Social-emotional competence as a protective factor against suicide risk in Chinese college students: a moderated mediation analysis of depression and self-esteem Authors: Yanhong Shen, Xinyu Yin, Beibei Xu, Shuanghu Fang Published: 2026 ABSTRACT This study investigates the relationship between social and emotional competence (SEC) and suicide risk in Chinese college students and whether depression is a mediating factor, and furthermore whether self-esteem is a moderating factor of direct and indirect pathways. A cross-sectional study among 24,103 college students from one Chinese university using the Delaware Social-Emotional Competency Scale (DSECS-S), Patient Health Questionnaire-9 (PHQ-9), Rosenberg Self-Esteem Scale (RSES), and Suicidal Behaviors Questionnaire-Revised (SBQ-R). Correlation, mediation, and moderated mediation analysis were conducted controlling for gender, health status, place of origin, and sibling status. The path from SEC to suicide risk is significantly and negatively associated (β = -0.23, p < 0.001). Depression partially mediates this relationship SEC is negatively associated with depression (β = -0.17, p < 0.001); depression, in turn, is positively associated with suicide risk (β = 0.29, p < 0.001). The indirect effect of SEC on suicide risk through depression is significant, indirect effect = -0.05, (95% CI [-0.06, -0.05]). Self-esteem significantly moderated all paths in the mediation model. Specifically, higher self-esteem strengthens the negative association between SEC and depression, strengthens the beneficial association between SEC and lower suicide risk, and attenuates the positive association between depression and suicide risk (simple slope).Simple slope analyses revealed that associations between SEC and depression, and between depression and suicide risk are weaker among students with higher self-esteem. A moderated mediation model in which SEC (social-emotional competencies) are associated with lower suicide risk directly and through depression was supported, with self-esteem acting as a protective moderator across pathways. Introduction Suicide risk context in college students Suicide remains a critical global public health challenge. Among individuals aged 15–29, it is the second leading cause of death worldwide [1], a demographic that includes college students. In China, the situation is particularly concerning, as the suicide rate among college students is reported to be 2 to 4 times higher than that of their non-student peers [2], making suicide the leading cause of non-accidental death in this population [3]. Recent research indicates a continued upward trend in suicide related outcomes among college students [4]. A large-scale meta-analysis of over 500 high-quality studies further examined mental health problems among Chinese college students between 2010 and 2020 [5]. The findings revealed a notable increase in anxiety, depression, sleep disturbances, and suicidal behaviors over the past decade. This alarming prevalence underscores the urgent need for a deeper understanding of the mechanisms underlying suicidal risk in this vulnerable group. Given the complex and often unpredictable nature of suicidal behavior, identifying modifiable protective factors and the pathways through which they operate is paramount for developing effective prevention strategies. Social-emotional competence as a foundational protective factor Social-emotional competence (SEC) represents a critical internal protective factor against suicide risk among college students [6]. It is defined as the capacity to acquire and apply emotional and social knowledge, attitudes, and skills to recognize and manage emotions, develop and implement goals, demonstrate empathy, maintain supportive relationships, and make responsible decisions [7]. Previous research shows that higher SEC predicts better academic performance [8, 9], subjective well-being [10], interpersonal relationships [6, 11], and mental health outcomes [12], while also playing a pivotal role in healthy social development [13]. Conversely, deficits in SEC are robustly linked to elevated suicide risk. According to the Interpersonal Theory of Suicide [14], thwarted belongingness and perceived burdensomeness are primary drivers of suicidal desire. Deficits in SEC can directly contribute to these states by impairing an individual’s ability to form and maintain meaningful social connections (thwarted belongingness) [15], and by fostering maladaptive cognitions that amplify perceived ineffectiveness (perceived