A test for multigroup comparison using partial least squares path modeling

Michael Klesel, Florian Schuberth, Jörg Henseler, Bjoern Niehaves

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Purpose: People seem to function according to different models, which implies that in business and social sciences, heterogeneity is a rule rather than an exception. Researchers can investigate such heterogeneity through multigroup analysis (MGA). In the context of partial least squares path modeling (PLS-PM), MGA is currently applied to perform multiple comparisons of parameters across groups. However, this approach has significant drawbacks: first, the whole model is not considered when comparing groups, and second, the family-wise error rate is higher than the predefined significance level when the groups are indeed homogenous, leading to incorrect conclusions. Against this background, the purpose of this paper is to present and validate new MGA tests, which are applicable in the context of PLS-PM, and to compare their efficacy to existing approaches. Design/methodology/approach: The authors propose two tests that adopt the squared Euclidean distance and the geodesic distance to compare the model-implied indicator correlation matrix across groups. The authors employ permutation to obtain the corresponding reference distribution to draw statistical inference about group differences. A Monte Carlo simulation provides insights into the sensitivity and specificity of both permutation tests and their performance, in comparison to existing approaches. Findings: Both proposed tests provide a considerable degree of statistical power. However, the test based on the geodesic distance outperforms the test based on the squared Euclidean distance in this regard. Moreover, both proposed tests lead to rejection rates close to the predefined significance level in the case of no group differences. Hence, our proposed tests are more reliable than an uncontrolled repeated comparison approach. Research limitations/implications: Current guidelines on MGA in the context of PLS-PM should be extended by applying the proposed tests in an early phase of the analysis. Beyond our initial insights, more research is required to assess the performance of the proposed tests in different situations. Originality/value: This paper contributes to the existing PLS-PM literature by proposing two new tests to assess multigroup differences. For the first time, this allows researchers to statistically compare a whole model across groups by applying a single statistical test.

Original languageEnglish
Pages (from-to)464-477
Number of pages13
JournalInternet Research
Volume29
Issue number3
DOIs
Publication statusPublished - 1 Jan 2019

Fingerprint

Social sciences
Statistical tests
Group
Partial least squares
Modeling
Industry
statistical test
performance
social science
simulation
Multi-group analysis
methodology
Values
Monte Carlo simulation
Euclidean distance

Keywords

  • Monte Carlo simulation
  • Multigroup analysis
  • Partial least squares path modelling
  • Permutation test

Cite this

Klesel, Michael ; Schuberth, Florian ; Henseler, Jörg ; Niehaves, Bjoern. / A test for multigroup comparison using partial least squares path modeling. In: Internet Research. 2019 ; Vol. 29, No. 3. pp. 464-477.
@article{e80d696426524bbab53fa842f779e323,
title = "A test for multigroup comparison using partial least squares path modeling",
abstract = "Purpose: People seem to function according to different models, which implies that in business and social sciences, heterogeneity is a rule rather than an exception. Researchers can investigate such heterogeneity through multigroup analysis (MGA). In the context of partial least squares path modeling (PLS-PM), MGA is currently applied to perform multiple comparisons of parameters across groups. However, this approach has significant drawbacks: first, the whole model is not considered when comparing groups, and second, the family-wise error rate is higher than the predefined significance level when the groups are indeed homogenous, leading to incorrect conclusions. Against this background, the purpose of this paper is to present and validate new MGA tests, which are applicable in the context of PLS-PM, and to compare their efficacy to existing approaches. Design/methodology/approach: The authors propose two tests that adopt the squared Euclidean distance and the geodesic distance to compare the model-implied indicator correlation matrix across groups. The authors employ permutation to obtain the corresponding reference distribution to draw statistical inference about group differences. A Monte Carlo simulation provides insights into the sensitivity and specificity of both permutation tests and their performance, in comparison to existing approaches. Findings: Both proposed tests provide a considerable degree of statistical power. However, the test based on the geodesic distance outperforms the test based on the squared Euclidean distance in this regard. Moreover, both proposed tests lead to rejection rates close to the predefined significance level in the case of no group differences. Hence, our proposed tests are more reliable than an uncontrolled repeated comparison approach. Research limitations/implications: Current guidelines on MGA in the context of PLS-PM should be extended by applying the proposed tests in an early phase of the analysis. Beyond our initial insights, more research is required to assess the performance of the proposed tests in different situations. Originality/value: This paper contributes to the existing PLS-PM literature by proposing two new tests to assess multigroup differences. For the first time, this allows researchers to statistically compare a whole model across groups by applying a single statistical test.",
keywords = "Monte Carlo simulation, Multigroup analysis, Partial least squares path modelling, Permutation test",
author = "Michael Klesel and Florian Schuberth and J{\"o}rg Henseler and Bjoern Niehaves",
note = "Klesel, M., Schuberth, F., Henseler, J., & Niehaves, B. (2019). A test for multigroup comparison using partial least squares path modeling. Internet Research, 29(3), 464-477. https://doi.org/10.1108/IntR-11-2017-0418",
year = "2019",
month = "1",
day = "1",
doi = "10.1108/IntR-11-2017-0418",
language = "English",
volume = "29",
pages = "464--477",
journal = "Internet Research",
issn = "1066-2243",
publisher = "Emerald Group Publishing Ltd.",
number = "3",

}

A test for multigroup comparison using partial least squares path modeling. / Klesel, Michael; Schuberth, Florian; Henseler, Jörg; Niehaves, Bjoern.

