Properties of second-order exponential models as multidimensional response models

Carolyn Jane Anderson, Hsiu Ting Yu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Second-order exponential (SOE) models have been proposed as item response models (e.g., Anderson et al., J. Educ. Behav. Stat. 35:422–452, 2010; Anderson, J. Classif. 30:276–303, 2013. doi: 10.1007/s00357-00357-013-9131-x; Hessen, Psychometrika 77:693–709, 2012. doi:10.1007/s11336-012-9277-1 Holland, Psychometrika 55:5–18, 1990); however, the philosophical and theoretical underpinnings of the SOE models differ from those of standard item response theory models. Although presented as reexpressions of item response theory models (Holland, Psychometrika 55:5–18, 1990), which are reflective models, the SOE models are formative measurement models. We extend Anderson and Yu (Psychometrika 72:5–23, 2007) who studied unidimensional models for dichotomous items to multidimensional models for dichotomous and polytomous items. The properties of the models for multiple latent variables are studied theoretically and empirically. Even though there are mathematical differences between the second-order exponential models and multidimensional item response theory (MIRT) models, the SOE models behave very much like standard MIRT models and in some cases better than MIRT models.

Original languageEnglish (US)
Title of host publicationQuantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016
EditorsWen-Chung Wang, Marie Wiberg, Steven A. Culpepper, Jeffrey A. Douglas, L. Andries van der Ark
PublisherSpringer New York LLC
Pages9-19
Number of pages11
ISBN (Print)9783319562933
DOIs
StatePublished - Jan 1 2017
Event81st annual meeting of the Psychometric Society, 2016 - Asheville, United States
Duration: Jul 11 2016Jul 15 2016

Publication series

NameSpringer Proceedings in Mathematics and Statistics
Volume196
ISSN (Print)2194-1009
ISSN (Electronic)2194-1017

Other

Other81st annual meeting of the Psychometric Society, 2016
CountryUnited States
CityAsheville
Period7/11/167/15/16

Fingerprint

Second-order Model
Exponential Model
Model Theory
Model
Multidimensional Model
Latent Variables

Keywords

  • Bi-variate exponential
  • Composite indicators
  • Dutch Identity
  • Formative models
  • Log-multiplicative association models
  • Reflective models
  • Skew normal

ASJC Scopus subject areas

  • Mathematics(all)

Cite this

Anderson, C. J., & Yu, H. T. (2017). Properties of second-order exponential models as multidimensional response models. In W-C. Wang, M. Wiberg, S. A. Culpepper, J. A. Douglas, & L. A. van der Ark (Eds.), Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016 (pp. 9-19). (Springer Proceedings in Mathematics and Statistics; Vol. 196). Springer New York LLC. https://doi.org/10.1007/978-3-319-56294-0_2

Properties of second-order exponential models as multidimensional response models. / Anderson, Carolyn Jane; Yu, Hsiu Ting.

Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016. ed. / Wen-Chung Wang; Marie Wiberg; Steven A. Culpepper; Jeffrey A. Douglas; L. Andries van der Ark. Springer New York LLC, 2017. p. 9-19 (Springer Proceedings in Mathematics and Statistics; Vol. 196).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Anderson, CJ & Yu, HT 2017, Properties of second-order exponential models as multidimensional response models. in W-C Wang, M Wiberg, SA Culpepper, JA Douglas & LA van der Ark (eds), Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016. Springer Proceedings in Mathematics and Statistics, vol. 196, Springer New York LLC, pp. 9-19, 81st annual meeting of the Psychometric Society, 2016, Asheville, United States, 7/11/16. https://doi.org/10.1007/978-3-319-56294-0_2
Anderson CJ, Yu HT. Properties of second-order exponential models as multidimensional response models. In Wang W-C, Wiberg M, Culpepper SA, Douglas JA, van der Ark LA, editors, Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016. Springer New York LLC. 2017. p. 9-19. (Springer Proceedings in Mathematics and Statistics). https://doi.org/10.1007/978-3-319-56294-0_2
Anderson, Carolyn Jane ; Yu, Hsiu Ting. / Properties of second-order exponential models as multidimensional response models. Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016. editor / Wen-Chung Wang ; Marie Wiberg ; Steven A. Culpepper ; Jeffrey A. Douglas ; L. Andries van der Ark. Springer New York LLC, 2017. pp. 9-19 (Springer Proceedings in Mathematics and Statistics).
@inproceedings{47863b20cb274edabc290d8cc9c3420d,
title = "Properties of second-order exponential models as multidimensional response models",
abstract = "Second-order exponential (SOE) models have been proposed as item response models (e.g., Anderson et al., J. Educ. Behav. Stat. 35:422–452, 2010; Anderson, J. Classif. 30:276–303, 2013. doi: 10.1007/s00357-00357-013-9131-x; Hessen, Psychometrika 77:693–709, 2012. doi:10.1007/s11336-012-9277-1 Holland, Psychometrika 55:5–18, 1990); however, the philosophical and theoretical underpinnings of the SOE models differ from those of standard item response theory models. Although presented as reexpressions of item response theory models (Holland, Psychometrika 55:5–18, 1990), which are reflective models, the SOE models are formative measurement models. We extend Anderson and Yu (Psychometrika 72:5–23, 2007) who studied unidimensional models for dichotomous items to multidimensional models for dichotomous and polytomous items. The properties of the models for multiple latent variables are studied theoretically and empirically. Even though there are mathematical differences between the second-order exponential models and multidimensional item response theory (MIRT) models, the SOE models behave very much like standard MIRT models and in some cases better than MIRT models.",
keywords = "Bi-variate exponential, Composite indicators, Dutch Identity, Formative models, Log-multiplicative association models, Reflective models, Skew normal",
author = "Anderson, {Carolyn Jane} and Yu, {Hsiu Ting}",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-56294-0_2",
language = "English (US)",
isbn = "9783319562933",
series = "Springer Proceedings in Mathematics and Statistics",
publisher = "Springer New York LLC",
pages = "9--19",
editor = "Wen-Chung Wang and Marie Wiberg and Culpepper, {Steven A.} and Douglas, {Jeffrey A.} and {van der Ark}, {L. Andries}",
booktitle = "Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016",

