Comparison of factor simplicity indices for dichotomous data

DETECT R, Bentler's simplicity index, and the loading simplicity index

Holmes Finch, Alan Kirk Stage, Patrick Monahan

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

A primary assumption underlying several of the common methods for modeling item response data is unidimensionality, that is, test items tap into only one latent trait. This assumption can be assessed several ways, using nonlinear factor analysis and DETECT, a method based on the item conditional covariances. When multidimensionality is identified, a question of interest concerns the degree to which individual items are related to the latent traits. In cases where an item response is primarily associated with one of these traits it is said that (approximate) simple structure exists, whereas when the item response is related to both traits, the structure is complex. This study investigated the performance of three indices designed to assess the underlying structure present in item response data, two of which are based on factor analysis and one on DETECT. Results of the Monte Carlo simulations show that none of the indices works uniformly well in identifying the structure underlying item responses, although the DETECT r-ratio might be promising in differentiating between approximate simple and complex structures under certain circumstances.

Original languageEnglish
Pages (from-to)41-64
Number of pages24
JournalApplied Measurement in Education
Volume21
Issue number1
DOIs
StatePublished - Jan 2008

Fingerprint

Statistical Factor Analysis
factor analysis
simulation
present
performance

ASJC Scopus subject areas

  • Education
  • Developmental and Educational Psychology
  • Psychology (miscellaneous)

Cite this

Comparison of factor simplicity indices for dichotomous data : DETECT R, Bentler's simplicity index, and the loading simplicity index. / Finch, Holmes; Stage, Alan Kirk; Monahan, Patrick.

In: Applied Measurement in Education, Vol. 21, No. 1, 01.2008, p. 41-64.

Research output: Contribution to journalArticle

@article{0346f7a9cf084383bb9061cc91108539,
title = "Comparison of factor simplicity indices for dichotomous data: DETECT R, Bentler's simplicity index, and the loading simplicity index",
abstract = "A primary assumption underlying several of the common methods for modeling item response data is unidimensionality, that is, test items tap into only one latent trait. This assumption can be assessed several ways, using nonlinear factor analysis and DETECT, a method based on the item conditional covariances. When multidimensionality is identified, a question of interest concerns the degree to which individual items are related to the latent traits. In cases where an item response is primarily associated with one of these traits it is said that (approximate) simple structure exists, whereas when the item response is related to both traits, the structure is complex. This study investigated the performance of three indices designed to assess the underlying structure present in item response data, two of which are based on factor analysis and one on DETECT. Results of the Monte Carlo simulations show that none of the indices works uniformly well in identifying the structure underlying item responses, although the DETECT r-ratio might be promising in differentiating between approximate simple and complex structures under certain circumstances.",
author = "Holmes Finch and Stage, {Alan Kirk} and Patrick Monahan",
year = "2008",
month = "1",
doi = "10.1080/08957340701796365",
language = "English",
volume = "21",
pages = "41--64",
journal = "Applied Measurement in Education",
issn = "0895-7347",
publisher = "Routledge",
number = "1",

}

TY - JOUR

T1 - Comparison of factor simplicity indices for dichotomous data

T2 - DETECT R, Bentler's simplicity index, and the loading simplicity index

AU - Finch, Holmes

AU - Stage, Alan Kirk

AU - Monahan, Patrick

PY - 2008/1

Y1 - 2008/1

N2 - A primary assumption underlying several of the common methods for modeling item response data is unidimensionality, that is, test items tap into only one latent trait. This assumption can be assessed several ways, using nonlinear factor analysis and DETECT, a method based on the item conditional covariances. When multidimensionality is identified, a question of interest concerns the degree to which individual items are related to the latent traits. In cases where an item response is primarily associated with one of these traits it is said that (approximate) simple structure exists, whereas when the item response is related to both traits, the structure is complex. This study investigated the performance of three indices designed to assess the underlying structure present in item response data, two of which are based on factor analysis and one on DETECT. Results of the Monte Carlo simulations show that none of the indices works uniformly well in identifying the structure underlying item responses, although the DETECT r-ratio might be promising in differentiating between approximate simple and complex structures under certain circumstances.

AB - A primary assumption underlying several of the common methods for modeling item response data is unidimensionality, that is, test items tap into only one latent trait. This assumption can be assessed several ways, using nonlinear factor analysis and DETECT, a method based on the item conditional covariances. When multidimensionality is identified, a question of interest concerns the degree to which individual items are related to the latent traits. In cases where an item response is primarily associated with one of these traits it is said that (approximate) simple structure exists, whereas when the item response is related to both traits, the structure is complex. This study investigated the performance of three indices designed to assess the underlying structure present in item response data, two of which are based on factor analysis and one on DETECT. Results of the Monte Carlo simulations show that none of the indices works uniformly well in identifying the structure underlying item responses, although the DETECT r-ratio might be promising in differentiating between approximate simple and complex structures under certain circumstances.

UR - http://www.scopus.com/inward/record.url?scp=42649109664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42649109664&partnerID=8YFLogxK

U2 - 10.1080/08957340701796365

DO - 10.1080/08957340701796365

M3 - Article

VL - 21

SP - 41

EP - 64

JO - Applied Measurement in Education

JF - Applied Measurement in Education

SN - 0895-7347

IS - 1

ER -