A framework of statistical tests for comparing mean and covariance structure models

Roy Levy, Gregory R. Hancock

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

Although statistical procedures are well-known for comparing hierarchically related (nested) mean and covariance structure models, statistical tests for comparing non-hierarchically related (nonnested) models have proven more elusive. Although isolated attempts at statistical tests of non-hierarchically related models have been made, none exist within the commonly used maximum likelihood estimation framework, thereby compromising these methods' accessibility and general applicability. Building on general theory developed by Vuong (1989) and techniques for establishing the relation between covariance structure models (Raykov & Penev, 1999), this work provides a general paradigm for conducting statistical tests on competing mean and covariance structure models. The proposed framework is appropriate for hierarchically related models as well as non-hierarchically related models. In developing the structure of the framework, key aspects of model equivalence, relation, and comparison are unified. An illustration demonstrates its use.

Original languageEnglish (US)
Pages (from-to)33-66
Number of pages34
JournalMultivariate Behavioral Research
Volume42
Issue number1
StatePublished - 2007

Fingerprint

statistical test
Covariance Structure
Statistical test
Statistical Models
Model
Framework
Statistical Tests
Equivalence relation
statistical method
equivalence
Accessibility
Maximum Likelihood Estimation
Paradigm
paradigm
Demonstrate

ASJC Scopus subject areas

  • Mathematics (miscellaneous)
  • Statistics and Probability
  • Social Sciences (miscellaneous)
  • Psychology(all)
  • Experimental and Cognitive Psychology

Cite this

A framework of statistical tests for comparing mean and covariance structure models. / Levy, Roy; Hancock, Gregory R.

In: Multivariate Behavioral Research, Vol. 42, No. 1, 2007, p. 33-66.

Research output: Contribution to journalArticle

@article{01aba5cde9684394b68fc525aded392a,
title = "A framework of statistical tests for comparing mean and covariance structure models",
abstract = "Although statistical procedures are well-known for comparing hierarchically related (nested) mean and covariance structure models, statistical tests for comparing non-hierarchically related (nonnested) models have proven more elusive. Although isolated attempts at statistical tests of non-hierarchically related models have been made, none exist within the commonly used maximum likelihood estimation framework, thereby compromising these methods' accessibility and general applicability. Building on general theory developed by Vuong (1989) and techniques for establishing the relation between covariance structure models (Raykov & Penev, 1999), this work provides a general paradigm for conducting statistical tests on competing mean and covariance structure models. The proposed framework is appropriate for hierarchically related models as well as non-hierarchically related models. In developing the structure of the framework, key aspects of model equivalence, relation, and comparison are unified. An illustration demonstrates its use.",
author = "Roy Levy and Hancock, {Gregory R.}",
year = "2007",
language = "English (US)",
volume = "42",
pages = "33--66",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "1",

}

TY - JOUR

T1 - A framework of statistical tests for comparing mean and covariance structure models

AU - Levy, Roy

AU - Hancock, Gregory R.

PY - 2007

Y1 - 2007

N2 - Although statistical procedures are well-known for comparing hierarchically related (nested) mean and covariance structure models, statistical tests for comparing non-hierarchically related (nonnested) models have proven more elusive. Although isolated attempts at statistical tests of non-hierarchically related models have been made, none exist within the commonly used maximum likelihood estimation framework, thereby compromising these methods' accessibility and general applicability. Building on general theory developed by Vuong (1989) and techniques for establishing the relation between covariance structure models (Raykov & Penev, 1999), this work provides a general paradigm for conducting statistical tests on competing mean and covariance structure models. The proposed framework is appropriate for hierarchically related models as well as non-hierarchically related models. In developing the structure of the framework, key aspects of model equivalence, relation, and comparison are unified. An illustration demonstrates its use.

AB - Although statistical procedures are well-known for comparing hierarchically related (nested) mean and covariance structure models, statistical tests for comparing non-hierarchically related (nonnested) models have proven more elusive. Although isolated attempts at statistical tests of non-hierarchically related models have been made, none exist within the commonly used maximum likelihood estimation framework, thereby compromising these methods' accessibility and general applicability. Building on general theory developed by Vuong (1989) and techniques for establishing the relation between covariance structure models (Raykov & Penev, 1999), this work provides a general paradigm for conducting statistical tests on competing mean and covariance structure models. The proposed framework is appropriate for hierarchically related models as well as non-hierarchically related models. In developing the structure of the framework, key aspects of model equivalence, relation, and comparison are unified. An illustration demonstrates its use.

UR - http://www.scopus.com/inward/record.url?scp=34547448618&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547448618&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:34547448618

VL - 42

SP - 33

EP - 66

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 1

ER -