Jarrah 2018; Njim et al. 2020). Depressed nursing students have been reported to miss classes more often and receive lower cumulative grade point averages than those without depression (Abu Ruz, Al‐Akash, and Jarrah 2018). Therefore, it is necessary to examine the degree of depression in nursing students continuously and to understand related factors and their relationships to reduce depression. Background Depression in nursing students is known to be affected by various factors such as economic status (Song, Lee, and Jung 2019), satisfaction regarding major (Byun, Kim, and Lee 2020), health status, academic performance, and satisfaction regarding university life (Koh and Jeon 2022). Amongst these, living conditions or societal events can affect depression. Various psychological stress, such as experiences of adverse events and alienation from groups to which one belongs, can cause depression (Kwon 2016). The relationship between stress and depression is based on the stress exposure model (Cole et al. 2006), according to which continuous exposure to the stress of adverse life events causes depression. The stress process model is a well‐known framework that explains the mechanisms by which stressors lead to health outcomes (Pearlin et al. 1981). In this model, it is important to distinguish between three components of stress: sources, outcomes, and mediators (Pearlin et al. 1981; Hish et al. 2019). The model consisted of three elements of stress. According to the stress process model (Pearlin et al. 1981), psychological issues induced by stress are affected through a process with various internal and external factors of individuals rather than by a single stress factor. These internal and external factors serve as mediators. Social support and self‐efficacy (Sharma and Kumra 2022) are known to serve mediating roles in the relationship between stress and depression in nursing students. However, attempts must be made to identify the mediating effects of more diverse factors. An individual has a finite capacity to manage stress. In other words, the use of energy and adaptability to stress depends on the genetic character, attitudes, and lifestyles, and stress responses can vary depending on the individual's cognitive assessment of the stress, even when experiencing similar stress situations or stress levels (Han 2020). Some nurses become dissatisfied with their jobs or transfer to different jobs because they cannot endure the negative emotions caused by excessive job stress and various difficulties in the nursing practice (Hayes et al. 2006). In contrast, others overcome negative psychological factors or challenging situations to build a satisfactory career and continue to develop as experts (Kim and Park 2020). That is, individual stress responses under stressful situations are manifested differently. Studies have been attempted on the relationship between stress responses and depression, such as whether individual depression has a moderating effect on stress response (Kim and Lee 2019) and to determine if stress responses and depression are positively correlated (Park et al. 2008). However, although numerous studies have examined stress in nursing students, few studies have investigated the relationship between stress and stress response. It was difficult to find studies that elucidate how stress response affects the relationship between depression and stress. Grit is a personal trait whereby one persistently strives to achieve long‐term goals and is defined as the continuous effort and undertaking of challenges with endurance and enthusiasm (Duckworth et al. 2007). Interest in grit has recently been increasing, and the U.S. Department of Education has selected grit as a critical factor for living in the 21st century (SRI International 2018). The importance of effort is emphasised because grit plays a significant role in achieving and succeeding in an individual's goals and can always change extrinsically. Therefore, grit significantly affects future career choices, career plans, and efforts to earn a desired job for most university students going through young adulthood. Grit is an essential competency for nursing students because to achieve long‐term goals, constant efforts and passion must support the university years and enable them to become successful nurses after graduation (Oh, Baek, and Jang 2022). The higher the grit of nursing students, the lower the stress (Kang and Choi 2022). Grit was also found to have a mediating effect on the stress and adaptation to college life of nursing students (Oh, Baek, and Jang 2022). Furthermore, grit and depression were found to be negatively correlated (Shin 2020). However, few studies have identified the effect of grit on the relationship between