burdensomeness) [16, 17]. Therefore, we propose that higher SEC serves as a critical buffer against suicide risk by mitigating these interpersonal deficits (H1). Depression as a key mechanistic pathway According to the Organization for Economic Cooperation and Development (OECD), SEC equips students with the requisite skills to navigate academic, social and digital challenges, thereby enhancing psychological resilience during crises. However, a proportion of college students currently exhibit deficiencies in emotional regulation, social awareness, and self-management, which may contribute to the high prevalence of depression symptoms [18–20], The “Report on National Mental Health Development in China (2021–2022),” indicates that 66.76% of college students experiencing symptoms of depression and anxiety. Depression is a significant risk factor for suicide [21–23], and both its severity and duration significantly increasing suicide risk [24]. The Motivational-Volitional Model of Suicidal Behavior posits that feelings of defeat and humiliation can lead to a state of entrapment, which, in the presence of poor problem-solving and negative future thinking (common in depression), fuels the motivation for suicide [25]. We argue that lower SEC may increase vulnerability to such states. Individuals with poor emotional regulation and social problem-solving skills (core components of SEC) are less equipped to cope with stressors, potentially leading to feelings of defeat and, subsequently, depressive symptoms. These depressive symptoms, in turn, may exacerbate feelings of hopelessness and entrapment, thereby increasing suicide risk. Thus, we hypothesize that depression mediates the relationship between SEC and suicide risk, serving as a critical explanatory pathway (H2). The moderating role of self-esteem Importantly, the strength of the relationships in the proposed SEC→Depression →Suicide Risk pathway may vary between individuals. Self-esteem, defined as an individual’s overall subjective evaluation of their own worth [26], is a crucial personal psychological resource that may serve as a critical moderator. According to the Social Buffering Hypothesis model [27], higher levels of self-esteem can help individuals buffer the adverse effects of stressful events. In the context of our model, individuals with high self-esteem may be less likely to experience severe depressive symptoms following SEC-related challenges [28, 29], as a positive self-view can mitigate the impact of social setbacks or perceived failures [30]. self-esteem may serve as a “buffer” between depression and suicide risk: even when experiencing depressive symptoms, individuals with a stronger sense of self-worth are more likely to maintain greater self-efficacy and future orientation [31], thus hindering the transformation of despair into suicidal desire. Empirical evidence supports that self-esteem indeed moderates the impact of various risk factors on depression and suicide [32]. Based on the theoretical analysis and empirical evidence outlined above, this study proposes that self-esteem plays a moderating role in all paths of the mediation model. Specifically, it not only buffers the negative impact of low SEC on depression (Path a) but also buffers the promoting effect of depression on suicide risk (Path b). Additionally, it may moderate the direct protective effect of SEC on suicide risk (Path c’). This comprehensive moderating effect constitutes the third hypothesis of this study (H3), as shown in Fig. 1. Fig. 1Research hypothesis model Research hypothesis model In summary, prior studies have established SEC’s direct effects on depression [33–35], and its association with suicide risk [36, 17], and identified depression as a mediator and self-esteem as a moderator in related contexts, an integrative model examining these constructs simultaneously is lacking. This study aims to address this gap by proposing and testing a moderate mediation model. Specifically, we investigate: (1) whether SEC is negatively associated with suicide risk among Chinese college students; (2) whether this relationship is mediated by depression; (3) whether self-esteem moderates the direct effect of SEC on suicide risk, the effect of SEC on depression, and the effect of depression on suicide risk. This investigation will provide a more nuanced understanding of the mechanisms linking SEC to suicide risk and identify for whom these protective pathways are strongest, offering valuable insights for targeted, evidence-based interventions. Methods Participants