Partial least squares path modeling using ordinal categorical indicators

Florian Schuberth, Jörg Henseler (Corresponding Author), Theo K. Dijkstra

Research output: Contribution to journalArticleAcademicpeer-review

66 Downloads (Pure)

Abstract

This article introduces a new consistent variance-based estimator called ordinal consistent partial least squares (OrdPLSc). OrdPLSc completes the family of variance-based estimators consisting of PLS, PLSc, and OrdPLS and permits to estimate structural equation models of composites and common factors if some or all indicators are measured on an ordinal categorical scale. A Monte Carlo simulation (N =500 ) with different population models shows that OrdPLSc provides almost unbiased estimates. If all constructs are modeled as common factors, OrdPLSc yields estimates close to those of its covariance-based counterpart, WLSMV, but is less efficient. If some constructs are modeled as composites, OrdPLSc is virtually without competition.
Original languageEnglish
Pages (from-to)9-35
Number of pages27
JournalQuality and quantity
Volume52
Issue number1
Early online date14 Sep 2016
DOIs
Publication statusPublished - 2018

Fingerprint

Partial Least Squares
Categorical
Common factor
Path
structural model
Composite
Modeling
Estimate
Estimator
Structural Equation Model
Population Model
simulation
Monte Carlo Simulation

Keywords

  • UT-Hybrid-D
  • Structural equation models
  • Consistent partial least squares
  • Ordinal categorical indicators
  • Common factors
  • Composites
  • Polychoric correlation

Cite this

@article{8ddc4e61b8ac4ebb9c49aa416278f77b,
title = "Partial least squares path modeling using ordinal categorical indicators",
abstract = "This article introduces a new consistent variance-based estimator called ordinal consistent partial least squares (OrdPLSc). OrdPLSc completes the family of variance-based estimators consisting of PLS, PLSc, and OrdPLS and permits to estimate structural equation models of composites and common factors if some or all indicators are measured on an ordinal categorical scale. A Monte Carlo simulation (N =500 ) with different population models shows that OrdPLSc provides almost unbiased estimates. If all constructs are modeled as common factors, OrdPLSc yields estimates close to those of its covariance-based counterpart, WLSMV, but is less efficient. If some constructs are modeled as composites, OrdPLSc is virtually without competition.",
keywords = "UT-Hybrid-D, Structural equation models, Consistent partial least squares, Ordinal categorical indicators, Common factors, Composites, Polychoric correlation",
author = "Florian Schuberth and J{\"o}rg Henseler and Dijkstra, {Theo K.}",
note = "Springer deal",
year = "2018",
doi = "10.1007/s11135-016-0401-7",
language = "English",
volume = "52",
pages = "9--35",
journal = "Quality and quantity",
issn = "0033-5177",
publisher = "Springer",
number = "1",

}

Partial least squares path modeling using ordinal categorical indicators. / Schuberth, Florian ; Henseler, Jörg (Corresponding Author); Dijkstra, Theo K.

In: Quality and quantity, Vol. 52, No. 1, 2018, p. 9-35.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Partial least squares path modeling using ordinal categorical indicators

AU - Schuberth, Florian

AU - Henseler, Jörg

AU - Dijkstra, Theo K.

N1 - Springer deal

PY - 2018

Y1 - 2018

N2 - This article introduces a new consistent variance-based estimator called ordinal consistent partial least squares (OrdPLSc). OrdPLSc completes the family of variance-based estimators consisting of PLS, PLSc, and OrdPLS and permits to estimate structural equation models of composites and common factors if some or all indicators are measured on an ordinal categorical scale. A Monte Carlo simulation (N =500 ) with different population models shows that OrdPLSc provides almost unbiased estimates. If all constructs are modeled as common factors, OrdPLSc yields estimates close to those of its covariance-based counterpart, WLSMV, but is less efficient. If some constructs are modeled as composites, OrdPLSc is virtually without competition.

AB - This article introduces a new consistent variance-based estimator called ordinal consistent partial least squares (OrdPLSc). OrdPLSc completes the family of variance-based estimators consisting of PLS, PLSc, and OrdPLS and permits to estimate structural equation models of composites and common factors if some or all indicators are measured on an ordinal categorical scale. A Monte Carlo simulation (N =500 ) with different population models shows that OrdPLSc provides almost unbiased estimates. If all constructs are modeled as common factors, OrdPLSc yields estimates close to those of its covariance-based counterpart, WLSMV, but is less efficient. If some constructs are modeled as composites, OrdPLSc is virtually without competition.

KW - UT-Hybrid-D

KW - Structural equation models

KW - Consistent partial least squares

KW - Ordinal categorical indicators

KW - Common factors

KW - Composites

KW - Polychoric correlation

U2 - 10.1007/s11135-016-0401-7

DO - 10.1007/s11135-016-0401-7

M3 - Article

VL - 52

SP - 9

EP - 35

JO - Quality and quantity

JF - Quality and quantity

SN - 0033-5177

IS - 1

ER -