stress and depression in nursing students. According to the review of previous studies above, the theoretical framework of this study is based on the three components of the stress process model: stress (sources), depression (health outcomes), and mediators (stress response, grit). The Study Aims and Objectives This study aims to explore the prevalence of depression amongst Korean nursing students, to identify the relationship between stress, stress response, grit, and depression in the same population, and to identify the mediating effect of stress response and grit in the relationship between stress and depression. Specifically, we hypothesised that stress response and grit would mediate the relationship between stress and depression in nursing students. Methods Design This is a cross‐sectional study. The hypothetical model of this study is presented in Figure 1. Paths a1, a2, b1, b2, and c represent the direct effects of each variable. M represents the mediating effect, where between stress and depression, M1 is the path for the mediating effect of stress response, and M2 is the path for the mediating effect of grit. Path diagram of the mediating effects of stress response and grit on the relationship between stress and depression. Instrument With Validity and Reliability For stress, the tool used on the scale of stress in university life was developed by Chon, Kim, and Yi (2000) for university students. The tool consists of 50 questions in eight sub‐areas, with seven questions for academics, seven questions for the economy, eight questions for the future, five questions for values, six questions for different sexes, five questions for friends, six questions for family, and six questions for professors. Each question was rated on a Likert scale ranging from one point for “not stressed at all” to five for “highly often stressed,” where the higher the score, the higher the stress level. The reliability of this tool at the time of its development was Cronbach's α = 0.97 (Chon, Kim, and Yi 2000) and 0.96 in this study. A stress response scale developed by Koh et al. (2001) for university students was used for stress response. The tool consists of 39 questions in seven sub‐areas, with six questions for tension, four questions for aggression, three questions for somatization, six questions for anger, eight questions for depression, five questions for fatigue, and seven questions for frustration. Each question was rated on a Likert scale ranging from one point for “not at all” to five points for “very much so,” where the higher the score, the higher the stress response. The reliability of this tool at the time of its development was Cronbach's α = 0.97 (Koh et al. 2001), and 0.97 in this study. The Korean Version of GRIT (K‐GRIT), developed by Lee et al. (2021), was used for grit. The tool consists of 20 questions in three sub‐areas, with 10 questions for effort–persistence, five for passion–enjoyment, and five for passion–meaning. Each question was rated on a Likert scale ranging from one point for “not at all” to five points for “very much so,” where the higher the score, the higher the grit towards long‐term goals. The reliability of this tool at the time of its development was Cronbach's α =0.84 ~ 0.91 (Lee et al. 2021) and.95 in this study. The tool used for depression was one modified and supplemented from the existing Beck Depression Inventory‐II (BDI‐II) by Beck, Steer, and Brown (1987) in accordance with the criteria for diagnosing depression in the Diagnostic and Statistical Manual of Mental Disorders, adapted into Korean by Lee and Song (1991). The tool consists of 21 questions in three sub‐areas, with 10 questions in the emotional aspect, six in the cognitive aspect, and five in the physical aspect. Each question was represented in a score range of zero to 63 using the sum of the scores of the 21 questions on a four‐point Likert scale of zero to three points. For each BDI‐II score section, zero to 13 points referred to “normal,” 14 to 19 referred to “mild depression,” 20 to 28 referred to “moderate depression,” and 29 to 63 referred to “severe depression.” In non‐clinical populations, scores exceeding 20 suggest the presence of depressive symptoms. The reliability of this tool was Cronbach's α = 0.89 (Lee and Song 1991) and 0.95 in this study. Sampling and Recruitment The participants of this study were university students from four nursing colleges located in the metropolitan area and Gangwon‐do in South Korea. All four nursing colleges have been accredited for five years by the Korea Nursing Education Accreditation Board, proving the excellence of their curriculum and learning outcomes, and these colleges maintain similar levels of education. Convenience sampling was