Two-level linear paired comparison models: estimation and identifiability issues

Abstract The method of paired comparisons became popular in psychological research with Thurstone’s [Psychometrika 65 (1927) 233] demonstration that attitudes can be scaled along a one-dimensional continuum. Despite a large number of applications of this method over the years, it has been noted only recently that paired comparison data do not only contain information about item mean differences but are also useful for studying how individuals differ in their evaluative judgments. We show that a mixed-effects, generalized linear model is well-suited for investigating such individuals differences and present a Monte Carlo EM algorithm for parameter estimation. In addition, we discuss identification issues in the specifications of different covariance structures because they impose important constraints on the interpretation of model parameters. An extensive analysis of a value study employing ordinal paired comparison illustrates the proposed statistical framework.

[1]  R. Duncan Luce,et al.  Individual Choice Behavior , 1959 .

[2]  M. Rokeach.  The Nature Of Human Values , 1974 .

[3]  R. A. Bradley,et al.  Rank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons , 1952 .

[4]  Wim P. M. Vijverberg,et al.  Monte Carlo evaluation of multivariate normal probabilities , 1997 .

[5]  U. Böckenholt,et al.  Maximum likelihood estimation of factor and ideal point models for paired comparison data , 2001 .

[6]  P. Doyle,et al.  Random utility models in marketing research: a survey , 2001 .

[7]  R. A. Bradley,et al.  RANK ANALYSIS OF INCOMPLETE BLOCK DESIGNS THE METHOD OF PAIRED COMPARISONS , 1952 .

[8]  H. A. David,et al.  The method of paired comparisons , 1966 .

[9]  Paul A. Ruud,et al.  Simulation of multivariate normal rectangle probabilities and their derivatives theoretical and computational results , 1996 .

[10]  N. T. Gridgeman.  Pair Comparison, with and without Ties , 1959 .

[11]  Ulf Böckenholt,et al.  Modeling within-subject dependencies in ordinal paired comparison data , 1997 .

[12]  R. Tsai,et al.  Remarks on the identifiability of thurstonian ranking models: Case V, case III, or neither? , 2000 .

[13]  B. R. Dansie.  Normal order statistics as permutation probability models , 1986 .

[14]  D. Rubin,et al.  Parameter expansion to accelerate EM: The PX-EM algorithm , 1998 .

[15]  Patrick Bossuyt,et al.  A Comparison of Probabilistic Unfolding Theories for Paired Comparisons Data , 1990 .

[16]  Hal S. Stern,et al.  Models for Distributions on Permutations , 1990 .

[17]  S. Schwartz,et al.  Toward A Universal Psychological Structure of Human Values , 1987 .

[18]  U. Böckenholt.  Thurstonian representation for partial ranking data , 1992 .

[19]  U. Böckenholt,et al.  Hierarchical modeling of paired comparison data. , 2001, Psychological methods.

[20]  P. Acker.  Transitivity revisited , 1990 .

[21]  L. V. Jones,et al.  The measurement and prediction of judgment and choice. , 1970 .

[22]  S. Schwartz.  Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries , 1992 .

[23]  D. Harville.  Matrix Algebra From a Statistician's Perspective , 1998 .

[24]  Xiao-Li Meng,et al.  Fast EM‐type implementations for mixed effects models , 1998 .

[25]  Michael W. Browne,et al.  Circumplex models for correlation matrices , 1992 .

[26]  H. Conte,et al.  Circumplex models of personality and emotions , 1997 .

[27]  Yoshio Takane,et al.  Analysis of Covariance Structures and Probabilistic Binary Choice Data , 1989 .

[28]  Albert Maydeu-Olivares,et al.  Limited information estimation and testing of Thurstonian models for preference data , 2002, Math. Soc. Sci..

[29]  Joseph G. Ibrahim,et al.  Monte Carlo Methods in Bayesian Computation , 2000 .

[30]  B. Mellers,et al.  Similarity and Choice. , 1994 .

[31]  R. D. Bock.  Remarks on the test of significance for the method of paired comparisons , 1958 .

[32]  William G. Cochran,et al.  Experimental Designs, 2nd Edition , 1950 .

[33]  Juan de Dios Ortúzar,et al.  Representation of heteroskedasticity in discrete choice models , 2000 .

[34]  S. Schwartz,et al.  Toward a theory of the universal content and structure of values: Extensions and cross-cultural replications. , 1990 .

[35]  L. L. Thurstone,et al.  An internal consistency check for scale values determined by the method of successive intervals , 1952 .

[36]  Rung-Ching Tsai.  Analysis of Paired Comparison Data Using Monte Carlo EM Algorithms , 2001 .

[37]  A. Kuk,et al.  MAXIMUM LIKELIHOOD ESTIMATION FOR PROBIT-LINEAR MIXED MODELS WITH CORRELATED RANDOM EFFECTS , 1997 .

[38]  L. Thurstone.  A law of comparative judgment. , 1994 .

[39]  K. Train,et al.  Forecasting new product penetration with flexible substitution patterns , 1998 .

[40]  M. Kendall,et al.  Rank Correlation Methods , 1949 .

[41]  N. T. Gridgeman,et al.  Statistics and Taste Testing , 1960 .

[42]  P. Lazarsfeld,et al.  Mathematical Thinking in the Social Sciences. , 1955 .

[43]  Tim B. Swartz,et al.  Approximating Integrals Via Monte Carlo and Deterministic Methods , 2000 .

[44]  Vassilis A. Hajivassiliou,et al.  Simulation Estimation Methods for Limited Dependent Variable Models , 1991 .