Depression, anxiety, and student satisfaction with university life among college students: a cross-lagged study

The study addresses the high global prevalence of depression and anxiety and their impact on college students’ well-being and academic life. University students are particularly vulnerable to mental health problems during this transitional stage, which can lead to poor academic performance, insomnia, dropout, and suicidality. Concurrently, student satisfaction with university life—students’ perceptions and evaluations of the overall campus environment—is an important indicator of physical and mental well-being and is relevant for institutional quality and reputation. The research aims to clarify the longitudinal relationships among depression, anxiety, and student satisfaction with university life using large-sample data from Chinese college students and a cross-lagged design. The study proposes three contributions: (1) it is among the few longitudinal, large-sample studies in China on these relationships; (2) it innovatively applies the customer satisfaction model to explain links among the constructs; and (3) it helps resolve conflicting findings regarding anxiety and satisfaction and deepens understanding of predictive relationships. Hypotheses: I) Depression and anxiety are negatively correlated with student satisfaction; II) Depression and anxiety negatively predict later student satisfaction; III) Student satisfaction negatively predicts later depression and anxiety.

The literature indicates that depression (persistent sadness, hopelessness) and anxiety (heightened worry) adversely affect students’ perceived quality of life and well-being. Many studies report negative correlations between depression/anxiety and life satisfaction, though findings on anxiety and satisfaction are mixed, with some showing strong negative links (especially during COVID-19) and others suggesting potential positive or non-significant relations depending on context and coping. Longitudinal evidence on predictive relations is limited: depression often predicts lower life satisfaction; anxiety, conceptualized as stemming from persistent negative thoughts, may reduce life satisfaction, though several studies report anxiety as a non-significant predictor. Conversely, higher life or study satisfaction has been found to predict lower depressive symptoms and, in some studies, lower anxiety. Moderators such as interpersonal relationships and social support can buffer adverse effects. The customer satisfaction model provides a theoretical lens: discrepancies between expectations and experiences (in educational services) can shape satisfaction, which may in turn influence mental health. Research gaps include a paucity of large-scale Chinese longitudinal studies, predominance of cross-sectional designs, limited integration of depression and anxiety within customer satisfaction frameworks, and conflicting results about anxiety–satisfaction links.

Design: Two-wave longitudinal survey with a one-year interval; cross-lagged panel models (CLPM) were used to examine bidirectional predictive relations among depression, anxiety, and student satisfaction with university life (SSUL). Participants: Chinese college students; N=2298 at Time 1 (T1, junior year), N=2070 followed at Time 2 (T2, senior year). Age 18–28 (M=21.550, SD=0.895). Participation was voluntary with informed consent; ethics approval TJUE-2022-188. Measures: - Depression: DASS-42 Depression subscale (14 items, 0–3). Higher scores indicate higher depression; categories per DASS norms. Cronbach’s alpha: T1=0.9004, T2=0.9141. - Anxiety: DASS-42 Anxiety subscale (14 items, 0–3). Higher scores indicate higher anxiety; categories per DASS norms. Alphas: T1=0.8477, T2=0.8746. - Student Satisfaction with University Life (SSUL): 9 items (Teaching facilities; Teachers’ research capabilities; Teaching abilities; Academic status in the country; Systematic nature of courses; Usefulness of courses; Extracurricular activities; Student–teacher relationships; Learning atmosphere). Each 1 (very poor) to 10 (excellent); summed score; higher indicates greater satisfaction. Alphas: T1=0.9234, T2=0.9239. Control variables: Gender, age, extroversion (single item 1–9), family social status (1–5 ladder from lower to upper class). Analysis: Descriptive statistics and correlations (Stata 15.0). CLPMs estimated in Mplus 8.3 with five models for each pair: autoregressive (M1), preceding (M2), outcome (M3), interaction (M4), control (M5, adds gender, age, extroversion, family social status). Model fit indices: CFI, TLI (>0.90 acceptable), RMSEA (<0.10), SRMR (<0.10). Chi-square/df not emphasized due to large N.