A cluster sampling method was employed to survey 25,828 college students from a comprehensive university in Anhui Province, China. Participants were recruited using natural classes as sampling units. Questionnaires were administered collectively during regular class meetings to ensure standardized and centralized administration. After excluding invalid questionnaires (e.g., all“1” or all“4” across items), and scores exceeding three standard deviations (scores exceeding 3 standard deviations on the total scores of four DSECS-S、SBQ-R、PHQ-9、RSES were excluded), 24,103 valid questionnaires were obtained, representing a valid response rate of 93.32%, the specific process is shown in Table 1. Table 1CONSORT-style data flow processStage N Initial sample25,828Excluded for response quality issues1312Excluded for 3SD (on 4 scales)413Final valid sample24,103Valid response rate93.32% Table 1CONSORT-style data flow processStage N Initial sample25,828Excluded for response quality issues1312Excluded for 3SD (on 4 scales)413Final valid sample24,103Valid response rate93.32% CONSORT-style data flow process The participants ranged in age from 17 to 24 years. The sample included 9,602 males (39.84%) and 14,501 females (60.16%), In terms of residential background, 8,123 participants were from urban areas (33.70%), 4,688 from county-towns (19.45%), and 11,292 from rural areas (46.85%). In terms of academic year, there were 6,791 freshmen (28.18%), 7,333 sophomores (30.42%), 7,005 juniors (29.06%), and 2,974 seniors (12.34%). Additionally, 16,175 participants were only children (67.11%), while 7,928 were non-only children (32.89%). Measures Delaware Social-Emotional Competency Scale (DSECS-S) The Delaware Social-Emotional Competency Scale, which was developed in collaboration with George Bear and Yang, was utilized. The scale comprises 12 items, covering four dimensions: responsible decision-making, relationship skills, social awareness, and self-management. The scale employs a 4-point Likert scale (1 = not at all like me, 4 = very like me), with higher scores denoting elevated levels of SEC [37]. In this study, the Cronbach’s alpha coefficient for the DSECS-S was 0.87. Patient Health Questionnaire 9 (PHQ-9) Depressive symptoms were measured using PHQ-9, which is based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) [38] The PHQ-9 is a self-report instrument comprising nine items that assess the frequency of depressive symptoms experienced over the past two weeks. Each item is rated on a 4-point Likert scale (0 = not at all, 3 = almost every day), with a total score ranging from 0 to 27. Higher scores indicate more severe depressive symptoms [39]. In this study, the Cronbach’s alpha coefficient for the PHQ-9 was 0.88. Rosenberg Self-Esteem Scale (RSES) This study employed the Rosenberg Self-Esteem Scale [26], which comprises 10 items. The scale employs a 4-point Likert scale (1 = strongly disagree, 4 = strongly agree), with higher scores indicative of elevated levels of self-esteem. In this study, the Cronbach’s alpha coefficient for the RSES was 0.88. Suicidal Behaviors Questionnaire-Revised (SBQ-R) This study employed the Suicidal Behavior Questionnaire-Revised, as developed by Osman et al. [40], to measure suicide risk. The scale consist of four items, with a total score ranging from 3 to 18. A higher score indicates higher risk of suicide. A score of 7 or above is indicative of a high risk of suicidal behavior [41]. In this study, the Cronbach’s alpha coefficient for the SBQ-R was 0.67, for scales with fewer than 10 items, reliability coefficients in the range of 0.60–0.70 are considered acceptable for research purposes [42]. Additionally, The McDonald’s ω coefficient for SBQ-R is 0.70, and the Spearman–Brown coefficient is 0.71. Confirmatory factor analysis and scale reliability To establish the construct validity of the key measures in our specific sample, we conducted confirmatory factor analyses (CFA) for the DSECS-S, PHQ-9, RSES, and SBQ-R. The model fit was satisfactory for all scales: DSECS-S (CFI = 0.96, TLI = 0.94, RMSEA = 0.07, SRMR = 0.03), PHQ-9 (CFI = 0.96, TLI = 0.93, RMSEA = 0.07, SRMR = 0.02), RSES (CFI = 0.99, TLI = 0.98, RMSEA = 0.05, SRMR = 0.02), and SBQ-R (CFI = 0.99, TLI = 0.98, RMSEA = 0.04, SRMR = 0.01). All indices met or exceeded conventional thresholds for good fit, supporting the structural validity of these scales in the present study. Statistical analysis SPSS version 28 software was used to data analysis. Analyses including tests for common method bias test, reliability