In: Internet Research, Vol. 29, No. 3, 01.01.2019, p. 464-477.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A test for multigroup comparison using partial least squares path modeling

AU - Klesel, Michael

AU - Schuberth, Florian

AU - Henseler, Jörg

AU - Niehaves, Bjoern

N1 - Klesel, M., Schuberth, F., Henseler, J., & Niehaves, B. (2019). A test for multigroup comparison using partial least squares path modeling. Internet Research, 29(3), 464-477. https://doi.org/10.1108/IntR-11-2017-0418

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Purpose: People seem to function according to different models, which implies that in business and social sciences, heterogeneity is a rule rather than an exception. Researchers can investigate such heterogeneity through multigroup analysis (MGA). In the context of partial least squares path modeling (PLS-PM), MGA is currently applied to perform multiple comparisons of parameters across groups. However, this approach has significant drawbacks: first, the whole model is not considered when comparing groups, and second, the family-wise error rate is higher than the predefined significance level when the groups are indeed homogenous, leading to incorrect conclusions. Against this background, the purpose of this paper is to present and validate new MGA tests, which are applicable in the context of PLS-PM, and to compare their efficacy to existing approaches. Design/methodology/approach: The authors propose two tests that adopt the squared Euclidean distance and the geodesic distance to compare the model-implied indicator correlation matrix across groups. The authors employ permutation to obtain the corresponding reference distribution to draw statistical inference about group differences. A Monte Carlo simulation provides insights into the sensitivity and specificity of both permutation tests and their performance, in comparison to existing approaches. Findings: Both proposed tests provide a considerable degree of statistical power. However, the test based on the geodesic distance outperforms the test based on the squared Euclidean distance in this regard. Moreover, both proposed tests lead to rejection rates close to the predefined significance level in the case of no group differences. Hence, our proposed tests are more reliable than an uncontrolled repeated comparison approach. Research limitations/implications: Current guidelines on MGA in the context of PLS-PM should be extended by applying the proposed tests in an early phase of the analysis. Beyond our initial insights, more research is required to assess the performance of the proposed tests in different situations. Originality/value: This paper contributes to the existing PLS-PM literature by proposing two new tests to assess multigroup differences. For the first time, this allows researchers to statistically compare a whole model across groups by applying a single statistical test.

AB - Purpose: People seem to function according to different models, which implies that in business and social sciences, heterogeneity is a rule rather than an exception. Researchers can investigate such heterogeneity through multigroup analysis (MGA). In the context of partial least squares path modeling (PLS-PM), MGA is currently applied to perform multiple comparisons of parameters across groups. However, this approach has significant drawbacks: first, the whole model is not considered when comparing groups, and second, the family-wise error rate is higher than the predefined significance level when the groups are indeed homogenous, leading to incorrect conclusions. Against this background, the purpose of this paper is to present and validate new MGA tests, which are applicable in the context of PLS-PM, and to compare their efficacy to existing approaches. Design/methodology/approach: The authors propose two tests that adopt the squared Euclidean distance and the geodesic distance to compare the model-implied indicator correlation matrix across groups. The authors employ permutation to obtain the corresponding reference distribution to draw statistical inference about group differences. A Monte Carlo simulation provides insights into the sensitivity and specificity of both permutation tests and their performance, in comparison to existing approaches. Findings: Both proposed tests provide a considerable degree of statistical power. However, the test based on the geodesic distance outperforms the test based on the squared Euclidean distance in this regard. Moreover, both proposed tests lead to rejection rates close to the predefined significance level in the case of no group differences. Hence, our proposed tests are more reliable than an uncontrolled repeated comparison approach. Research limitations/implications: Current guidelines on MGA in the context of PLS-PM should be extended by applying the proposed tests in an early phase of the analysis. Beyond our initial insights, more research is required to assess the performance of the proposed tests in different situations. Originality/value: This paper contributes to the existing PLS-PM literature by proposing two new tests to assess multigroup differences. For the first time, this allows researchers to statistically compare a whole model across groups by applying a single statistical test.

KW - Monte Carlo simulation

KW - Multigroup analysis

KW - Partial least squares path modelling

KW - Permutation test

UR - http://www.scopus.com/inward/record.url?scp=85063036478&partnerID=8YFLogxK

UR - http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=Alerting&SrcApp=Alerting&DestApp=WOS_CPL&DestLinkType=FullRecord&UT=WOS:000475944800005

U2 - 10.1108/IntR-11-2017-0418

DO - 10.1108/IntR-11-2017-0418

M3 - Article

VL - 29

SP - 464

EP - 477

JO - Internet Research

JF - Internet Research

SN - 1066-2243

IS - 3

ER -