Updating latent class imputations with external auxiliary variables

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Latent class models are often used to assign values to categorical variables that cannot be measured directly. This "imputed" latent variable is then used in further analyses with auxiliary variables. The relationship between the imputed latent variable and auxiliary variables can only be correctly estimated if these auxiliary variables are included in the latent class model. Otherwise, point estimates will be biased. We develop a method that correctly estimates the relationship between an imputed latent variable and external auxiliary variables, by updating the latent variable imputations to be conditional on the external auxiliary variables using a combination of multiple imputation of latent classes and the so-called three-step approach. In contrast with existing "one-step" and "three-step" approaches, our method allows the resulting imputations to be analyzed using the familiar methods favored by substantive researchers.

LanguageEnglish
Pages750-761
JournalStructural Equation Modeling
Volume25
Issue number5
DOIs
StatePublished - 2018

Keywords

  • Latent class analysis
  • misclassification
  • multiple imputation
  • three-step approach
  • MULTIPLE-IMPUTATION
  • INFERENCE
  • VALUES
  • MODELS
  • SAMPLES

Cite this

@article{e6faa244bbd84af59c55abe0ea55024e,
title = "Updating latent class imputations with external auxiliary variables",
abstract = "Latent class models are often used to assign values to categorical variables that cannot be measured directly. This {"}imputed{"} latent variable is then used in further analyses with auxiliary variables. The relationship between the imputed latent variable and auxiliary variables can only be correctly estimated if these auxiliary variables are included in the latent class model. Otherwise, point estimates will be biased. We develop a method that correctly estimates the relationship between an imputed latent variable and external auxiliary variables, by updating the latent variable imputations to be conditional on the external auxiliary variables using a combination of multiple imputation of latent classes and the so-called three-step approach. In contrast with existing {"}one-step{"} and {"}three-step{"} approaches, our method allows the resulting imputations to be analyzed using the familiar methods favored by substantive researchers.",
keywords = "Latent class analysis, misclassification, multiple imputation, three-step approach, MULTIPLE-IMPUTATION, INFERENCE, VALUES, MODELS, SAMPLES",
author = "L. Boeschoten and Oberski, {Daniel L.} and {de Waal}, T. and J.K. Vermunt",
year = "2018",
doi = "10.1080/10705511.2018.1446834",
language = "English",
volume = "25",
pages = "750--761",
journal = "Structural Equation Modeling",
issn = "1070-5511",
publisher = "ROUTLEDGE JOURNALS, TAYLOR & FRANCIS LTD",
number = "5",

}

Updating latent class imputations with external auxiliary variables. / Boeschoten, L.; Oberski, Daniel L.; de Waal, T.; Vermunt, J.K.

In: Structural Equation Modeling, Vol. 25, No. 5, 2018, p. 750-761.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Updating latent class imputations with external auxiliary variables

AU - Boeschoten,L.

AU - Oberski,Daniel L.

AU - de Waal,T.

AU - Vermunt,J.K.

PY - 2018

Y1 - 2018

N2 - Latent class models are often used to assign values to categorical variables that cannot be measured directly. This "imputed" latent variable is then used in further analyses with auxiliary variables. The relationship between the imputed latent variable and auxiliary variables can only be correctly estimated if these auxiliary variables are included in the latent class model. Otherwise, point estimates will be biased. We develop a method that correctly estimates the relationship between an imputed latent variable and external auxiliary variables, by updating the latent variable imputations to be conditional on the external auxiliary variables using a combination of multiple imputation of latent classes and the so-called three-step approach. In contrast with existing "one-step" and "three-step" approaches, our method allows the resulting imputations to be analyzed using the familiar methods favored by substantive researchers.

AB - Latent class models are often used to assign values to categorical variables that cannot be measured directly. This "imputed" latent variable is then used in further analyses with auxiliary variables. The relationship between the imputed latent variable and auxiliary variables can only be correctly estimated if these auxiliary variables are included in the latent class model. Otherwise, point estimates will be biased. We develop a method that correctly estimates the relationship between an imputed latent variable and external auxiliary variables, by updating the latent variable imputations to be conditional on the external auxiliary variables using a combination of multiple imputation of latent classes and the so-called three-step approach. In contrast with existing "one-step" and "three-step" approaches, our method allows the resulting imputations to be analyzed using the familiar methods favored by substantive researchers.

KW - Latent class analysis

KW - misclassification

KW - multiple imputation

KW - three-step approach

KW - MULTIPLE-IMPUTATION

KW - INFERENCE

KW - VALUES

KW - MODELS

KW - SAMPLES

U2 - 10.1080/10705511.2018.1446834

DO - 10.1080/10705511.2018.1446834

M3 - Article

VL - 25

SP - 750

EP - 761

JO - Structural Equation Modeling

T2 - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

IS - 5

ER -