and validity test, descriptive statistics, one-way ANOVA, and t-test, partial correlation analysis, mediating effect test, and moderating effect test. Ethical considerations Ethical approval The study was approved by the Ethical Committee of Anhui Normal University with the approval code of AHNU-ET2025063. Informed consent procedure Prior to the survey, all participants were fully informed about the research purpose, confidentiality protocols, and their right to withdraw at any time, all participants provided informed consent. Suicide risk management protocol Given the sensitive nature of the topic (suicide risk and depression), we implemented a structured crisis referral procedure: Participants scoring above the clinical threshold on the suicide (SBQ-R) or depression (PHQ-9) scales were flagged for follow-up. These students were invited for individual assessments by university counselors at the Mental Health Education Center. Based on the assessment outcomes, students were either offered on-campus counseling or referred to specialized psychiatric hospitals for further treatment. Results Common method bias test A single-factor test was employed to conduct exploratory factor analysis on 30 items across four scales. Five factors were identified with eigenvalues greater than 1. The first factor explained the largest proportion of variance at 27.30%, which is below the critical threshold of 40% [43], indicating that there was no significant influence of common method bias. Descriptive statistics and correlation analysis The pertinent analysis indicates a correlation between SEC, self-esteem, depression, and suicide risk (the average values, standard deviations, and correlation coefficients of each variable are shown in Table 2). Table 2Descriptive statistics and correlation analysis of variablesM ± SD12341.social emotional competence39.74 ± 5.4812. self-esteem31.39 ± 5.430.51***13.depression3.15 ± 3.45-0.23***-0.36***14.suicide risk4.29 ± 1.85-0.21***-0.33***0.40***1***p < 0.001 Table 2Descriptive statistics and correlation analysis of variablesM ± SD12341.social emotional competence39.74 ± 5.4812. self-esteem31.39 ± 5.430.51***13.depression3.15 ± 3.45-0.23***-0.36***14.suicide risk4.29 ± 1.85-0.21***-0.33***0.40***1***p < 0.001 Descriptive statistics and correlation analysis of variables ***p < 0.001 Normality test To ensure that the density distribution of the measured data follows a normal distribution, normality tests are conducted on the relevant variables. Normality testing includes assessments of kurtosis and skewness. Kline suggests that as long as the skewness falls between ± 3 and the kurtosis falls between ± 10, the sample data of the questionnaire can be considered to approximately follow a normal distribution [44]. The following table illustrates the normality test for variables that may exhibit high skewness and discrete distributions (Table 3). Table 3Normality test statisticsMinMaxMeanstd.SkewnessKurtosisStatisticsstd.Statisticsstd.depression-1.152.84-0.060.880.890.02-0.040.03suicide risk-2.902.68-0.060.841.100.020.650.03 Table 3Normality test statisticsMinMaxMeanstd.SkewnessKurtosisStatisticsstd.Statisticsstd.depression-1.152.84-0.060.880.890.02-0.040.03suicide risk-2.902.68-0.060.841.100.020.650.03 Normality test statistics Introduction of covariates Gender(1 = Male, 2 = Female, M = 1.61, SD = 0.49), Health status (0 = Healthy, 1 = Unhealthy, M = 0.05, SD = 0.22), Place of origin (1 = Urban, 2 = Township, 3 = Rural, M = 2.13, SD = 0.89), Only child status (1 = Only child, 2 = Not only child, M = 1.67, SD = 0.47).It should be noted that health status reflects participants’ subjective evaluation of their physical health status. “0” indicates “healthy,” and “1” indicates “unhealthy.” Mediation effect test After controlling for gender, health status, place of origin, and siblings, which identified as influential factors for SEC, suicide risk, and self-esteem [45, 46], the mediation effect of depression on the relationship between SEC and suicide risk was tested. As shown in Fig. 2, SEC had a significant negative predictive effect on suicide risk (β = -0.23, p < 0.001). Upon the inclusion of depression in the mediation model, SEC was still found to significantly negatively predict suicide risk (β = -0.18, p < 0.001). Additionally, SEC was found to significantly negatively predict depression (β = -0.17, p < 0.001), while depression was found to significantly positively predict suicide risk (β = 0.29, p < 0.001). Furthermore, the bias-corrected bootstrap method was