employed. While a truly random sample could not be achieved due to the recruitment process, efforts were made to include a broad and representative sample of students. Participation was voluntary, and all eligible students who responded and consented to participate were included in the study. This approach aimed to minimise selection bias and ensure that the sample was as representative as possible given the constraints. Sample Size and Power The sample size for this study was determined using G*Power 3.1.9.2 software. The study's primary objective was to explore the mediating effects of stress response and grit on the relationship between stress and depression. To this end, the sample size was calculated based on an F test (Cohen 1988). We aimed to detect a medium effect size of 0.15 with a significance level of 0.05 and a statistical power of 0.90. This calculation accounted for 15 predictors, including demographic variables as well as stress, stress response, and grit. The medium effect size was selected to ensure that the study was sensitive to meaningful differences and associations between variables. Population and Sample After announcing the study on the online student body community, nursing students who wished to participate voluntarily could click the link by volition for the questionnaire. After connection through the link, it was assumed that they had accepted to participate in the study if they had read the explanation for the study and marked their consent. The survey was structured so the participant could proceed with it upon marking their consent. Inclusion and/or Exclusion Criteria The inclusion and exclusion criteria were determined by referring to a previous study conducted on nursing students (Koh and Jeon 2022). The criteria for inclusion were those aged 19 or older and currently enrolled in the nursing course. The criteria for exclusion were nursing students who were currently or had been in the past diagnosed with depression or taking depression‐related drugs. Participants were screened using initial survey questions to exclude those with a history of depression diagnosis or current use of depression‐related medications. This screening was based on self‐reported data requested at the beginning of the online questionnaire. Data Sources/Collection The data collection period spanned from February 24 to March 30, 2023, using an online questionnaire after the Institutional Review Board's (IRB's) deliberation. Permission to publish the study was obtained from the department heads of the targeted nursing colleges after the study was explained. The study was then advertised through various channels, including official university emails, online student portals, and social media groups associated with nursing students. Interested students were directed to an online platform on which they reviewed detailed information about the study and provided their informed consent. The collected data were anonymized to be used solely for research purposes. All data related to this study will be stored in the researcher's secured folder for 3 years following data analysis and then discarded. Amongst the participants who had responded to the online questionnaire, only those who had consented to the collection of their personal information regarding telephone numbers were given a set number of $10 online coffee coupons. The survey took 10–15 min. Data Analysis Data analysis was performed using SPSS (ver.26) statistical software. An independent t‐test and one‐way ANOVA identified differences in depression according to the demographic characteristics of the participants, and then the Scheffé post hoc test was conducted. Pearson's correlation coefficient analysed the correlations between stress, stress response, grit, and depression. The Kolmogorov–Smirnov test, skewness, and kurtosis were utilised to test the normality of the major variables. Although the Kolmogorov–Smirnov test did not satisfy the normality assumption, it has been shown that this test can be overly sensitive to large sample sizes. Therefore, we also examined the skewness and kurtosis values. According to West, Finch, and Curran (1995), skewness and kurtosis values within the range of −1 to +1 suggest that normality can be assumed. In this study, the skewness and kurtosis values for the major variables were within this range, indicating that the data could be considered approximately normally distributed. A regression analysis was performed using The PROCESS macro for SPSS (Hayes 2013) to verify the mediating effect. In this study, the ranges of skewness and kurtosis of major variables fell within that of normal distribution but did not satisfy the Kolmogorov–Smirnov normality assumption. Therefore, Model Four was used to verify