The data for this study were obtained from a nationwide serial survey—the Chinese Social Mentality Survey (CSMS)—conducted by the Chinese Academy of Social Sciences (CASS). The survey was administered through the “Questionnaire Treasure” APP in January, 2021. Duringthe questionnaire collection process, screening questions, trap questions, and logic check questions were set up, and only participants who passed all of them were finally included in the sample. A total of 1,200 questionnaires were distributed in the study, and 1,010 valid samples were obtained, resulting in an effective response rate of 84.17%. Respondents were drawn from 5 places including Xi’an city in Shanxi Province, Jinnan District in Tianjin City, Zhengzhou City, Anyang City, Xuchang City in Henan Province. Detailed demographic information of the participants is provided in Table 1.

Mediation effects were tested using the bias-corrected percentile Bootstrap with 5000 repetitions of resampling in the study method (35). The mediation model analysis was performed using SPSS 25.0 in combination with the SPSS Macro PROCESS (36). Pearson correlation analysis was employed to explore relationships between variables. Further, a path analysis was conducted to assess a hypothesized mediation model concerning the effect of various factors on depression.

The Harman’s single-factor test was conducted to test the common method bias of the used items in the self-reported scales of community cohesion, social support, resilience, and depression. The results showed that the first common factor accounted for 34.70% of the total variance, which was below the critical threshold of 40% (37). Confirmatory factor analysis was used to further examine common method bias using Mplus8.3, and the fit result of the single-factor model was very poor (χ²/df = 12.90, CFI = 0.62, TLI = 0.60, SRMR = 0.11, RMSEA = 0.11). However, the indicators in four-factor model show good fit (χ²/df = 4.74, CFI = 0.88, TLI = 0.87, SRMR = 0.05, RMSEA = 0.06). The findings indicated that there was no significant common method bias in the current study.

Our correlational analysis revealed significant relationships among community cohesion, social support, resilience, and depression. Specifically, community cohesion negatively correlated with depression scores, indicating that greater community cohesion was associated with less depression (r = -0.21, p < 0.01). Moreover, social support scores were negatively linked to depression scores (r = -0.27, p < 0.05), and positively correlated with psychological resilience (r = 0.32, p < 0.01), suggesting that social support contributes to less depression and high psychological resilience. Psychological resilience was negatively correlated with depression scores (r = -0.37, p 185 < 0.01), indicating that higher psychological resilience is associated with less depression. Also, community cohesion is significantly related to social support (r = 0.39, p < 0.001) and psychological resilience (r = 0.35, p < 0.001) (Table 2).

Using the SPSS Macro PROCESS developed by Hayes (36), the mediating roles of social support and psychological resilience in the relationship between community cohesion and depression among adults was analyzed, while controlling for gender, age, education years and income Regression analysis results (as presented in Table 3, Figure 1) was shown that community cohesion positively affected social support (β = 0.54, t = 13.63, p < 0.001, 95% CI = [0.46, 0.62]); Social support (β = 0.13, t =6.91, p < 0.001, 95% CI = [0.09, 0.17])and community cohesion (β = 0.23, t = 8.79, p < 0.001, 95% CI = [0.18, 0.28]) had positive effect on psychological resilience. Both social support (β = -0.09, t = -5.05, p < 0.001, 95% CI = [-0.12, -0.05] )and psychological resilience (β = -0.25, t = -8.82, p < 0.001, 95% CI = [-0.30, -0.19]) negatively affected adult depression. Community cohesion demonstrated no significant direct effect on depression (β = -0.04, t = -1.67, p > 0.05, 95% CI = [-0.09, 0.01]), but a significant total negative effect on depression (β = -0.16, t = -7.17, p < 0.001, 95% CI = [-0.21, -012]).