employed to ascertain the statistical significance of the mediation effect. The results demonstrated that the mediation effect value was − 0.05 (Boot SE = 0.002, [-0.06, -0.05]), indicating that depression plays a partial mediating role in the relationship between college students’ SEC and suicide risk. Fig. 2Mediating model of depression on the impact of SEC on suicide risk Mediating model of depression on the impact of SEC on suicide risk The moderating effect of self-esteem The moderating effect of self-esteem on each path in the mediation model was evaluated using Model 59 in the PROCESS 4.0 macro. As evidenced by the results presented in Table 4, the moderating effect of self-esteem was found to be statistically significant for each path within the mediation model. In Path 1, SEC was found to significantly negatively predict depression (β = -0.03, p < 0.001). Furthermore, the results indicated that self-esteem significantly negatively predicted depression (β = -0.31, p < 0.001), and that the interaction between SEC and self-esteem also significantly negatively predicted depression (β = -0.02, p < 0.001). In Path 2, SEC significantly negatively predicted suicide risk (β = -0.17, p < 0.001), while depression significantly positively predicted suicide risk (β = 0.25, p < 0.001). Self-esteem was identified as a moderating variable, exhibiting a significant negative predictive effect on suicide risk (β = -0.06, p < 0.001), and the interaction between SEC and self-esteem also demonstrated a significant negative predictive effect on suicide risk (β = -0.03, p < 0.001). Furthermore, the interaction between self-esteem and depression was found to significantly negatively predict suicide risk (β = -0.05, p < 0.001). Table 4Mediation effect test with moderationEquation1:Depression(M)Equation2:Suicide risk(Y)βt95%CIβt95%CISex0.2624.91***[0.24,0.28]0.1716.48***[0.15,0.19]Health0.8637.94***[0.81,0.90]0.114.57***[0.06,0.15]Place of origin0.1118.04***[0.10,0.12]-0.05-8.87***[-0.06,-0.04]Siblings0.021.75[-0.002,0.04]-0.01-1.10[-0.03,0.01]Social-Emotional Competence(X)-0.03-5.01***[-0.05, -0.02]-0.17-25.43***[-0.18, -0.16]Self-esteem(W)-0.31-53.67***[-0.32, -0.30]-0.06-9.69***[-0.07, -0.05]X*W-0.02-4.01***[-0.04, -0.01]-0.03-4.81***[-0.04, -0.02]Depression(M)0.2537.14***[0.24,0.26]M*W-0.05-8.45***[-0.06, -0.04]R20.230.17F1024.69***563.34***The 'Health' variable represents participants' self-reported physical health status Table 4Mediation effect test with moderationEquation1:Depression(M)Equation2:Suicide risk(Y)βt95%CIβt95%CISex0.2624.91***[0.24,0.28]0.1716.48***[0.15,0.19]Health0.8637.94***[0.81,0.90]0.114.57***[0.06,0.15]Place of origin0.1118.04***[0.10,0.12]-0.05-8.87***[-0.06,-0.04]Siblings0.021.75[-0.002,0.04]-0.01-1.10[-0.03,0.01]Social-Emotional Competence(X)-0.03-5.01***[-0.05, -0.02]-0.17-25.43***[-0.18, -0.16]Self-esteem(W)-0.31-53.67***[-0.32, -0.30]-0.06-9.69***[-0.07, -0.05]X*W-0.02-4.01***[-0.04, -0.01]-0.03-4.81***[-0.04, -0.02]Depression(M)0.2537.14***[0.24,0.26]M*W-0.05-8.45***[-0.06, -0.04]R20.230.17F1024.69***563.34***The 'Health' variable represents participants' self-reported physical health status Mediation effect test with moderation The 'Health' variable represents participants' self-reported physical health status To further examine the moderating effect of self-esteem on the relationship between SEC and depression, two groups were formed based on high and low levels of self-esteem, and a simple slope test was conducted. As illustrated in Fig. 3, for college students with low self-esteem, a notable reduction in depression was observed as SEC improved. For college students with high self-esteem, the improvement in SEC was also associated with a significant reduction in depressive symptoms, with a greater decrease than that observed in students with low self-esteem (simple slope=-0.01, t=-1.25, 95%CI= [-0.03, 0.01]). Furthermore, in comparison to college students with low self-esteem, a stronger negative predictive effect on depression was observed among college students with high self-esteem (simple slope=-0.06, t=-5.97, 95%CI= [-0.075, -0.04]) in response to an improvement in SEC. Fig. 3The moderating effect of self-esteem in the relationship between SEC and depression. Notes:High self-esteem's slope is -0.06; low self-esteem's simple slope is -0.01 The moderating effect of self-esteem in the relationship between SEC and depression. Notes:High self-esteem's slope is -0.06; low self-esteem's simple slope is -0.01 The second step of the study examined the moderating effect