the effect of one parameter in the relationship between independent and dependent variables amongst the 76 path models that Hayes (2013) presented using the PROCESS macro. Model 4 is designed to verify the mediating effect of one variable in the relationship between an independent variable and a dependent variable. This model was selected because it allows for the assessment of the direct and indirect effects of stress on depression through the mediating variables of stress response and grit. We chose Model 4 because it is the most appropriate for testing simple mediation hypotheses, where only one mediator is considered at a time in the analysis. The bootstrapping method was used for the statistical significance of the mediating effect. The 95% percentile confidence interval (boot lower limit‐boot upper limit) was computed with 10,000 bootstrap samples. Bootstrapping is particularly useful in mediation analysis as it does not assume the normality of the sampling distribution of the indirect effect and provides more accurate confidence intervals. Using a large number of bootstrap samples, such as 10,000, enhances the stability and reliability of the estimates (Hayes 2013). This method provides robust estimates of the indirect effects even when the normality assumption is violated, thereby enhancing the reliability of the mediation analysis results. Ethical Considerations The plan and questionnaire for this study were reviewed and approved by the Institutional Review Board (IRB) of G University. Prior to data collection, informed consent was obtained from all participants. Participants were provided with a detailed explanation of the study's purpose, procedures, potential risks, and benefits via an online link at which participants could read the details at their own pace. They were informed that their participation was voluntary and that they could withdraw at any time without any penalty. If participants agreed to take part in the study, they indicated their consent by clicking an “Agree” button at the end of the explanation page. Clicking this button was considered equivalent to providing informed consent. Written consent was therefore implied through this digital agreement process. Additionally, the tools used in this study were utilised after obtaining consent from the original author via email. Several measures were implemented to ensure the confidentiality and anonymity of the participants. All collected data were stored securely on a password‐protected server accessible only to the research team. Participants were assigned unique identification codes to replace personally identifiable information. This process ensured that individual responses could not be traced back to any specific participant. Additionally, any identifying information provided by the participants was separated from the survey data and stored separately to further protect their anonymity. Data analysis was performed on anonymized datasets, and results were reported in aggregate form to prevent the identification of individual participants. Results Differences in Depression According to General Characteristics of Nursing Students The respondents were mostly female (88.0%). The mean age of the participants was 22.3 ± 3.36. The grades were senior (54 students, 29.4%), junior (47 students, 25.5%), sophomore (44 students, 23.9%), and freshman (39 students, 21.2%). The majority of the respondents did not have a religion (116 students, 63.0%), and the most common residence status was “rented room” (74 students, 40.2%). The most common economic status was average (106 students, 57.6%), high (22.3%), and low (37 students, 20.2%). The most common grade point average for the entire university year was 3.5 or higher but lower than 4.0 (89 students, 48.4%), and the most common motivation for applying to the department of nursing was employment rate consideration (84 students, 45.7%). 92.8% of the subjects seemed to be satisfied with their major. Depression varied significantly according to sociodemographic and internet‐related characteristics. Participants with high depression had low economic status (F = 4.96 and p = 0.008), may be in poor health (F = 13.34 and p < 0.001), a lower previous grade point average (F = 3.33 and p = 0.021), if their motivation for applying to the Department of Nursing came according to the grade (F = 3.53 and p = 0.016), and if they were dissatisfied with their major (F = 13.87 and p = 0.001) (Table 1). TABLE 1Differences in depression according to sociodemographic and internet‐related characteristics (N = 184).CharacteristicCategories n (%)DepressionM ± SD † Difference of M ± SD t or F (p) post hocGenderMale22 (12.0)12.95 ± 10.93‐0.09 ± 0.35 (vs. Female) −0.04 (0.971) Female162 (88.0)13.04 ± 