A fully conditional specification approach to multilevel imputation of categorical and continuous variables

Craig K. Enders, Brian T. Keller, Roy Levy

Research output: Contribution to journalArticle

39 Scopus citations

Abstract

Specialized imputation routines for multilevel data are widely available in software packages, but these methods are generally not equipped to handle a wide range of complexities that are typical of behavioral science data. In particular, existing imputation schemes differ in their ability to handle random slopes, categorical variables, differential relations at Level-1 and Level-2, and incomplete Level-2 variables. Given the limitations of existing imputation tools, the purpose of this manuscript is to describe a flexible imputation approach that can accommodate a diverse set of 2-level analysis problems that includes any of the aforementioned features. The procedure employs a fully conditional specification (also known as chained equations) approach with a latent variable formulation for handling incomplete categorical variables. Computer simulations suggest that the proposed procedure works quite well, with trivial biases in most cases. We provide a software program that implements the imputation strategy, and we use an artificial data set to illustrate its use.

Original languageEnglish (US)
Pages (from-to)298-317
Number of pages20
JournalPsychological Methods
Volume23
Issue number2
DOIs
StatePublished - Jun 1 2018

Keywords

  • Imputation software
  • Missing data
  • Multilevel models
  • Multiple imputation

ASJC Scopus subject areas

  • Psychology (miscellaneous)

Fingerprint Dive into the research topics of 'A fully conditional specification approach to multilevel imputation of categorical and continuous variables'. Together they form a unique fingerprint.

  • Cite this