Skip to main content

Table 1 Average percentage of missing values imputed outside the specified range using linear regression imputation, by scoring method

From: Comparison of methods for imputing limited-range variables: a simulation study

  GHQ raw scale GHQ transformed
  % imputed below lower bound % imputed above upper bound % imputed below lower bound % imputed above upper bound
Scoring method MCAR MAR MCAR MAR MCAR MAR MCAR MAR
Likert 2.0 2.0 0.0 0.0 0.0 0.0 0.2 0.2
C-GHQ 12.2 11.6 0.1 0.1 7.9 7.4 1.8 0.2
Standard 23.0 24.0 0.0 0.0 14.3 14.9 6.5 7.5
  1. Note: Mean number of missing values per dataset is 238.