Estimation of Item Parameters

KB Christensen

Estimation of Item Parameters

Section of Biostatistics

Abstract

This chapter describes item parameter estimation in dichotomous and polytomous Rasch models. It discusses three different ways of estimating item parameters: (1) conditional maximum likelihood (CML) estimation, (2) pairwise CML estimation and (3) marginal maximum likelihood (MML) estimation. Much research interest has focused on the possibility of doing marginal inference without having to make distributional assumptions about the latent variable and the chapter touches upon the extended likelihood function. Reduced rank parameterization is discussed next. Finally, the chapter illustrates the methods using data from a study validating a French translation of the Diabetes Health Profile (DHP).

Original language	Undefined/Unknown
Title of host publication	In:Rasch Models in Health
Editors	Mesbah M Christensen KB Kreiner S
Number of pages	13
Publisher	Wiley
Publication date	4 Mar 2013
Pages	49-61
Publication status	Published - 4 Mar 2013

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Cite this

@inbook{c42304a0dfcd42e5be29a04a1573e6e7,

title = "Estimation of Item Parameters",

abstract = "This chapter describes item parameter estimation in dichotomous and polytomous Rasch models. It discusses three different ways of estimating item parameters: (1) conditional maximum likelihood (CML) estimation, (2) pairwise CML estimation and (3) marginal maximum likelihood (MML) estimation. Much research interest has focused on the possibility of doing marginal inference without having to make distributional assumptions about the latent variable and the chapter touches upon the extended likelihood function. Reduced rank parameterization is discussed next. Finally, the chapter illustrates the methods using data from a study validating a French translation of the Diabetes Health Profile (DHP).",

author = "KB Christensen",

year = "2013",

month = mar,

day = "4",

language = "Udefineret/Ukendt",

pages = "49--61",

editor = "{Christensen KB Kreiner S}, {Mesbah M}",

booktitle = "In:Rasch Models in Health",

publisher = "Wiley",

address = "USA",

}

TY - CHAP

T1 - Estimation of Item Parameters

AU - Christensen, KB

PY - 2013/3/4

Y1 - 2013/3/4

N2 - This chapter describes item parameter estimation in dichotomous and polytomous Rasch models. It discusses three different ways of estimating item parameters: (1) conditional maximum likelihood (CML) estimation, (2) pairwise CML estimation and (3) marginal maximum likelihood (MML) estimation. Much research interest has focused on the possibility of doing marginal inference without having to make distributional assumptions about the latent variable and the chapter touches upon the extended likelihood function. Reduced rank parameterization is discussed next. Finally, the chapter illustrates the methods using data from a study validating a French translation of the Diabetes Health Profile (DHP).

AB - This chapter describes item parameter estimation in dichotomous and polytomous Rasch models. It discusses three different ways of estimating item parameters: (1) conditional maximum likelihood (CML) estimation, (2) pairwise CML estimation and (3) marginal maximum likelihood (MML) estimation. Much research interest has focused on the possibility of doing marginal inference without having to make distributional assumptions about the latent variable and the chapter touches upon the extended likelihood function. Reduced rank parameterization is discussed next. Finally, the chapter illustrates the methods using data from a study validating a French translation of the Diabetes Health Profile (DHP).

M3 - Bidrag til bog/antologi

SP - 49

EP - 61

BT - In:Rasch Models in Health

A2 - Christensen KB Kreiner S, Mesbah M

PB - Wiley

ER -