Objective:To preliminarily establish a valuable threshold for the college students’ psychological Suzhi questionnaire simplified version (CSPSz-SV) by means of the latent profile analysis and the receiver operating characteristic(ROC) curve.Methods:The CSPSz-SV was applied to investigate 1 870 college freshmen from undergraduate colleges. The identification threshold value of the CSPSz-SV was identified by combining latent profile analysis (LPA) and the ROC curve.Both Mplus 8.3 and R 4.3.1 software were used for the data analysis.Results:Based on LPA results, the college freshmen were divided into three categories according to the levels of psychological Suzhi: the low psychological Suzhi group, the medium psychological Suzhi group, and the high psychological Suzhi group.Following the LPA, the ROC analysis was conducted to obtain the optimal critical value (91) of the psychological Suzhi level for college freshmen (Youden's index was 0.934, sensitivity was 0.979, specificity was 0.955, positive predictive value was 0.855, negative predictive value was 0.994, accuracy was 0.960, and AUC was 0.991, respectively). College freshmen were divided into the abnormal psychological Suzhi group and the normal psychological Suzhi group in accordance with this critical value, and the difference between the two groups was statistically significant (Cohen's d=1.74-2.18, P<0.001). Conclusion:The threshold value established and obtained in this study can be reviewed as a reference for evaluating the psychological Suzhi level of college freshmen.

Objective:To preliminarily establish a valuable threshold for the college students’ psychological Suzhi questionnaire simplified version (CSPSz-SV) by means of the latent profile analysis and the receiver operating characteristic(ROC) curve.Methods:The CSPSz-SV was applied to investigate 1 870 college freshmen from undergraduate colleges. The identification threshold value of the CSPSz-SV was identified by combining latent profile analysis (LPA) and the ROC curve.Both Mplus 8.3 and R 4.3.1 software were used for the data analysis.Results:Based on LPA results, the college freshmen were divided into three categories according to the levels of psychological Suzhi: the low psychological Suzhi group, the medium psychological Suzhi group, and the high psychological Suzhi group.Following the LPA, the ROC analysis was conducted to obtain the optimal critical value (91) of the psychological Suzhi level for college freshmen (Youden's index was 0.934, sensitivity was 0.979, specificity was 0.955, positive predictive value was 0.855, negative predictive value was 0.994, accuracy was 0.960, and AUC was 0.991, respectively). College freshmen were divided into the abnormal psychological Suzhi group and the normal psychological Suzhi group in accordance with this critical value, and the difference between the two groups was statistically significant (Cohen's d=1.74-2.18, P<0.001). Conclusion:The threshold value established and obtained in this study can be reviewed as a reference for evaluating the psychological Suzhi level of college freshmen.

Objective:To test the reliability of the short version of the depression-anxiety-stress scale (DASS-21) in adolescents through generalizability theory.Methods:Total of 458 adolescents from three middle schools in Guizhou and Sichuan province were surveyed with the DASS-21 from November to December 2021.Based on multivariate generalization theory(MGT), the generalization coefficients, reliability indexes and variance contribution ratios were calculated.mGENOVA 2.1 was used for data analysis.Results:The generalizability coefficient and reliability index for the total scale of DASS-21 were 0.927 and 0.922, respectively.The generalizability coefficients of depression, anxiety and stress subscales were 0.808-0.840, and the dependability indexes were 0.789-0.832.The contribution ratios of depression, anxiety and stress subscales to the general score were 33.20%, 33.94% and 32.86%, respectively.The stress subscale (34.95%) contributed the most to the relative error variance, and the anxiety subscale (36.98%) contributed the most to the absolute error variance.Conclusion:The DASS-21 has excellent test reliability in adolescents, and it can be used both as the norm-referenced test and the criterion-referenced test in the actual assessment of negative emotions in adolescents.