Partial least squares path modeling using ordinal categorical indicators

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 $$=500$$=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.

[1]  F. Bookstein,et al.  Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory , 1982 .

[2]  G. Cantaluppi,et al.  A Partial Least Squares Algorithm Handling Ordinal Variables , 2014 .

[3]  Looking at the Antecedents of Perceived Switching Costs. A PLS Path Modeling Approach with Categorical Indicators , 2005 .

[4]  Jeffrey M. Wooldridge,et al.  Introductory Econometrics: A Modern Approach , 1999 .

[5]  K. Pearson.  Mathematical contributions to the theory of evolution. VIII. On the correlation of characters not quantitatively measurable , 2022, Proceedings of the Royal Society of London.

[6]  Ulf Olsson,et al.  Maximum likelihood estimation of the polychoric correlation coefficient , 1979 .

[7]  Francisco Pablo Holgado Tello,et al.  Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables , 2010 .

[8]  C. Saunders,et al.  Editor's comments: PLS: a silver bullet? , 2006 .

[9]  Studies of the polychoric correlation and other correlation measures for ordinal variables , 1992 .

[10]  Christian Homburg,et al.  The loss of the marketing department’s influence: is it really happening? And why worry? , 2015 .

[11]  Marko Sarstedt,et al.  PLS-SEM: Indeed a Silver Bullet , 2011 .

[12]  F. Drasgow,et al.  The polyserial correlation coefficient , 1982 .

[13]  Mikko Rönkkö,et al.  Matrix-Based Partial Least Squares Estimation , 2015 .

[14]  Sönke Albers,et al.  PLS and Success Factor Studies in Marketing , 2010 .

[15]  T. Dijkstra.  Latent Variables and Indices: Herman Wold’s Basic Design and Partial Least Squares , 2010 .

[16]  Taeke Klaas Dijkstra.  Latent variables in linear stochastic models , 1981 .

[17]  Peter M. Bentler,et al.  A three-stage estimation procedure for structural equation models with polytomous variables , 1990 .

[18]  J. Carroll.  The nature of the data, or how to choose a correlation coefficient , 1961 .

[19]  Y. Takane,et al.  Generalized Structured Component Analysis: A Component-Based Approach to Structural Equation Modeling , 2014 .

[20]  Jörg Henseler,et al.  Consistent and asymptotically normal PLS estimators for linear structural equations , 2014 .

[21]  F. Bookstein,et al.  Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory: , 1982 .

[22]  K. Pearson.  On the Measurement of the Influence of “Broad Categories” on Correlation , 1913 .

[23]  Analyzing Ordered Categorical Data derived from Elliptically Symmetric Distributions , 1998 .

[24]  V. Savalei,et al.  When can categorical variables be treated as continuous? A comparison of robust continuous and categorical SEM estimation methods under suboptimal conditions. , 2012, Psychological methods.

[25]  Marko Sarstedt,et al.  An assessment of the use of partial least squares structural equation modeling in marketing research , 2012 .

[26]  Peter M. Bentler,et al.  Structural equation models with continuous and polytomous variables , 1992 .

[27]  Peter F. Halpin,et al.  Manifest and Latent Variates , 2008 .

[28]  Jörg Henseler,et al.  Consistent Partial Least Squares Path Modeling , 2015, MIS Q..

[29]  T. Dijkstra.  Some comments on maximum likelihood and partial least squares methods , 1983 .

[30]  Walter Krämer,et al.  Review of Modern applied statistics with S, 4th ed. by W.N. Venables and B.D. Ripley. Springer-Verlag 2002 , 2003 .

[31]  Myrsini Katsikatsou,et al.  Pairwise likelihood estimation for factor analysis models with ordinal data , 2012, Comput. Stat. Data Anal..

[32]  H. Wold.  Path Models with Latent Variables: The NIPALS Approach , 1975 .

[33]  Los Angeles,et al.  Multivariate Ordinal Data Analysis with Pairwise Likelihood and Its Extension to SEM , 2007 .

[34]  Theo K. Dijkstra.  Consistent Partial Least Squares estimators for linear and polynomial factor models. A report of a belated, serious and not even unsuccessful attempt. Comments are invited , 2011 .

[35]  Fred L. Bookstein,et al.  Partial least squares analysis in developmental psychopathology , 1989, Development and Psychopathology.

[36]  Sik-Yum Lee,et al.  Two-step estimation of multivariate polychoric correlation , 1987 .

[37]  V. Savalei.  What to Do About Zero Frequency Cells When Estimating Polychoric Correlations , 2011 .