Social support and psychological resilience jointly mediated the relationship between community cohesion and depression, with a total mediation effect value of -0.12, accounting for 75% of the total effect (-0.12/-0.16). Specifically, the mediation effect comprised three indirect pathways: Pathway 1 (Community Cohesion → Social Support → Depression) demonstrated a significant mediating effect of -0.05 (95% CI [-0.07, -0.03]), accounting for 31.25% of the total effect (-0.05/-0.16); Pathway 2 (Community Cohesion → Psychological Resilience → Depression) showed a significant mediating effect of −0.06(95% CI [-0.08, -0.04]), accounting for 37.50% of the total effect (-0.06/-0.16); Pathway 3 (Community Cohesion → Social Support → Psychological Resilience → Depression) produced a significant mediating effect of −0.02(95% CI [-0.03, -0.01], accounting for 6.25% of the total effect (-0.02/-0.16) (Table 4).

Correlation analysis has shown that there was a negative relation between community cohesion and depression, which validated Hypothesis 1 of this study. Consistent with previous studies (7, 9), the finding suggested individuals living in higher community cohesion may have less risk of getting depression. According to Social Disorganization Theory, cohesion among members in a neighborhood, is a proximate mechanism through which neighborhood may affect individuals (6). On the one hand, communities with high cohesion provide residents with more resources and convenient facilities, and this can make residents adopt and reinforce healthy behaviors (46). On the other hand, highly cohesive communities can reduce individuals’ sense of self-reliance by offering more opportunities for interaction and enhancing the sense of purpose (21, 38). Moreover, a socially cohesive community can enhance individuals’ restorative experiences (39), thereby exerting a protective effect on their physical and mental health. Regression analysis revealed that the total effect of community cohesion on depression was significant, but the direct effect was not. This finding helps explain Abada et al.’s observation that perceived cohesion was unrelated to depression (40). We believe the reason is although community cohesion does not directly affect depression, it may exert an indirect influence through mediating variables such as social support and psychological resilience (as demonstrated in this study).

Consistent with prior research (41), our findings revealed that social support mediated association between community cohesion and depression, which validated Hypothesis 2 of this study. According to the social capital theory (42), communities with strong cohesion are more likely to develop “bridging social capital,” which facilitates cross-group support. Social support can encourage positive health behaviors through established social norms (41) and enhance self-efficacy, or promote confidence (43), which are protective factors against individual depression.

Our finding demonstrated that psychological resilience served as a significant mediating factor in the relationship between community cohesion and depression, which validated Hypothesis 3 of this study. This finding aligns with prior research (25, 26), which suggests that in a community with higher cohesion, individuals are more likely to develop greater psychological resilience, which is a protective factor against depression (44). This is in accordance with findings that children who lived in relatively high crime neighborhoods with low social cohesion were less likely to be resilient versus non-resilient to maltreatment (45). Ruiz and his colleagues confirmed that the association between low social cohesion and increased depressive symptoms operated via a psychological, but not a health behavioral, pathway (10). Our study extends their work by investigating how community cohesion influences depression through psychological resilience.

This study revealed a weaker but significant chain mediation of social support and psychological resilience, which validated Hypothesis 4 of this study. According to Conservation of Resources Theory (29), community cohesion may provide social support and a climate of encouragement, feelings of being understood and mutual respect, which in turn, help individuals to adopt healthy behaviors, enhance sense of control or mastery and use active coping strategies (14, 46, 47). The findings support Bronfenbrenner’s Ecological Systems Theory (48) and provide a comprehensive mechanism through which community cohesion fosters social support, and social support bolsters resilience, ultimately alleviating depression.

In summary, Individuals in communities with high cohesion can establish good social relationships with others based on mutual trust, gain more emotional and material social support among residents. When they encounter difficulties and stress in life, they will actively seek neighborhood support or develop the ability to recover quickly from difficulties, which can prevent or alleviate depression. While this model suggested in this study can reveal the mechanisms underlying the prevention or alleviation of depression well, it also has phenomena that it cannot explain. For instance, in collectivist cultures, individuals in communities with high cohesion are more sensitive to group norms and thus tend to suppress their own emotions, which will cause individuals to develop negative emotions (49) Therefore, this model will be improved in future research.