10.58Age (year)< 2162 (33.7)13.85 ± 9.921.10 ± 1.41 (vs. ≥ 23) 0.29 (0.750) 21 ∼ < 2366 (35.9)12.50 ± 10.68−0.13 ± 0.65 (vs. ≥ 23)≥ 2356 (30.4)12.75 ± 11.33M ± SD22.3 ± 3.36GradeFreshman39 (21.2)14.59 ± 10.601.04 ± 0.83 (vs. Senior) 0.81 (0.490) Sophomore44 (23.9)13.55 ± 10.100.42 ± 1.33 (vs. Senior)Junior47 (25.5)11.15 ± 10.10−1.98 ± 1.33 (vs. Senior)Senior54 (29.4)13.13 ± 11.43ReligionYes68 (37.0)12.50 ± 10.56−0.84 ± 0.09 (vs. No) −0.052 (0.603) No116 (63.0)13.34 ± 10.65Residence statusLiving with parents71 (38.6)12.92 ± 11.64−0.40 ± 1.58 (vs. Dormitory) 0.05 (0.949) Rented room74 (40.2)12.69 ± 9.81−0.63 ± 0.25 (vs. Dormitory)Dormitory39 (21.2)13.32 ± 10.06Economic statusHigha 41 (22.3)9.93 ± 10.49−7.34 ± 2.19 (vs. Low) 4.96 (0.008) a < c Moderateb 106 (57.6)12.75 ± 9.41−4.52 ± 3.27 (vs. Low)Lowc 37 (20.2)17.27 ± 12.68Health statusGooda 78 (42.4)9.23 ± 9.14−9.29 ± 2.30 (vs. Poor) 13.34 (< 0.001) a < b < c Moderateb 81 (44.0)14.42 ± 10.26−1.86 ± 0.81 (vs. poor)Poorc 25 (13.6)20.40 ± 11.28Grade point average for the entire university year≥ 4.067 (36.4)10.78 ± 8.64−12.39 ± 5.73 (vs. < 3.0) 3.33 (0.021) a < d 3.5 ∼ < 4.089 (48.4)13.52 ± 10.97−9.65 ± 3.40 (vs. < 3.0)3.0 ∼ < 3.522 (12.0)15.18 ± 11.80−8.00 ± 2.57 (vs. < 3.0)< 3.06 (3.2)23.17 ± 14.37Motivation for applying to the department of nursingAccording to one's aptitudea 62 (33.7)10.37 ± 10.13−7.39 ± 0.83 (vs. According to grades) 3.53 (0.016) a < d Parent recommendationb 21 (11.4)16.95 ± 12.45−0.81 ± 1.49 (vs. According to grades)Employment rate considerationc 84 (45.7)13.06 ± 9.89−4.70 ± 1.07 (vs. According to grades)According to gradesd 17 (9.2)17.76 ± 10.96Satisfaction with majorSatisfieda 105 (57.0)9.71 ± 9.20−9.29 ± 2.30 (vs. Not satisfied) 13.87 (0.001) a < b, c Moderateb 66 (35.9)17.14 ± 10.69−1.86 ± 0.81 (vs. Not satisfied)Not satisfiedc 13 (7.1)19.00 ± 11.50 † M ± SD, mean ± standard deviation. Differences in depression according to sociodemographic and internet‐related characteristics (N = 184). −0.04 (0.971) 0.29 (0.750) 0.81 (0.490) −0.052 (0.603) 0.05 (0.949) 4.96 (0.008) a < c 13.34 (< 0.001) a < b < c 3.33 (0.021) a < d 3.53 (0.016) a < d 13.87 (0.001) a < b, c M ± SD, mean ± standard deviation. Scores of Variables The descriptive data reported included means, standard deviations, skewness, and kurtosis statistics of the average stress scores, stress response, grit, and depression. Based on the cut‐off point of 20, the prevalence of depression was identified in 23.8% of the participants (45 participants), of which 13.7% (26 participants) had moderate depression and 10.1% (19 participants) had severe depression (Table 2). TABLE 2Scores of variables and depression stages (n = 184).Variables (range)Depression group (score range) n (%)ItemsMin ~ MaxTotal M ± SDM ± SDSkewnessKurtosisStress (1 ∼ 5)5050 ~ 197110.35 ± 31.932.21 ± 0.640.35−0.67Stress response (1 ∼ 5)3939 ~ 16481.19 ± 31.362.08 ± 0.800.85−0.17Grit (1 ∼ 5)2025 ~ 10065.78 ± 13.113.29 ± 0.66−0.360.53Depression (0 ∼ 63)230 ~ 4413.03 ± 10.600.950.15Normal (0 ~ 13)118 (62.4)4.25 ± 2.80Mild (14 ~ 19)26 (13.8)12.24 ± 1.86Moderate a (20 ~ 28)26 (13.7)18.96 ± 2.20Severe a (29 ~ 63)19 (10.1)31.45 ± 5.87Abbreviations: M, mean; SD, standard deviation. a Depression of cut‐off value = 20. Scores of variables and depression stages (n = 184). Abbreviations: M, mean; SD, standard deviation. Depression of cut‐off value = 20. Correlations of Variables Depression was significantly positively correlated with stress (r = 0.58 and p < 0.001) and stress response (r = 0.75 and p < 0.001). However, depression was significantly negatively correlated with grit (r = −0.35 and p < 0.001) (Table 3). TABLE 3Pearson correlations amongst stress, stress response, grit, and depression (n = 184).VariableStressStress responseGritDepression r (p) r (p) r (p) r (p)Stress1Stress response0.63 (< 0.001)1Grit−0.30 (< 0.001)−0.33 (< 0.001)1Depression0.58 (< 0.001)0.75 (< 0.001)−0.35 (< 0.001)1 Pearson correlations amongst stress, stress response, grit, and depression (n = 184). Mediating Effects of Stress Response and Grit in the Stress and Depression Upon reviewing the multicollinearity between independent variables prior to analysis, no multicollinearity issue was found because the tolerance was greater than 0.10 (ranging from 0.58 to 0.88) and the variance inflation factor (VIF) was less than 10 (ranging from 1.140 to 1.717). These results indicate that multicollinearity is not a concern in our model (George 2011). In addition, the Durbin‐Watson value was close to the 2.00 threshold at 1.87, indicating no residual autocorrelation problem, which suggests that the residuals are independent (Durbin and Watson 1951). For the normality test, a normal distribution was confirmed by inspecting the P–P plot, which