Understanding Suicidal Ideation Among University Students: An Ecological Analysis of Risk and Protective Factors

Abstract

Suicidal ideation among college students is a multicausal public health concern shaped by individual, family, social, and academic factors. Although prior research has consistently documented associations between psychological distress and suicide risk in this population, most studies analyze risk factors in isolation, leaving the joint contribution of variables across ecological levels insufficiently understood. The present study, framed within Bronfenbrenner’s ecological model of human development and Joiner’s interpersonal theory of suicide, examined the simultaneous influence of clinical, protective, academic, and victimization variables on suicidal ideation in college students. A cross-sectional design was employed with a sample of 553 undergraduate and graduate students (68.9% women; mean age = 20.93 years, SD = 3.81) from the University of Málaga (Spain), recruited through snowball and convenience sampling. Participants completed validated self-report measures of suicidal ideation (Paykel Suicide Scale), depression and anxiety (HADS), eating disorders (CBCAR), self-esteem (RSE), academic stress (SISCO), academic procrastination (APS-SF), bullying, cyberbullying, and social exclusion (QAEU), social support (QFSSS), family satisfaction, healthy habits, and family history of mental illness. Data were analyzed through Pearson and Spearman correlations and a robust multiple linear regression (ordinary least squares with HC3 robust standard errors and bias-corrected and accelerated bootstrap confidence intervals from 5000 resamples), complemented by an ordinal regression as a sensitivity analysis. The regression model accounted for 44.1% of the variance in suicidal ideation, with depression, anxiety, eating disorders, family history of mental illness, and bullying showing significant positive associations and self-esteem the strongest protective association. An ordinal regression sensitivity analysis yielded a consistent pattern, supporting the robustness of these findings. Family and social support, although correlated with lower ideation, lost statistical weight in the multivariate models, suggesting that their contribution operates indirectly, mainly through depression and self-esteem. Findings support the integration of clinical, social, and personal-resource components into college prevention programs, prioritizing early detection of psychopathology, anti-bullying protocols, interventions to strengthen self-esteem, and greater visibility of campus psychological services.