Descriptive and correlations: - Depression and SSUL were significantly negatively correlated at both time points (T1 r=-0.146, T2 r=-0.088, p<0.05; small effects). - Anxiety and SSUL were negatively correlated at T1 (r=-0.114, p<0.05; small) but not at T2 (r=-0.030, ns). - Depression and anxiety were strongly positively correlated at both T1 (r=0.684) and T2 (r=0.756) (large effects; p<0.05). - SSUL increased from T1 to T2 (means: 61.270 vs. 63.673; t-test p<0.05). Cross-lagged models: Depression ↔ SSUL (Tables 2–3) - Model fit acceptable across models (e.g., M1: CFI=0.967, TLI=0.956, RMSEA=0.076, SRMR=0.032). - Depression T1 → SSUL T2: negative, significant (β≈-0.060 to -0.064; p≈0.002–0.003) across M2, M4, M5. - SSUL T1 → Depression T2: negative, significant (β≈-0.067 to -0.069; p=0.001) across M3, M4, M5. - Autoregressive paths significant (e.g., Depression β≈0.55–0.57; SSUL β≈0.61–0.63). - Indicates a bidirectional negative relationship between depression and SSUL over time. Cross-lagged models: Anxiety ↔ SSUL (Tables 4–5) - Model fit acceptable (e.g., M1: CFI=0.969, TLI=0.959, RMSEA=0.071, SRMR=0.022). - Anxiety T1 → SSUL T2: not significant (β≈-0.021 to -0.024; p>0.26) across M2, M4, M5. - SSUL T1 → Anxiety T2: negative, significant (β≈-0.042 to -0.046; p≈0.030–0.047) across M3, M4, M5. - Autoregressive paths significant (Anxiety β≈0.59; SSUL β≈0.62). Overall: - Depression at T1 predicts lower SSUL at T2; anxiety at T1 does not predict SSUL. - Higher SSUL at T1 predicts lower depression and anxiety at T2. - Findings robust after controlling for gender, age, extroversion, and family social status.

The study clarifies longitudinal dynamics among depression, anxiety, and student satisfaction with university life (SSUL) in Chinese college students. Consistent with Hypothesis I, depression correlates negatively with SSUL, while anxiety’s concurrent relation with SSUL is small and not consistent across time points. Cross-lagged analyses support Hypothesis II partially: depression (but not anxiety) negatively predicts later SSUL. In line with Hypothesis III, SSUL negatively predicts later depression and anxiety. These results suggest that depressive symptoms exert a more stable detrimental influence on future satisfaction than anxiety does, whereas enhancing SSUL can reduce subsequent internalizing symptoms. The customer satisfaction model helps explain how negative affect may lower perceived service quality and life evaluations over time, while improved educational environments and student experiences can mitigate mental health problems. Practically, universities should enhance teaching quality, campus climate, student–teacher relationships, and support services to raise SSUL and thereby buffer depression and anxiety. Attention to students with higher depressive symptoms is especially warranted given their downstream impact on satisfaction.

- Depression and student satisfaction with university life are negatively correlated; anxiety shows no consistent negative correlation with satisfaction across time points. - Depression predicts lower future student satisfaction with university life, whereas anxiety does not significantly predict later satisfaction. - Student satisfaction with university life negatively predicts both subsequent depression and anxiety, indicating that improving satisfaction can help reduce these symptoms. Future research should expand to diverse contexts and refine measurement, including validated instruments for satisfaction and personality/social status, and explore moderators and mechanisms.

- Self-report measures for depression, anxiety, SSUL, extroversion, and family social status may introduce measurement error. - Although depression and SSUL are significantly negatively correlated, the effect sizes are small. - Sample comprised higher-grade (junior-to-senior) university students; results may not generalize to lower-grade cohorts or other educational stages. - The SSUL instrument (9-item scale) was not previously validated; retest reliability and prior references are lacking. - Extroversion and social status were assessed via single unvalidated items.