showed that the residuals closely follow the diagonal line, indicating that the residuals are approximately normally distributed. For the homogeneity of variance (homoscedasticity) test, the scatterplot of standardised residuals against predicted values was examined. The scatterplot did not show any discernible pattern, indicating that the residuals have constant variance across all levels of the independent variables, thereby confirming homoscedasticity (Field 2013). To test the mediating effect of stress response and grit in the relationship between stress and depression in nursing students, “economic status,” “health status,” “grade point average,” “motivation for applying to the department of nursing,” and “satisfaction regarding their major,” which displayed significant differences in the univariate analysis of depression in nursing students, were controlled and analysed using PROCESS macro in a parallel multiple mediator model. The hypothetical model demonstrated a total explanatory power of 63.0% and was statistically significant (F = 18.51 and p < 0.001). Direct effects from stress to stress response (a1: B = 0.73, SE = 0.08, t = 9.26, and p < 0.001) and direct effects from stress to grit (b1: B = −0.09, SE = 0.08, t = 22.82, and p = < 0.001) were statistically significant. Direct effects from stress to depression (a2: B = 2.70, SE = 1.03, t = 2.61, and p = 0.009), direct effects from stress response to depression (b2: B = 8.04, SE = 0.83, t = 9.68, and p = < 0.001), and direct effects from grit to depression (c: B = ‐1.73, SE = 0.83, t = −2.09, and p = 0.039) were also statistically significant. Upon testing by bootstrapping the mediating effect of stress response and grit in the relationship between the stress and depression of the participants, the mediating effect of stress response was statistically significant (M1: B = 6.35 and 95% CI = 4.60 ~ 8.23), but that for grit was not (M2: B = 0.53 and 95% CI = −0.03 ~ 1.28) (Table 4). The final model based on the results of this study is presented in Figure 2. TABLE 4Mediation effects of stress response and grit in the stress and depression (N = 184, bootstrap samples = 10,000).EffectPathPath B a SE b t (p)SE95% CI c LL d UL e DirectStress → Stress responsea1 0.730.089.26 (< 0.001)0.650.93Stress → Gritb1 −0.090.0822.82 (< 0.001)−0.45−0.16Stress → Depressiona2 2.701.032.61 (0.009)0.664.73Stress response → Depressionb2 8.040.839.68 (< 0.001)6.409.68Grit → Depressionc−1.730.83−2.09 (0.039)−3.37−0.09EffectPathPathEffectBoot SE95% BootCI c LL d UL e IndirectStress → Stress response → DepressionM1 (a1a2)6.350.934.608.23Stress → Grit → DepressionM2 (b1b2)0.530.33−0.031.28 Note: Adj. R 2 = 0.63, F = 18.51, p < 0.001, bootstrap samples = 10,000. a Unstandardized coefficients. b Standard error. c Percentile 95% confidence interval. d Lower limit. e Upper limit. Mediation effects of stress response and grit in the stress and depression (N = 184, bootstrap samples = 10,000). Note: Adj. R 2 = 0.63, F = 18.51, p < 0.001, bootstrap samples = 10,000. Unstandardized coefficients. Standard error. Percentile 95% confidence interval. Lower limit. Upper limit. Final path diagram of the mediating effects of stress response and grit on the relationship between stress and depression. Discussion This study was conducted to explore the prevalence of depression amongst Korean nursing students and to identify the mediating effect of stress response and grit in the relationship between stress and depression in the population. We identified 45 (23.8%) students as being depressed when a cut‐off value of 20 was used to determine the presence of depression (Beck, Steer, and Brown 1987). This study found more participants to be depressed than in a study of medical students in South Korea using the same tool, where 9.4% of them were found to be depressed (Roh et al. 2009). These differences may be related to the time of investigation. According to a recent survey, the number of patients with depression in South Korea had increased by 8.0% annually to 36.1% in 2021 compared to 2017. By age group, patients with depression in their 20s accounted for the largest percentage at 19.0% of all patients (Health Insurance Review and Assessment Service 2022). In a recent survey, the prevalence of depression has increased in all age groups, where the rise in the 20s may be because of COVID‐19. Activity restrictions, severance of relationships, and anxiety regarding employment and career may have intensified depression at a period when social relations should have been actively formed (Cengiz, Gurdap, and Işik 2022). There have been reports that COVID‐19 has negatively affected the mental health of undergraduate nursing students (Taylor, Thomas‐Gregory, and Hofmeyer 2020). In