The Role of Extended IRT Models for Composite Indicators Construction

Composite indicators are the conventional approach to socio-economic evaluation and are obtained by the combination into a single metrics of a number of individual indicators, each assessing a dimension of the latent phenomenon at issue. When dealing with the measurement of complex latent phenomenon, an objective is to seek information on the different dimensions contributing to characterise such multidimensional phenomenon. At the same time, it is also important to highlight potential differences in the behaviours of the units at issue (i.e., citizens, students, Countries, etc.) as regards such ascertained dimensions. For this twofold purpose, we describe the potentials of an extended Item Response Theory model, which allows us to: (i) retain the multidimensional structure of the data, and (ii) classify units in homogeneous groups, with very similar behaviours in terms of each dimension of the phenomenon at issue. The potentials of such model are illustrated considering the data from the Italian Multipurpose Survey, and specifically the part of it reserved to peoples’ habits as regards to their spare time and holidays.

[1]  Richard Scheines,et al.  When Are Multidimensional Data Unidimensional Enough for Structural Equation Modeling? An Evaluation of the DETECT Multidimensionality Index , 2015 .

[2]  Saisana Michaela,et al.  Higher Education Rankings: Robustness Issues and Critical Assessment , 2008 .

[3]  Mara Boni,et al.  Early approach to theoretical thinking: gears in primary school , 1999 .

[4]  B. Lindsay,et al.  Semiparametric Estimation in the Rasch Model and Related Exponential Response Models, Including a Simple Latent Class Model for Item Analysis , 1991 .

[5]  B. W. Ang,et al.  Comparing MCDA Aggregation Methods in Constructing Composite Indicators Using the Shannon-Spearman Measure , 2009 .

[6]  Tanja Srebotnjak,et al.  Measuring Uncertainties in Composite Indicators of Sustainability , 2009 .

[7]  Sarah Amsler.  University ranking: a dialogue on turning towards alternatives , 2014 .

[8]  Francesco Bartolucci,et al.  A Class of Multidimensional Latent Class IRT Models for Ordinal Polytomous Item Responses , 2012, 1201.4667.

[9]  Susan E. Embretson,et al.  A multidimensional latent trait model for measuring learning and change , 1991 .

[10]  Michela Gnaldi,et al.  Measuring University Performance by Means of Composite Indicators: A Robustness Analysis of the Composite Measure Used for the Benchmark of Italian Universities , 2016 .

[11]  Richard M. Luecht,et al.  Unidimensional Calibrations and Interpretations of Composite Traits for Multidimensional Tests , 1992 .

[12]  K. Holzinger,et al.  The Bi-factor method , 1937 .

[13]  Sarah S. Amsler,et al.  University ranking as social exclusion , 2012 .

[14]  Gregory Camilli,et al.  A Conceptual Analysis of Differential Item Functioning in Terms of a Multidimensional Item Response Model , 1992 .

[15]  A. Saltelli,et al.  Composite Indicators between Analysis and Advocacy , 2007 .

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

[17]  Roderick P. McDonald,et al.  Factor Analysis and Related Methods , 1985 .

[18]  Kristopher J Preacher,et al.  Item factor analysis: current approaches and future directions. , 2007, Psychological methods.

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

[20]  Francesco Bartolucci,et al.  A class of multidimensional IRT models for testing unidimensionality and clustering items , 2007 .

[21]  G. Rasch.  On General Laws and the Meaning of Measurement in Psychology , 1961 .

[22]  Anton K. Formann,et al.  Linear Logistic Latent Class Analysis and the Rasch Model , 1995 .

[23]  Paul F. Lazarsfeld,et al.  Latent Structure Analysis. , 1969 .

[24]  H. Hotelling.  Analysis of a complex of statistical variables into principal components. , 1933 .

[25]  Union européenne,et al.  Handbook on Constructing Composite Indicators: Methodology and User Guide , 2008 .

[26]  Francesco Bartolucci,et al.  Statistical Analysis of Questionnaires: A Unified Approach Based on R and Stata , 2015 .

[27]  M. Fattore,et al.  From Composite Indicators to Partial Orders: Evaluating Socio-Economic Phenomena Through Ordinal Data , 2012 .

[28]  Stefano Tarantola,et al.  Uncertainty and sensitivity analysis techniques as tools for the quality assessment of composite indicators , 2005 .

[29]  T. Lecerf,et al.  Orthogonal higher order structure and confirmatory factor analysis of the French Wechsler Adult Intelligence Scale (WAIS-III). , 2011, Psychological assessment.

[30]  H. Harman.  Modern factor analysis , 1961 .