Suicidal Ideation in Graduate Students: Psychometric Validation of the Brazilian Version of the Frequency of Suicidal...

Graduate students are a particularly vulnerable group in terms of suicidal behavior, which highlights the need for psychometrically robust instruments for assessing suicidal ideation within academic environments.

To evaluate the validity evidence of the Frequency of Suicidal Ideation Inventory - Brazilian version among Brazilian graduate students.

A total of 701 participants participated in this psychometric study. The construct validity of the model was evaluated using Confirmatory Factor Analysis, as well as the model fit indices and unidimensionality metrics. The reliability and precision of the scores were assessed using McDonald's omega coefficient, marginal reliability, determinacy and sensitivity indices, as well as the factor replicability index. The correlation with the Kessler Psychological Distress Scale was used to evaluate convergent validity, whereas criterion-related validity was assessed using ROC analysis, taking self-reported suicide attempts within the past 12 months into account as the external criterion.

This instrument demonstrated a unidimensional structure, which was supported by high factor loadings (λ = .806-.928) and substantial communalities (0.649-0.861). The presence of a dominant general factor was confirmed by the unidimensionality index (UniCo = 0.999; Explained Common Variance (ECV) = 0.958; Mean of Item Residual Absolute Loadings (MIREAL) = 0.176). The model fit was satisfactory (Comparative Fit Index (CFI) = 0.998; Tucker-Lewis Index (TLI) = 0.995; Standardized Root Mean Square Residual (SRMR) = 0.031), although the RMSEA was high (0.096), likely due to the low degrees of freedom in the model. The reliability and precision of scores were excellent (ω = .958; H = 0.954; marginal reliability = 0.950; factor determinacy = 0.975). Convergent validity was supported by a moderate positive correlation with the Kessler Psychological Distress Scale (ρ = 0.583; p < .001). An ROC analysis indicated excellent discriminatory capability (AUC = 0.921), sensitivity of 91.67%, and specificity of 86.79%.

The results demonstrate robust evidence of structural validity, high score precision, and excellent discriminatory performance, supporting the use of the FSII-Br as a reliable tool for assessing suicidal ideation among Brazilian graduate students.