Skip to main content

Table 1 Parameter estimates from the simple regression models

From: Algorithms for converting estimates of child malnutrition based on the NCHS reference into estimates based on the WHO Child Growth Standards

  

Simple linear regression model in the logit1 scale: logit (PWHO) = a+b*logit (PNCHS)

Algorithm

N

Intercept a (95% CI)

Slope b (95% CI)

Correlation coefficient (R2)

Mean square error

Underweight

271

-0.177 (-0.231, -0.124)2

0.987 (0.955, 1.019)2

0.931

0.073

Stunting

271

0.216 (0.198, 0.235)2

0.925 (0.908, 0.941)2

0.979

0.012

Wasting

271

0.026 (-0.038, 0.090)

0.928 (0.905, 0.951)2

0.959

0.030

Overweight

256

0.235 (0.117, 0.353)2

0.912 (0.880, 0.944)2

0.925

0.064

  1. 1 logit (p) = log [P/(1-P)], where P is either PWHO or PNCHS, the WHO and NCHS prevalence estimates, respectively.
  2. 2 p-value < 0.05