addition, previous studies have demonstrated that women experience depression 2.1 times more than men (Health Insurance Review and Assessment Service 2022). Women made up 88% of the participants of this study, which may have contributed to the high prevalence of depression. Stress and depression in nursing students affect individual health and well‐being and can later lead to the provision of poor‐quality patient care if not correctly managed (Aloufi et al. 2021). Therefore, it is currently very crucial to adequately manage high‐risk groups by closely investigating relationships with factors related to the depression of nursing students. In this study, the variables that demonstrated significant differences in the univariate analysis of depression of nursing students were economic status, health status, grade point average, motivation for applying for the department of nursing, and satisfaction regarding their major. This was consistent with previous studies that reported factors influencing depression in nursing students could be academic performance, academic stress, interest in their majors, occupational future, interpersonal relationship, frequency of exercise, self‐reported health status, and social support (Xu et al. 2014). As academic issues were found to be a significant risk factor for depression in nursing students in various previous studies, it is urgent to diagnose high‐risk groups of depression in the early stages for nursing students and provide appropriate interventions. The aetiology of depression is multifactorial (Xu et al. 2014). Likewise, in this study, depression was positively correlated with stress and stress response, while it was negatively correlated with grit. This study tested the mediating effect of stress response and grit in the relationship between stress and depression in nursing students. The direct effect of stress on stress response and the direct effect of stress on grit were statistically significant. The direct effect of stress on depression, the direct effect of stress response to depression, and the direct effect of grit on depression were also statistically significant. The presence of direct effects between stress and grit and grit and depression were consistent with previous studies that reported that individuals with high levels of grit could have relatively low stress (Shin, Koo, and Park 2018). In a previous study of nurses, those with low grit noted that they experience higher levels of stress and mental health problems, such as anxiety, depression, and other negative emotions (Peng and Wu 2022). Previous studies have demonstrated that an individual with grit responds more appropriately to emotional exhaustion, which makes them less prone to depression (Jung et al. 2023). Grit affects an individual's attitude towards failure or adversity because it is a positive personality trait (Duckworth et al. 2007). Stress responses can vary from person to person even if the level of stress they experience is similar (Han 2020). Some people perceive stress as challenging but continue to persevere (Kim and Park 2020). Most of the stressors experienced by nursing students are uncontrollable. Therefore, this study aimed to identify mediating factors that can prevent stress from causing negative outcomes such as depression. We hypothesized that stress response and grit would mediate the relationship between stress and depression in nursing students. Upon testing the mediating effect of stress response and grit in the relationship between stress and depression in the study participants, it was found that stress response plays a mediating role between stress and depression, unlike grit. According to the stress‐vulnerability model, symptoms of depression are experienced as a natural emotional response under stress. At this point, factors of cognitive vulnerability can serve as agents that control the relationship between stress and depression (Ingram, Miranda, and Segal 1998). In other words, an individual's form of stress response can serve as a factor that controls the effect of stress and depression (Kim and Park 2013). Individuals may show varying stress responses even when experiencing the same stress level. Therefore, to reduce depression in nursing students, personalised strategies to reduce stress and stress response are necessary. It was an unexpected finding that the mediating effect of grit was not significant in the relationship between stress and depression. Previous studies have demonstrated that the interaction between maladjustment to university life and grit significantly moderates the indirect effect that traverses from career stress to depression through maladjustment to university life (Shin and Park 2021). However, it has been noted that