Table 3.

Sensitivity and specificity of risk-perception measures (n = 566)

Approach A
Approach B
Approach C
Approach D
NVCNVCNVCNVC
Very high sensitivity0.890.370.890.910.750.860.890.311.000.890.370.90
Very high specificity0.990.930.990.990.980.960.361.001.000.901.000.99
Very low sensitivity0.740.810.74
Very low specificity0.960.950.96
  • NOTE: Approach A: Using cutoff of 50% for numeric, very high for verbal, 6 on seven-point comparative measure (7—much higher than average) for predicting high risk, or 5% for numeric, very low for verbal, 2 on seven-point comparative measure (1—much lower than average) for predicting low risk. Approach B: Using cutoff of 50% numeric, very high for verbal, 6 on seven-point comparative measure (7—much higher than average) for predicting high risk, and dropping participants with a 50% perceived risk on the numerical scale for predicting high risk. Approach C: Using cutoff of 20% for numeric, very high for verbal, 6 on seven-point comparative measure (7—much higher than average) for predicting high risk. Approach D: Using cutoff of 20% for numeric, very high for verbal, 6 on seven-point comparative measure (7—much higher than average) and dropping participants with a 50% perceived risk on the numerical scale for predicting high risk.

    Abbreviations: N, numeric; V, verbal; C, comparative.