Skip to main content

Table 2 ROC analysis of the SAPAS-AV and subgroup comparison of AUC

From: Screening for personality disorder in incarcerated adolescent boys: preliminary validation of an adolescent version of the standardised assessment of personality – abbreviated scale (SAPAS-AV)

Sample

AUC (95% CI)

Standard error

Comparison of AUC difference

p 1

Whole sample at baseline (N = 80)

0.89**** (0.80 to 0.95)

0.04

0.74

Whole sample at follow-up (N = 80)

0.90**** (0.81 to 0.96)

0.03

Group 1 (n = 40)

0.89**** (0.75 to 0.97)

0.06

0.93

Group 2 (n = 40)

0.89**** (0.75 to 0.96)

0.05

Site 1 (n = 34)

0.86**** (0.70 to 0.96)

0.07

0.47

Site 2 (n = 38)

0.92**** (0.78 to 0.98)

0.05

  1. Note. ****AUC is significantly different from the non-informative line (AUC = 0.50) at the level of p < 0.0001. 1Two-tailed z-test comparing differences between AUC for the two ROC-curves.