while people of higher grit can control individual internal factors such as individual needs, impulses, and efforts towards long‐term goals, there are limitations in the factors that are difficult for them to wield command over, such as relationships with professors and peers (Shin and Park 2021). According to a previous study, grit alone may not be sufficient for students with depressive symptoms to succeed in college (Kilgore et al. 2023). Grit is defined as the ability to persevere towards a goal despite adversity (Kilgore et al. 2023). However, the results of this study may suggest that not all stress can be controlled by grit. The stress investigated in this study involves a mix of external factors such as peers, family members, professors, and individual factors, based on which it is believed that the mediating effect of grit on stress and depression did not appear. Therefore, external resources are necessary to properly manage the stress of nursing students. Strength and Limitations of the Work This study makes a significant contribution to the literature because it provides the variables that demonstrated significant differences in the univariate analysis of depression in nursing students. As nurses are exposed to a high‐stress job environment, managing stress appropriately is very important for maintaining their mental health. Therefore, understanding the factors that affect stress and depression during undergraduate nursing school could contribute to producing healthy nurses in the future. This study has the following limitations. First, due to the cross‐sectional design, causal inferences are necessarily limited. Second, we relied on self‐reported data; subsequent studies should objectively measure stress response and depression using physiological parameters. Third, it is difficult to generalise the findings, as the participants were from four universities in South Korea. Implications for Nursing Practice Nursing students experience high stress from their school days and into their professional practice. This stress is directly related to depression. Accumulated stress can be a greater risk factor for mental health management in the future. Therefore, nursing students should be able to manage stress symptoms that have an intermediate mediating effect in the relationship with depression as well as stress. This experience may help nursing students protect their mental health when they become clinical nurses. Recommendations for Further Research Further, the possibility of grit as a covariate cannot be excluded, and there may be bias due to self‐report questionnaires. Hence, it is necessary to investigate different populations using longitudinal studies to establish causality or additional variables that could mediate the relationship between stress and depression. Moreover, replication studies in various cultural and educational settings should be conducted to enhance the generalizability of the findings. Conclusions The prevalence of depression amongst Korean nursing students was high, of which 13.7% had moderate depression and 10.1% had severe depression. The direct effects from stress to stress response, from stress to grit, from stress to depression, from stress response to depression, and from grit to depression were statistically significant. The mediating effect of stress response in the relationship between stress and depression was statistically significant, but that for grit was not. These results suggest that preparing strategies against depression amongst nursing students is an urgent matter. Therefore, it is necessary to investigate students' stress responses and provide personalised interventions, as well as reduce established stressors. Author Contributions Conceptualization, J.J. and S.C.; data curation, J.J. and S.C.; formal analysis, J.J.; investigation, J.J. and S.C.; methodology, J.J. and S.C.; resources and software, J.J. and S.C.; supervision, J.J.; validation, J.J. and S.C.; roles/writing – original draft, J.J. and S.C.; writing – review and editing, J.J. and S.C. Consent Informed consent was obtained from all subjects involved in the study. Institutional Review Board Statement: The study was approved by the Institutional Review Board of the researcher's university (No. GWNUIRB‐2023‐8). Conflicts of Interest The authors declare no conflicts of interest. Acknowledgements This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors. The authors affirm that the methods used in the data analyses are suitably applied to their data within their study design and context, and the statistical findings have been implemented and interpreted correctly. Funding: The authors received no specific funding for this work. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.