Skip to main content

Table 5 Type 1 error rate for non-response imputation, dichotomizing before multiply imputing, and multiply imputing before dichotomizing when missing =30%1

From: Imputation strategies when a continuous outcome is to be dichotomized for responder analysis: a simulation study

 

Null response profile 1

Null response profile 2

Dropout model 1

Dropout model 4

Dropout model 1

Dropout model 4

NRI

0.06

0.16

0.05

0.31

DTI MI

0.03

0.04

0.03

0.04

DTI MI with CV

0.03

0.04

0.03

0.04

IBD MI

0.02

0.02

0.03

0.03

IBD MI with CV

0.03

0.02

0.03

0.04

  1. NRI: Non-response imputation; DTI MI: Dichotomize then impute multiple imputation; IBD MI: Impute before dichotomizing multiple imputation
  2. 1Using Dropout model 1 and 4
  3. 2 Null response profile 1: =(65, 65, 65, 65)′; Null response profile 2: μA = (65, 67, 69, 71)′ and μB = (65, 63, 68, 71)′
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us