Skip to main content

Table 5 Classification performance for the obstetric data example (N = 10574): logistic regression (LR) and SEM with Q-metric input (see Figure 4)

From: Beyond logistic regression: structural equations modelling for binary variables and its application to investigating unobserved confounders

Correctly classified (%)

Dependent variables

 

Caesarian section

Low birthweight

Special Care Baby Unit

 

LR *

SEM

LR *

SEM

LR *

SEM

All outcomes

85.7

83.8

95.4

83.9

95.3

82.7

Events

0.0

15.5

26.7

72.7

36.5

69.0

Non events

100.0

95.2

99.1

84.6

99.2

83.6

  1. * Note: Logistic regression used the same outcome and predictor variables as the SEM model in Figure 4 but needed separate logistic models for each dependent variable.
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us