The effect of simple imputation on inferences about population means when data are missing in biomedical research due to detection limits

Hongyue Wang*, Guangqing Chen, Xiang Lu, Hui Zhang, Changyong Feng

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Summary: The sample geometric mean has been widely used in biomedical and psychosocial research to estimate and compare population geometric means. However, due to the detection limit of measurement instruments, the actual value of the measurement is not always observable. A common practice to deal with this problem is to replace missing values by small positive constants and make inferences based on the imputed data. However, no work has been carried out to study the effect of this naïve imputation method on inference. In this report, we show that this simple imputation method may dramatically change the reported outcomes of a study and, thus, make the results uninterpretable, even if the detection limit is very small.

Original languageEnglish (US)
Pages (from-to)319-325
Number of pages7
JournalShanghai Archives of Psychiatry
Volume27
Issue number5
DOIs
StatePublished - Oct 1 2015

Keywords

  • Population geometric mean
  • Sample geometric mean
  • Two-sample test

ASJC Scopus subject areas

  • Neurology
  • Psychiatry and Mental health

Fingerprint Dive into the research topics of 'The effect of simple imputation on inferences about population means when data are missing in biomedical research due to detection limits'. Together they form a unique fingerprint.

  • Cite this