Item Response Theory with Estimation of the Latent Population Distribution Using Spline-Based Densities

The purpose of this paper is to introduce a new method for fitting item response theory models with the latent population distribution estimated from the data using splines. A spline-based density estimation system provides a flexible alternative to existing procedures that use a normal distribution, or a different functional form, for the population distribution. A simulation study shows that the new procedure is feasible in practice, and that when the latent distribution is not well approximated as normal, two-parameter logistic (2PL) item parameter estimates and expected a posteriori scores (EAPs) can be improved over what they would be with the normal model. An example with real data compares the new method and the extant empirical histogram approach.

[1]  Trevor Hastie,et al.  [Flexible Parsimonious Smoothing and Additive Modeling]: Discussion , 1989 .

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

[3]  James O. Ramsay,et al.  Multicategorical spline model for item response theory , 1992 .

[4]  Frederick F. Stephan,et al.  The Expected Value and Variance of the Reciprocal and Other Negative Powers of a Positive Bernoullian Variate , 1945 .

[5]  J. Ramsay.  Kernel smoothing approaches to nonparametric item characteristic curve estimation , 1991 .

[6]  J. Coyne,et al.  Examining symptom expression as a function of symptom severity: item performance on the Hamilton Rating Scale for Depression. , 2001, Psychological assessment.

[7]  Efficient nonparametric approaches for estimating the operating characteristics of discrete item responses , 1998 .

[8]  James O. Ramsay,et al.  Differential equation models for statistical functions , 2000 .

[9]  L. A. Goodman,et al.  Measures of association for cross classifications , 1979 .

[10]  Shelby J. Haberman,et al.  Dispersion of Categorical Variables and Penalty Functions: Derivation, Estimation, and Comparability , 1995 .

[11]  James O. Ramsay,et al.  Binomial Regression with Monotone Splines: A Psychometric Application , 1989 .

[12]  Carl de Boor,et al.  A Practical Guide to Splines , 1978, Applied Mathematical Sciences.

[13]  Tae-Je Seong.  Sensitivity of Marginal Maximum Likelihood Estimation of Item and Ability Parameters to the Characteristics of the Prior Ability Distributions , 1990 .

[14]  R. D. Bock,et al.  Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm , 1981 .

[15]  L. J. Savage.  Elicitation of Personal Probabilities and Expectations , 1971 .

[16]  R. Hambleton,et al.  Fundamentals of Item Response Theory , 1991 .

[17]  A. Liebetrau.  Measures of association , 1983 .

[18]  John R. Boulet.  The effect of nonnormal ability distributions on IRT parameter estimation using full-information and limited-information methods. , 1996 .

[19]  G. Marsaglia,et al.  A New Class of Random Number Generators , 1991 .

[20]  F. J. Anscombe,et al.  On estimating binomial response relations , 1956 .

[21]  David Thissen,et al.  Item Response Theory for Items Scored in Two Categories , 2001 .

[22]  L. Kirisci,et al.  The Robustness of BILOG to Violations of the Assumptions of Unidimensionality of Test Items and Normality of Ability Distribution. , 1995 .

[23]  F. O’Sullivan.  Fast Computation of Fully Automated Log-Density and Log-Hazard Estimators , 1988 .

[24]  Robert J. Mislevy,et al.  BILOG 3 : item analysis and test scoring with binary logistic models , 1990 .

[25]  J. W. Getzels,et al.  Role conflict and personality. , 1955, Journal of personality.

[26]  H. Chernoff.  A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations , 1952 .

[27]  Dean Follmann,et al.  Consistent estimation in the rasch model based on nonparametric margins , 1988 .

[28]  E. B. Andersen,et al.  Estimating the parameters of the latent population distribution , 1977 .

[29]  J. Friedman,et al.  FLEXIBLE PARSIMONIOUS SMOOTHING AND ADDITIVE MODELING , 1989 .

[30]  Wendy M. Yen,et al.  A comparison of the efficiency and accuracy of BILOG and LOGIST , 1987 .

[31]  Georg Rasch,et al.  Probabilistic Models for Some Intelligence and Attainment Tests , 1981, The SAGE Encyclopedia of Research Design.

[32]  Tue Tjur,et al.  A Connection between Rasch's Item Analysis Model and a Multiplicative Poisson Model , 1982 .

[33]  L. Kirisci,et al.  Robustness of Item Parameter Estimation Programs to Assumptions of Unidimensionality and Normality , 2001 .

[34]  C. J. Stone,et al.  Logspline Density Estimation for Censored Data , 1992 .

