Validity, Reliability, and Factor Structure of the Suicide Crisis Scale in Turkish.

Abstract

In our study, we aimed to adapt the Suicide Crisis Inventory (SCI), which can be used specifically to assess the acute phase of suicide, to the Turkish population by examining its Turkish validity and reliability in a non-clinical sample. In this cross-sectional study, a total of 300 university students aged 18-24 years were evaluated online using the Socio-demographic and Clinical Data Form, the SCI, and the Suicide Behavior Questionnaire (SBQ). Criterion validity, discriminative validity, and factor analyses (exploratory and confirmatory) were conducted for the validity of the SCI, and internal consistency and item-total correlations were examined for reliability analyses. Additionally, a linear regression model was constructed to assess the predictive validity of the SCI. The predictive validity of past SCI scores was evaluated using a simple regression model. When the linear regression model was tested with SCI scores as the independent variable and SBQ scores as the dependent variable [F(1-298) = 203.625; P = .000], it was found that the independent variable explained 41% of the variance in the dependent variable (r = 0.637; r 2 = 0.406). SCI scores significantly predicted SBQ scores (t = 14.270; B = 0.047; Bsth = 0.003; β = 0.647; P = .000). In the validity analysis, the items removed from the scale could be evaluated for the total score, as they did not belong to any factor as originally specified. When items were removed, the total item reliability was Cronbach's alpha = 0.981. We believe that the SCI will be a useful tool in assessing short-term suicide risk in a Turkish sample and in conducting scientific research. The SCI was found to be sufficient for use in a Turkish sample for the evaluation of short-term suicide risk, considering some limitations.