Skip to main content

Table 2 Estimated multilevel model for proportion of ages heaped

From: Trends in Demographic and Health Survey data quality: an analysis of age heaping over time in 34 countries in Sub Saharan Africa between 1987 and 2015

 

Model I: random intercept model

Model II: random slope model

Parameter estimate

95% confidence interval

Parameter estimate

95% confidence interval

Fixed effect parameters

 Survey year (centred)

0.004

(− 0.004, 0.012)

0.003

(− 0.010, 0.015)

 Intercept

− 3.032

 

− 3.025

 

Random effect parameters

 Random intercept ν 0j

0.836

(0.416,1.256)

0.786

(0.390,1.182)

 Random slope ν 1j

–

–

0.001

(0.000, 0.001)

 Intercept-slope covariance

–

–

0.004

(− 0.007, 0.015)

  1. Model based on 2nd order PQL RIGLS