Skip to main content

Table 4 Characteristics of the multiple logisitic regression models in VALID sampler

From: Development and validation of a dietary screener for carbohydrate intake in endurance athletes

model

Sensitivity

Specificity

False positives

False negatives

PPV

NPV

c statistic

5 variablesa

52.6

82.2

12.0

15.4

58.8

78.2

0.71

10 variables

75.4

86.4

9.1

8.0

72.9

87.9

0.90

15 variables

89.5

87.3

8.6

3.4

77.3

94.5

0.94

  1. PPV positive predictive value, NPV negative predictive value
  2. a % (all such values)