Abstract
A common problem in the meta analysis of continuous data is that some studies do not report sufficient information to calculate the standard deviation (SDs) of the treatment effect. One of the approaches in handling this problem is through imputation. This article examines the empirical implications of imputing the missing SDs on the standard error (SE) of the overall meta analysis estimate. The simulation results show that if the SDs are missing under Missing Completely at Random and Missing at Random mechanism, imputation is recommended. With non random missing, imputation can lead to overestimation of the SE of the estimate.
Notes
: the effect estimate based on all data; SE
all
: the SE based on all data;
: the between-study-variance based on all data;
: the SE based on MNAR data;
: the between-study-variance based on MNAR data;
: the SE based on single-value imputed data;
: the between-study-variance based on single-value imputed data;
: the difference in between-study variation based on all data and between study variance based on single-value imputed data.