On using multiple imputation for exploratory factor analysis of incomplete data

Vahid Nassiri; Anikó Lovik; Geert Molenberghs; Geert Verbeke

doi:10.3758/s13428-017-1013-4

On using multiple imputation for exploratory factor analysis of incomplete data

Behav Res Methods. 2018 Apr;50(2):501-517. doi: 10.3758/s13428-017-1013-4.

Authors

Vahid Nassiri¹, Anikó Lovik², Geert Molenberghs^{2

3}, Geert Verbeke^{2

3}

Affiliations

¹ KU Leuven, BioStat, 3000, Leuven, Belgium. vahid.nassiri@kuleuven.be.
² KU Leuven, BioStat, 3000, Leuven, Belgium.
³ Universiteit Hasselt, Biostat, 3590, Diepenbeek, Belgium.

PMID: 29392587
DOI: 10.3758/s13428-017-1013-4

Abstract

A simple multiple imputation-based method is proposed to deal with missing data in exploratory factor analysis. Confidence intervals are obtained for the proportion of explained variance. Simulations and real data analysis are used to investigate and illustrate the use and performance of our proposal.

Keywords: Exploratory factor analysis; Missing data; Multiple imputation; Principal component analysis.

Publication types

Research Support, Non-U.S. Gov't
Review

MeSH terms

Algorithms
Computer Simulation
Confidence Intervals
Data Interpretation, Statistical*
Factor Analysis, Statistical*
Humans
Principal Component Analysis