[38]  J. Kettenring,et al.  Canonical Analysis of Several Sets of Variables , 2022 .

[39]  S. Wold,et al.  Orthogonal projections to latent structures (O‐PLS) , 2002 .

[40]  Giorgio Russolillo,et al.  Non-Metric Partial Least Squares , 2012 .

[41]  T. Dijkstra,et al.  Consistent Partial Least Squares for Nonlinear Structural Equation Models , 2013, Psychometrika.

[42]  Forrest W. Young.  Quantitative analysis of qualitative data , 1981 .

[43]  B. Muthén.  A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators , 1984 .

[44]  Michel Tenenhaus,et al.  PLS path modeling , 2005, Comput. Stat. Data Anal..

[45]  Peter B. Wylie,et al.  Effects of Coarse Grouping and Skewed Marginal Distributions on the Pearson Product Moment Correlation Coefficient , 1976 .

[46]  M. Sarstedt,et al.  A new criterion for assessing discriminant validity in variance-based structural equation modeling , 2015 .

[47]  Jan Faber,et al.  Consistent estimation of correlations between observed interval variables with skewed distributions , 1988 .

[48]  Yves Rosseel,et al.  lavaan: An R Package for Structural Equation Modeling , 2012 .

[49]  G. Yule.  On the Association of Attributes in Statistics: With Illustrations from the Material of the Childhood Society, &c , 1900 .

[50]  D. Straub,et al.  Editor's comments: a critical look at the use of PLS-SEM in MIS quarterly , 2012 .

[51]  Shawn Bauldry,et al.  Three Cs in measurement models: causal indicators, composite indicators, and covariates. , 2011, Psychological methods.

[52]  Peter M. Bentler,et al.  Full maximum likelihood analysis of structural equation models with polytomous variables , 1990 .

[53]  Ulf Olsson,et al.  Measuring Correlation in Ordered Two-Way Contingency Tables , 1980 .

[54]  Wai-Yin Poon,et al.  Maximum likelihood estimation of multivariate polyserial and polychoric correlation coefficients , 1988 .

[55]  Edward E. Rigdon,et al.  Rethinking Partial Least Squares Path Modeling: In Praise of Simple Methods , 2012 .

[56]  William N. Venables,et al.  Modern Applied Statistics with S , 2010 .

[57]  P. Homer,et al.  Corrections for coarsely categorized measures: LISREL's polyserial and polychoric correlations , 1987 .

[58]  A. R. de Leon.  Pairwise likelihood approach to grouped continuous model and its extension , 2005 .

[59]  P. Coelho,et al.  The Choice between a Fivepoint and a Ten-point Scale in the Framework of Customer Satisfaction Measurement , 2007 .

[60]  Joerg Henseler.  Why generalized structured component analysis is not universally preferable to structural equation modeling , 2012 .

[61]  R. P. McDonald,et al.  Path Analysis with Composite Variables. , 1996, Multivariate behavioral research.

[62]  Jan-Bernd Lohmöller,et al.  Latent Variable Path Modeling with Partial Least Squares , 1989 .

[63]  Christian Derquenne,et al.  A modified PLS path modeling algorithm handling reflective categorical variables and a new model building strategy , 2007, Comput. Stat. Data Anal..

[64]  Marc A. Tomiuk,et al.  A Comparative Study on Parameter Recovery of Three Approaches to Structural Equation Modeling , 2010 .

[65]  Jacob Cohen,et al.  Applied multiple regression/correlation analysis for the behavioral sciences , 1979 .

[66]  F. Holgado-Tello,et al.  Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables , 2008 .

[67]  Kristina Höök,et al.  Strong concepts: Intermediate-level knowledge in interaction design research , 2012, TCHI.

[68]  Herman Wold,et al.  Soft modelling: The Basic Design and Some Extensions , 1982 .

[69]  G. Cantaluppi.  A Partial Least Squares Algorithm Handling Ordinal Variables also in Presence of a Small Number of Categories , 2012, 1212.5049.

[70]  Marko Sarstedt,et al.  Goodness-of-fit indices for partial least squares path modeling , 2013, Comput. Stat..

[71]  K. Bollen,et al.  Pearson's R and Coarsely Categorized Measures , 1981 .

[72]  Detmar W. Straub,et al.  Common Beliefs and Reality About PLS , 2014 .

[73]  Fritz Drasgow,et al.  Polychoric and Polyserial Correlations , 2006 .

[74]  M. Tenenhaus.  Component-based Structural Equation Modelling , 2008 .