}

TY - GEN

T1 - Properties of second-order exponential models as multidimensional response models

AU - Anderson, Carolyn Jane

AU - Yu, Hsiu Ting

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Second-order exponential (SOE) models have been proposed as item response models (e.g., Anderson et al., J. Educ. Behav. Stat. 35:422–452, 2010; Anderson, J. Classif. 30:276–303, 2013. doi: 10.1007/s00357-00357-013-9131-x; Hessen, Psychometrika 77:693–709, 2012. doi:10.1007/s11336-012-9277-1 Holland, Psychometrika 55:5–18, 1990); however, the philosophical and theoretical underpinnings of the SOE models differ from those of standard item response theory models. Although presented as reexpressions of item response theory models (Holland, Psychometrika 55:5–18, 1990), which are reflective models, the SOE models are formative measurement models. We extend Anderson and Yu (Psychometrika 72:5–23, 2007) who studied unidimensional models for dichotomous items to multidimensional models for dichotomous and polytomous items. The properties of the models for multiple latent variables are studied theoretically and empirically. Even though there are mathematical differences between the second-order exponential models and multidimensional item response theory (MIRT) models, the SOE models behave very much like standard MIRT models and in some cases better than MIRT models.

AB - Second-order exponential (SOE) models have been proposed as item response models (e.g., Anderson et al., J. Educ. Behav. Stat. 35:422–452, 2010; Anderson, J. Classif. 30:276–303, 2013. doi: 10.1007/s00357-00357-013-9131-x; Hessen, Psychometrika 77:693–709, 2012. doi:10.1007/s11336-012-9277-1 Holland, Psychometrika 55:5–18, 1990); however, the philosophical and theoretical underpinnings of the SOE models differ from those of standard item response theory models. Although presented as reexpressions of item response theory models (Holland, Psychometrika 55:5–18, 1990), which are reflective models, the SOE models are formative measurement models. We extend Anderson and Yu (Psychometrika 72:5–23, 2007) who studied unidimensional models for dichotomous items to multidimensional models for dichotomous and polytomous items. The properties of the models for multiple latent variables are studied theoretically and empirically. Even though there are mathematical differences between the second-order exponential models and multidimensional item response theory (MIRT) models, the SOE models behave very much like standard MIRT models and in some cases better than MIRT models.

KW - Bi-variate exponential

KW - Composite indicators

KW - Dutch Identity

KW - Formative models

KW - Log-multiplicative association models

KW - Reflective models

KW - Skew normal

UR - http://www.scopus.com/inward/record.url?scp=85020863216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020863216&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-56294-0_2

DO - 10.1007/978-3-319-56294-0_2

M3 - Conference contribution

AN - SCOPUS:85020863216

SN - 9783319562933

T3 - Springer Proceedings in Mathematics and Statistics

SP - 9

EP - 19

BT - Quantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016

A2 - Wang, Wen-Chung

A2 - Wiberg, Marie

A2 - Culpepper, Steven A.

A2 - Douglas, Jeffrey A.

A2 - van der Ark, L. Andries

PB - Springer New York LLC

ER -