서브메뉴
검색
Investigating Model and Metric Misspecification in Latent Variable Models and Evaluating Practical Implications
Investigating Model and Metric Misspecification in Latent Variable Models and Evaluating Practical Implications
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211153129
- ISBN
- 9798346874065
- DDC
- 371
- 저자명
- Huang, Qi.
- 서명/저자
- Investigating Model and Metric Misspecification in Latent Variable Models and Evaluating Practical Implications
- 발행사항
- [Sl] : The University of Wisconsin - Madison, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 147 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-06, Section: B.
- 주기사항
- Advisor: Bolt, Daniel M.
- 학위논문주기
- Thesis (Ph.D.)--The University of Wisconsin - Madison, 2024.
- 초록/해제
- 요약Latent variable models are often appealing in educational and psychological assessments, where focal constructs are not only generally viewed as being measured with error, but also lack the well-defined metrics associated with most physical measurements. Although the flexibility of metrics in latent variable models makes it easier for models to statistically fit the data at hand, it can also mask model misspecification. In this dissertation, I will consider three different scenarios where either the metric or measurement model (or both) is misspecified, discuss the consequences of overlooking each type of misspecification, and propose corresponding solutions. Specifically, the three studies examine: (1) misspecification of item response theory (IRT) models induced by item complexity; (2) misspecification of IRT in both metric and measurement model under unipolarity of the latent construct; (3) misspecification of cognitive diagnosis models (CDMs) in the presence of latent skill continuity.Overall, these studies draw attention to potentially underappreciated issues in measurement, expand the perspectives that people can take on measurement, and promote the use of a more contemporary toolbox for consideration of varied metrics and models, possibly allowing educational and psychological theory to more significantly inform the modeling process.
- 일반주제명
- Quantitative psychology
- 일반주제명
- Experimental psychology
- 기타저자
- The University of Wisconsin - Madison Educational Psychology
- 기본자료저록
- Dissertations Abstracts International. 86-06B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017165146
■00520250211153129
■006m o d
■007cr#unu||||||||
■020 ▼a9798346874065
■035 ▼a(MiAaPQ)AAI31767399
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a371
■1001 ▼aHuang, Qi.
■24510▼aInvestigating Model and Metric Misspecification in Latent Variable Models and Evaluating Practical Implications
■260 ▼a[Sl]▼bThe University of Wisconsin - Madison▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a147 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-06, Section: B.
■500 ▼aAdvisor: Bolt, Daniel M.
■5021 ▼aThesis (Ph.D.)--The University of Wisconsin - Madison, 2024.
■520 ▼aLatent variable models are often appealing in educational and psychological assessments, where focal constructs are not only generally viewed as being measured with error, but also lack the well-defined metrics associated with most physical measurements. Although the flexibility of metrics in latent variable models makes it easier for models to statistically fit the data at hand, it can also mask model misspecification. In this dissertation, I will consider three different scenarios where either the metric or measurement model (or both) is misspecified, discuss the consequences of overlooking each type of misspecification, and propose corresponding solutions. Specifically, the three studies examine: (1) misspecification of item response theory (IRT) models induced by item complexity; (2) misspecification of IRT in both metric and measurement model under unipolarity of the latent construct; (3) misspecification of cognitive diagnosis models (CDMs) in the presence of latent skill continuity.Overall, these studies draw attention to potentially underappreciated issues in measurement, expand the perspectives that people can take on measurement, and promote the use of a more contemporary toolbox for consideration of varied metrics and models, possibly allowing educational and psychological theory to more significantly inform the modeling process.
■590 ▼aSchool code: 0262.
■650 4▼aEducational tests & measurements
■650 4▼aQuantitative psychology
■650 4▼aExperimental psychology
■653 ▼aCognitive diagnosis model
■653 ▼aItem response theory
■653 ▼aLatent variable model
■653 ▼aModel misspecification
■690 ▼a0288
■690 ▼a0632
■690 ▼a0623
■71020▼aThe University of Wisconsin - Madison▼bEducational Psychology.
■7730 ▼tDissertations Abstracts International▼g86-06B.
■790 ▼a0262
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17165146▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
