ABSTRACT The comparison between criterions of oblique rotation in factor analysis. Tae-kyung, Lim Department of Psychology Graduate School of Sungkyunkwan University The purpose of this study was to investigate the effects of criterions of obli...

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
https://www.riss.kr/link?id=T10841785
서울 : 성균관대학교, 2007
학위논문(석사) -- 성균관대학교 일반대학원 , 심리학과 , 2007. 2
2007
한국어
150 판사항(22)
서울
117 p. : 삽도 ; 26 cm.
지도교수: 이순묵.
참고문헌: p.82-85
0
상세조회0
다운로드다국어 초록 (Multilingual Abstract)
ABSTRACT The comparison between criterions of oblique rotation in factor analysis. Tae-kyung, Lim Department of Psychology Graduate School of Sungkyunkwan University The purpose of this study was to investigate the effects of criterions of obli...
ABSTRACT
The comparison between criterions of oblique rotation in factor analysis.
Tae-kyung, Lim
Department of Psychology
Graduate School of
Sungkyunkwan University
The purpose of this study was to investigate the effects of criterions of oblique rotation on recovery of population factors under conditions in which the common factor model holds in the population. Which are relatively little studied among a variety of Monte Carlo studies on recovery of population factors.
The general approach used in this study involved the following steps: ⑴ 12 population correlation matrices were generated, which matrices were defined as having specific desired properties and known factor structures. The matrices were defined by ⒜ two levels of inter factor correlation(0.30, 0.50), ⒝ two levels of factor loading(0.40, 0.60), ⒞ three levels of number of factors(3, 4, 5), and the number of measured variables was held ‘number of factors×10’. ⑵ 30 sample correlation matrices were generated from each population; ⑶ the sample correlation matrices were factor analysed, using maximum likelihood factor analysis and various criterions of rotation(Geomin, Infomax, Orthoblque, Direct oblimin), specifying the number of factors retained as equal to the known number of factor in the population; the sample factor solutions were evaluated to determine how various aspects of those solutions were affected by criterion and other properties of the data. For each of the rotated sample solutions, measures of congruence between sample and population factors the variations of variable complexity(decrease and increase) were obtained. This Indexes were treated as dependent variables in four-way ANOVAs using three between factor(inter factor correlations, factor loading, number of factors), one within factor(criterions) as independent variables.
As results of the analysis of the dependent variable, factor loadings and inter factor correlations become higher, the role of criterion become more important.
Finally, implications and limitations of the study along with suggestions for future research are discussed.
목차 (Table of Contents)