of self-esteem on the relationship between depression and suicide risk. As illustrated in Fig. 4, for college students with low self-esteem, there was a notable increase in suicide risk with elevated levels of depression. For college students with high self-esteem (simple slope = 0.20, t = 19.52, 95%CI=[0.18, 0.22]), the suicide risk also increased significantly with higher levels of depression; however, the increase was less pronounced than that observed in students with low self-esteem (simple slope = 0.30, t = 41.25, 95%CI=[0.29, 0.31]). Depression had a more pronounced positive predictive effect on suicide risk among college students with low self-esteem, in comparison to those with high self-esteem. Fig. 4The moderating effect of self-esteem in the relationship between depression and suicide risk. Notes: High self-esteem's slope is 0.20; low self-esteem's simple slope is 0.30 The moderating effect of self-esteem in the relationship between depression and suicide risk. Notes: High self-esteem's slope is 0.20; low self-esteem's simple slope is 0.30 The third step was to investigate the moderating effect of self-esteem on the relationship between SEC and suicide risk. As illustrated in Fig. 5, for college students with low self-esteem, there was a notable decline in suicide risk as SEC improved. For college students with high self-esteem (simple slope=-0.20, t = 20.43, 95%CI=[-0.22, -0.18]), a significant inverse relationship was also observed between suicide risk and SEC, with a greater decrease in risk than that observed in college students with low self-esteem (simple slope=-0.14, t=-17.55, 95%CI=[-0.16,-0.12]). The negative predictive effect of SEC on suicide risk was more pronounced among college students with high self-esteem in comparison to those with low self-esteem. Fig. 5The moderating effect of self-esteem in the relationship between SEC and suicide risk. Notes: High self-esteem's slope is -0.20; low self-esteem's simple slope is -0.14 The moderating effect of self-esteem in the relationship between SEC and suicide risk. Notes: High self-esteem's slope is -0.20; low self-esteem's simple slope is -0.14 Discussion As the 21st century progresses, the suicide risk of Chinese college students is still a severe public health issue [47]. Accordingly, the present study was focused on college students and examined the effect of social-emotional competence (SEC) on suicide risk and the psychological mechanisms underlying that effect. The findings increase understanding of suicidal behavior among college students and have empirical implications for university prevention and intervention efforts. The impact of social emotional competence on college students’ suicide risk The present study demonstrated that SEC directly and negatively predicts the suicide risk among college students, thereby supporting the Interpersonal Theory of Suicide [48]. This theory emphasizes belongingness as a fundamental human need and posits that they are vulnerable to suicide if their interpersonal needs remain unfulfilled (as opposed to unmet needs in general). Specifically, suicidal ideation and behavior increases if people become frustrated in their desire to develop social relations. Furthermore, social emotional competence (SEC) encompasses the essential abilities for navigating interpersonal relationships, managing emotional experiences, solving daily life problems, and adapting to growth and development needs [49]. Deficits in SEC may hinder individuals’ ability to meet their basic need for belongingness, thereby increasing risk for suicide. Compared to high school settings, college students encounter a more diverse, complicated, and competitive interpersonal landscape, in which students negotiate the demands of coursework, social role transitions, and emotional stressors that require competent social-emotional skills. However, this period also coincides with a heightened susceptibility to the emergence of mental health concerns [50]. When students do not possess sufficient SEC, they may frequently encounter persistent interpersonal challenges, dysregulated emotions, and poor coping resources. Under these conditions suicidal behavior may sometimes be seen to escape from aversive states and/or as a maladaptive coping response [51, 52]. The mediating role of depression in the influence of social emotional competence on college students’ suicide risk In alignment with prior research that controlled for gender and age [53] or gender and grade [54], this study confirms that depression serves as a