Latent Variable Selection for Multidimensional Item Response Theory Models via $$L_{1}$$L1 Regularization

We develop a latent variable selection method for multidimensional item response theory models. The proposed method identifies latent traits probed by items of a multidimensional test. Its basic strategy is to impose an $$L_{1}$$L1 penalty term to the log-likelihood. The computation is carried out by the expectation–maximization algorithm combined with the coordinate descent algorithm. Simulation studies show that the resulting estimator provides an effective way in correctly identifying the latent structures. The method is applied to a real dataset involving the Eysenck Personality Questionnaire.

[1]  A. Béguin,et al.  MCMC estimation and some model-fit analysis of multidimensional IRT models , 2001 .

[2]  Terry A. Ackerman.  Unidimensional IRT Calibration of Compensatory and Noncompensatory Multidimensional Items , 1989 .

[3]  Mark D. Reckase,et al.  The Use of the General Rasch Model with Multidimensional Item Response Data. , 1982 .

[4]  I. Johnstone,et al.  Adapting to Unknown Smoothness via Wavelet Shrinkage , 1995 .

[5]  C. L. Mallows.  Some comments on C_p , 1973 .

[6]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[7]  Roderick P. McDonald,et al.  Linear Versus Models in Item Response Theory , 1982 .

[8]  R. Tibshirani,et al.  PATHWISE COORDINATE OPTIMIZATION , 2007, 0708.1485.

[9]  Walter D. Way,et al.  The Comparative Effects of Compensatory and Noncompensatory Two-Dimensional Data on Unidimensional IRT Estimates , 1988 .

[10]  Li Cai,et al.  HIGH-DIMENSIONAL EXPLORATORY ITEM FACTOR ANALYSIS BY A METROPOLIS–HASTINGS ROBBINS–MONRO ALGORITHM , 2010 .

[11]  Colin Fraser,et al.  NOHARM: Least Squares Item Factor Analysis. , 1988, Multivariate behavioral research.

[12]  Daniel M. Bolt,et al.  Estimation of Compensatory and Noncompensatory Multidimensional Item Response Models Using Markov Chain Monte Carlo , 2003 .

[13]  Trevor Hastie,et al.  Regularization Paths for Generalized Linear Models via Coordinate Descent. , 2010, Journal of statistical software.

[14]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[15]  K. Jöreskog.  A general approach to confirmatory maximum likelihood factor analysis , 1969 .

[16]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[17]  S. Embretson.  A general latent trait model for response processes , 1984 .

[18]  Robert A. Forsyth,et al.  An Examination of the Characteristics of Unidimensional IRT Parameter Estimates Derived From Two-Dimensional Data , 1985 .

[19]  R. Mckinley.  Confirmatory Analysis of Test Structure Using Multidimensional Item Response Theory. , 1989 .

[20]  Alberto Maydeu-Olivares,et al.  Item diagnostics in multivariate discrete data. , 2015, Psychological methods.

[21]  Terry A. Ackerman.  Using multidimensional item response theory to understand what items and tests are measuring , 1994 .

[22]  C. Mallows.  Some Comments on Cp , 2000, Technometrics.

[23]  H. Akaike.  A new look at the statistical model identification , 1974 .

[24]  Roy Levy,et al.  An Overview of Software for Conducting Dimensionality Assessment in Multidimensional Models , 2012 .

[25]  Sybil B. G. Eysenck,et al.  Re-introduction to cross-cultural studies of the EPQ , 2013 .

[26]  M. Reckase,et al.  Development and Application of a Multivariate Logistic Latent Trait Model , 1972 .

[27]  M. Reckase.  The Past and Future of Multidimensional Item Response Theory , 1997 .

[28]  R. P. McDonald,et al.  Nonlinear factor analysis. , 1967 .

[29]  M. R. Novick,et al.  Statistical Theories of Mental Test Scores. , 1971 .

[30]  S. Embretson,et al.  Item response theory for psychologists , 2000 .

[31]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[32]  M. Reckase.  Multidimensional Item Response Theory , 2009 .

[33]  E. Muraki,et al.  Full-Information Item Factor Analysis , 1988 .

[34]  Li Cai.  High-dimensional Exploratory Item Factor Analysis by A Metropolis–Hastings Robbins–Monro Algorithm , 2010 .