Maximum likelihood versus multiple imputation for missing data in small longitudinal samples with nonnormality

Tacksoo Shin; Mark L Davison; Jeffrey D Long

doi:10.1037/met0000094

Back

Maximum likelihood versus multiple imputation for missing data in small longitudinal samples with nonnormality

Journal article

Peer reviewed

Maximum likelihood versus multiple imputation for missing data in small longitudinal samples with nonnormality

Tacksoo Shin, Mark L Davison and Jeffrey D Long

Psychological methods, Vol.22(3), pp.426-449

09/2017

DOI: 10.1037/met0000094

PMID: 27709974

View Online

Abstract

The study examined the performance of maximum likelihood (ML) and multiple imputation (MI) procedures for missing data in longitudinal research when fitting latent growth models. A Monte Carlo simulation study was conducted with conditions of small sample size, intermittent missing data, and nonnormality. The results indicated that ML tended to display slightly smaller degrees of bias than MI across missing completely at random (MCAR) and missing at random (MAR) conditions. Although specification of prior information in the MI imputation-posterior (I-P) phase influenced the performance of MI, especially with nonnormal small samples and missing not at random (MNAR), the impact of this tight specification was not dramatic. Several corrected ML test statistics showed proper rejections rates across research designs, whereas posterior predictive p values for MI methods were more likely to be influenced by distribution shape and yielded higher rejection rates in MCAR and MAR than in MNAR. In conclusion, ML appears to be preferable to MI in research conditions with small missing samples and multivariate nonnormality whether or not strong prior information for the I-P phase of MI analysis is available. (PsycINFO Database Record

Bias

Data Interpretation, Statistical

Humans

Likelihood Functions

Models, Statistical

Monte Carlo Method

Probability

Psychological Tests - statistics & numerical data

Sample Size

Details

Title: Subtitle: Maximum likelihood versus multiple imputation for missing data in small longitudinal samples with nonnormality
Creators: Tacksoo Shin - Myongji University
Mark L Davison - University of Minnesota
Jeffrey D Long - University of Iowa
Resource Type: Journal article
Publication Details: Psychological methods, Vol.22(3), pp.426-449
DOI: 10.1037/met0000094
PMID: 27709974
ISSN: 1082-989X
eISSN: 1939-1463
Language: English
Date published: 09/2017
Academic Unit: Psychiatry; Biostatistics
Record Identifier: 9984280838502771

Metrics

38 Record Views

98 Times Cited - Web of Science