mediating factor in the relationship between SEC and suicide risk among college students. Higher level SEC is associated with fewer depressive symptoms among college students, which in turn is associated with lower suicide risk. Teasdale [55] has argued that during depressive episodes negative mood states become cognitively linked with dysfunctional beliefs and biased modes of information processing, and these links are strengthened across further depressions. The result is a higher probability that suicidal ideation, one of the strongest markers for suicidal behavior, will appear. Consistent with this framework, previous study shown, that depression acts as a mediator in the influence of various psychological risk factors on suicide [56–58]. Our results add to this literature by showing that SEC influences suicide risk through its effect on depression. Based on the cognitive model of suicidal behavior, people who find themselves trapped in a cognitive–emotional cycle based on hopelessness will use fewer adaptive problem-solving strategies. In such contexts, suicide may feel fixed and inevitable and result in impaired judgment. In this sense, suicidal behavior is a maladaptive attempt to relieve the hopelessness that characterizes depression [59]. Empirically, depressed college students are much more likely to develop suicidal behavior than college students with low or no depressive symptoms [60]. Taken together, these studies show that lower SEC associates with higher depressive symptoms, which in turn predict higher suicide risk in college students. Furthermore, the results indicate that depression partly mediates the relationship of SEC and suicide risk, thereby corroborating (albeit indirectly) the Differential Activation Theory of Suicidality. Disruptions in cognitive-emotional functioning might increase suicide risk directly as well as through negative emotional states (i.e., depression). Whereas SEC is a core competency for a group of students involved in positive relationships and engagement in college academic life, impairment in SEC may lead to social inadequacies with peers. Students without strong interpersonal skills face greater difficulty obtaining acceptance attempts as well as failure to reciprocate friendship with peers [61, 62]. Furthermore, they exhibit diminished interest in academic, manifesting as reduced school engagement and poorer academic conduct [61]. In such instances, individuals may resort to suicide as a means of attempting to alter their circumstances, particularly in the context of interpersonal and academic challenges [63]. Conversely, these experiences indicate that college students’ needs for social interaction and academic achievement are unmet, which increases the likelihood of them experiencing depression and negative emotions. The most central nodes in the network structure of adolescent suicidal behavior are depressive symptoms and bullying. Therefore, college students with high levels of depression are more likely to engage in suicidal behaviors [63]. The moderating role of self-esteem This mediating model linking social-emotional competence (SEC) to suicide risk among college students is moderated by self-esteem at every pathway. First, regarding the direct path of the mediation model, college students with high self-esteem had a stronger (or more negative) association with SEC in predicting suicide risk compared with students with low self-esteem. Self-esteem is a relatively stable inner psychological trait, with one’s perceived sense of personal worth at its core [64]. The interpersonal competence is an important aspect of SEC, and those with higher self-esteem show more confidence and ability in social interaction and communication, increasing the likelihood of being accepted in peer relations and positive interpersonal experience overall [65]. Thus, when college students present with low levels of SEC, high self-esteem may lead to compensatory experiences of interpersonal value, thereby buffering suicide risk. High SEC, when students possess it, means high self-esteem further heightens this effect, such that the risk of suicidality is lower. Second, self-esteem has a prominent moderating effect on the first stage of the mediating pathway; namely, the negative relationship of the SEC with depression is stronger among college students with high self-esteem than among those with low self-esteem. Specifically, self-esteem has independent and negative predictive effects on depression, and the beneficial effect of interaction between the SEC