Maximizing the information and validity of a linear composite in the factor analysis model for continuous item responses

Fuente: Redalyc
Saved in:
Bibliographic Details
Main Author: Pere J. Ferrando
Format: Artículo científico
Language:en
Published: Universitat de València 2008
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1876464471912742912
author Pere J. Ferrando
author_facet Pere J. Ferrando
contents Maximizing the information and validity of a linear composite in the factor analysis model for continuous item responses Pere J. Ferrando Psicología This paper develops results and procedures for obtaining linear compositesof factor scores that maximize: (a) test information, and (b) validity withrespect to external variables in the multiple factor analysis (FA) model. Itreat FA as a multidimensional item response theory model, and useAckerman’s multidimensional information approach based on maximumlikelihood (ML) estimation of trait levels. This approach, when applied tothe FA model, leads to particularly simple results as far as maximizing testinformation is concerned. Developments concerned with validity appear tobe new, and I use ML results in the context of error-in-variables regression.Graphical procedures for representing both type of results are proposed. Thedevelopments are illustrated with two empirical examples in personalitymeasurement 2008 artículo científico 0211-2159 https://www.redalyc.org/articulo.oa?id=16929205 en http://www.redalyc.org/revista.oa?id=169 Psicológica application/pdf Universitat de València Psicológica (España) Num.2 Vol.29
format Artículo científico
id redalyc_16929205
institution Redalyc
language en
publishDate 2008
publisher Universitat de València
spellingShingle Maximizing the information and validity of a linear composite in the factor analysis model for continuous item responses
Pere J. Ferrando
Psicología
Maximizing the information and validity of a linear composite in the factor analysis model for continuous item responses Pere J. Ferrando Psicología This paper develops results and procedures for obtaining linear compositesof factor scores that maximize: (a) test information, and (b) validity withrespect to external variables in the multiple factor analysis (FA) model. Itreat FA as a multidimensional item response theory model, and useAckerman’s multidimensional information approach based on maximumlikelihood (ML) estimation of trait levels. This approach, when applied tothe FA model, leads to particularly simple results as far as maximizing testinformation is concerned. Developments concerned with validity appear tobe new, and I use ML results in the context of error-in-variables regression.Graphical procedures for representing both type of results are proposed. Thedevelopments are illustrated with two empirical examples in personalitymeasurement 2008 artículo científico 0211-2159 https://www.redalyc.org/articulo.oa?id=16929205 en http://www.redalyc.org/revista.oa?id=169 Psicológica application/pdf Universitat de València Psicológica (España) Num.2 Vol.29
title Maximizing the information and validity of a linear composite in the factor analysis model for continuous item responses
topic Psicología
url https://www.redalyc.org/articulo.oa?id=16929205