서브메뉴
검색
Factored Regression Specification for Composites With Categorical Item-Level Missing Data.
Factored Regression Specification for Composites With Categorical Item-Level Missing Data.
상세정보
- 자료유형
- 학위논문
- Control Number
- 0017162477
- International Standard Book Number
- 9798382838557
- Dewey Decimal Classification Number
- 151
- Main Entry-Personal Name
- Alacam, Egamaria.
- Publication, Distribution, etc. (Imprint
- [S.l.] : University of California, Los Angeles., 2024
- Publication, Distribution, etc. (Imprint
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- Physical Description
- 107 p.
- General Note
- Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
- General Note
- Advisor: Du, Han.
- Dissertation Note
- Thesis (Ph.D.)--University of California, Los Angeles, 2024.
- Summary, Etc.
- 요약Composites are widely used in the behavioral and social sciences where multiple items measure a construct of interest. Researchers often use composites to measure abstract concepts such as depression and anxiety. However, missing items are prevalent in the field, either due to participants skipping items or a planned missingness design. A researcher can choose either an item-level missing data treatment or a scale-level missing data treatment, but studies have demonstrated that item-level missing data treatment is superior because it maximizes power and precision. Item-level missing data handling though, can be challenging because missing data models can become very complex especially when there are many total items and small sample sizes. Recently in the literature, there have been many studies focused on advancing factored regression specifications and a recently published paper applied this to composite scores. The method was very favorable compared to other gold standard methods, but simulation studies had limited scope on categorical items. This dissertation extends the factored regression specification to examine how it performs under various scenarios categorical item distribution types and response format options. Overall, the simulation results suggest that the proposed method can be very effective compared to the gold standard methods under most conditions, especially when the number of items is very large, and the sample size is relatively small. A real data analysis illustrates the application of the proposed methods and the other gold standard methods.
- Subject Added Entry-Topical Term
- Quantitative psychology.
- Subject Added Entry-Topical Term
- Statistics.
- Subject Added Entry-Topical Term
- Psychology.
- Subject Added Entry-Topical Term
- Epidemiology.
- Index Term-Uncontrolled
- Composites
- Index Term-Uncontrolled
- Factored regression specification
- Index Term-Uncontrolled
- Missing data
- Index Term-Uncontrolled
- Gold standard methods
- Index Term-Uncontrolled
- Depression
- Added Entry-Corporate Name
- University of California, Los Angeles Psychology 0780
- Host Item Entry
- Dissertations Abstracts International. 85-12B.
- Electronic Location and Access
- 로그인을 한후 보실 수 있는 자료입니다.
- Control Number
- joongbu:654668
MARC
008250224s2024 us ||||||||||||||c||eng d■001000017162477
■00520250211152017
■006m o d
■007cr#unu||||||||
■020 ▼a9798382838557
■035 ▼a(MiAaPQ)AAI31331845
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a151
■1001 ▼aAlacam, Egamaria.
■24510▼aFactored Regression Specification for Composites With Categorical Item-Level Missing Data.
■260 ▼a[S.l.]▼bUniversity of California, Los Angeles. ▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a107 p.
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-12, Section: B.
■500 ▼aAdvisor: Du, Han.
■5021 ▼aThesis (Ph.D.)--University of California, Los Angeles, 2024.
■520 ▼aComposites are widely used in the behavioral and social sciences where multiple items measure a construct of interest. Researchers often use composites to measure abstract concepts such as depression and anxiety. However, missing items are prevalent in the field, either due to participants skipping items or a planned missingness design. A researcher can choose either an item-level missing data treatment or a scale-level missing data treatment, but studies have demonstrated that item-level missing data treatment is superior because it maximizes power and precision. Item-level missing data handling though, can be challenging because missing data models can become very complex especially when there are many total items and small sample sizes. Recently in the literature, there have been many studies focused on advancing factored regression specifications and a recently published paper applied this to composite scores. The method was very favorable compared to other gold standard methods, but simulation studies had limited scope on categorical items. This dissertation extends the factored regression specification to examine how it performs under various scenarios categorical item distribution types and response format options. Overall, the simulation results suggest that the proposed method can be very effective compared to the gold standard methods under most conditions, especially when the number of items is very large, and the sample size is relatively small. A real data analysis illustrates the application of the proposed methods and the other gold standard methods.
■590 ▼aSchool code: 0031.
■650 4▼aQuantitative psychology.
■650 4▼aStatistics.
■650 4▼aPsychology.
■650 4▼aEpidemiology.
■653 ▼aComposites
■653 ▼aFactored regression specification
■653 ▼aMissing data
■653 ▼aGold standard methods
■653 ▼aDepression
■690 ▼a0632
■690 ▼a0621
■690 ▼a0766
■690 ▼a0463
■71020▼aUniversity of California, Los Angeles▼bPsychology 0780.
■7730 ▼tDissertations Abstracts International▼g85-12B.
■790 ▼a0031
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162477▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.