and self-esteem toward reduced depression is more pronounced among college students. The results of the present study are generally in line with those of Liu and Wang [66], while the former study controlled for sex and age, the present study controlled for the health state, place of origin, and whether the student has siblings or not. Finally, self-esteem also moderates the relation between depression and suicide risk. Depression predicts suicide risk more strongly for college students that have low self-esteem than it does for those high in self-esteem, lending partial support to the cognitive model of suicidal behavior [59]. High self-esteem acts as a protective buffer in which suicide risk is dampened in the presence of depressive symptoms [67]. There is consistent evidence that high self-esteem and well-being relates to a lower likelihood of suicidal behavior [68]. When self-esteem remains chronically low, cognitive processing may become rigidified and polarized about the self, leading to increased vulnerability to suicidal behavior in the presence of depression [69]. Research implications These findings have important implications for university administrators, families, and students in college. First, universities should place a higher priority on student mental health-particularly the prevention of suicidal behavior by incorporating SEC training, facilitating and encouraging student participation in social activities, and building psychological crisis intervention services. Improved SEC may alleviate depressive symptoms, thereby buffering the association between depression and suicide risk. Second, parents should teach their children to develop self-management, communication, and decision-making skills at an early age, as a stronger SEC may enhance individuals’ ability to cope with stressors and adversity later in life. Finally, the constructive coping with life stressors (e.g., through developing problem-solving skills, improving emotional regulation, and reducing negative emotions) was associated with a lower likelihood of depression, which is consistent with a protective pattern against suicide risk. Research limitations Despite these contributions, there are several limitations that should be acknowledged. First, this study employed a cross-sectional design, which only reflects the associations among variables at a single time point and cannot determine temporal order or causal direction. Although the theoretical model was used to examine the pathways linking social and emotional competence, depression, self-esteem, and suicide risk, strict causal inferences cannot be drawn. Future research may adopt longitudinal or experimental designs to further verify the causal mechanisms among these variables. Second, this study employed Harman’s single-factor test, which has its own limitations: it primarily serves as a screening tool and may not fully identify complex forms of method variance. Future research could adopt more robust statistical methods (e.g., the latent common method factor approach, marker variable techniques), to better separate method variance. Third, as the sample was drawn from a single university, the generalizability of the findings may be limited. Future studies involving multi-site designs and more diverse institutional settings are recommended to enhance the external validity of the results. Finally, this study did not control for other plausible confounders relevant to suicide/depression (e.g., adverse childhood experiences), their absence may influence the comparability of our findings with studies that account for a broader set of factors. Conclusion The study examined the association between social-emotional competence and suicide risk among Chinese college students using a moderated-mediation strategy. The SEC score is significantly negatively associated with suicide risk, while depression partially mediates that relationship; higher scores in SEC are associated with fewer depressive symptoms, leading to lower suicide risk. In addition, self-esteem acts as a moderator of all paths in the mediation model. Higher levels of self-esteem result in a stronger negative association between SEC and depression, a stronger protective association between SEC and suicide risk, and a weaker positive association between depression and suicide risk. Compared to students with low self-esteem, students with high self-esteem will show greater declines in depression and suicide risk as SEC rises, and a weaker rise in suicide risk in the face of depressive symptomology.