[35]  Marie Davidian,et al.  The Nonlinear Mixed Effects Model with a Smooth Random Effects Density , 1993 .

[36]  S. Haberman,et al.  Prediction functions for categorical panel data , 1995 .

[37]  Edwin J. C. G. van den Oord,et al.  Estimating Johnson Curve Population Distributions in MULTILOG , 2005 .

[38]  M Davidian,et al.  Linear Mixed Models with Flexible Distributions of Random Effects for Longitudinal Data , 2001, Biometrics.

[39]  M. Cox.  The Numerical Evaluation of B-Splines , 1972 .

[40]  J. S. Roberts,et al.  Characteristics of MML/EAP Parameter Estimates in the Generalized Graded Unfolding Model , 2002 .

[41]  Lee J. Cronbach,et al.  Psychological tests and personnel decisions , 1958 .

[42]  R. Mislevy.  Estimating latent distributions , 1984 .

[43]  S. Reise,et al.  Parameter Recovery in the Graded Response Model Using MULTILOG , 1990 .

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

[45]  R. R. Bahadur,et al.  On Deviations of the Sample Mean , 1960 .

[46]  David Thomas,et al.  The Art in Computer Programming , 2001 .

[47]  Ivo W. Molenaar,et al.  Loglinear Rasch Model Tests , 2005 .

[48]  A. Gallant,et al.  Semi-nonparametric Maximum Likelihood Estimation , 1987 .

[49]  Noel A Cressie,et al.  Characterizing the manifest probabilities of latent trait models , 1983 .

[50]  J. Ramsay,et al.  Examining scale discriminability in the BDI and CES-D as a function of depressive severity. , 1995 .

[51]  S. Haberman.  Analysis of Dispersion of Multinomial Responses , 1982 .

[52]  Item Parameter Recovery for the Nominal Response Model , 1999 .

[53]  A. Raftery.  Bayes Factors and BIC , 1999 .

[54]  N. L. Johnson,et al.  Systems of frequency curves generated by methods of translation. , 1949, Biometrika.

[55]  A. Zwinderman,et al.  Robustness of Marginal Maximum Likelihoo Estimation in the Rasch odel , 1990 .

[56]  P. Holland.  On the sampling theory roundations of item response theory models , 1990 .

[57]  D. Thissen,et al.  Loglinear Item Response Models, with Applications to Data from Social Surveys , 1989 .

[58]  B. Silverman,et al.  On the Estimation of a Probability Density Function by the Maximum Penalized Likelihood Method , 1982 .

[59]  J. Coyne,et al.  Shortening the CES-D to improve its ability to detect cases of depression , 1997 .

[60]  Paul H. C. Eilers,et al.  Flexible smoothing with B-splines and penalties , 1996 .

[61]  E. B. Andersen,et al.  Asymptotic Properties of Conditional Maximum‐Likelihood Estimators , 1970 .

[62]  B. Wright,et al.  The essential process in a family of measurement models , 1984 .

[63]  P. Lachenbruch.  On Expected Probabilities of Misclassification in Discriminant Analysis, Necessary Sample Size, and a Relation with the Multiple Correlation Coefficient , 1968 .

[64]  C. D. Boor,et al.  On Calculating B-splines , 1972 .

[65]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[66]  A. H. Stroud,et al.  Numerical Quadrature and Solution of Ordinary Differential Equations: A Textbook for a Beginning Course in Numerical Analysis , 1974 .

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

[68]  J. Ramsay.  A similarity-based smoothing approach to nondimensional item analysis , 1995 .

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

[70]  Norman Verhelst,et al.  Maximum Likelihood Estimation in Generalized Rasch Models , 1986 .

[71]  James O. Ramsay,et al.  Nonparametric Item Response Function Estimates with the EM Algorithm , 2002 .

[72]  C. J. Stone,et al.  A study of logspline density estimation , 1991 .

[73]  Marie Davidian,et al.  A Monte Carlo EM algorithm for generalized linear mixed models with flexible random effects distribution. , 2002, Biostatistics.

[74]  R. Darrell Bock,et al.  Fitting a response model forn dichotomously scored items , 1970 .

[75]  G. Nürnberger.  Approximation by Spline Functions , 1989 .

[76]  P. Vijn,et al.  The Rasch Model as a Loglinear Model , 1981 .

[77]  Clement A. Stone,et al.  Recovery of Marginal Maximum Likelihood Estimates in the Two-Parameter Logistic Response Model: An Evaluation of MULTILOG , 1992 .