Structural Equation Modeling for One-Way MANOVA Design Using the Amos Graphics Software

• 발행기관: 한국자료분석학회
• 저자명: 이중재(Jung-Jae Lee) 오현숙(Hyun Sook Oh)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.17 No.2, 597~611쪽, 전체 15쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2015.04.30

영문 초록

Bagozzi, Yi (1989) developed powerful procedures for the analysis of MANOVA and MANCOVA designs on the use of structural equation modeling with the Lisrel program. The present paper showed how the Bagozzi, Yi s (1989) procedure for one-way MANOVA and MANCOVA analysis could be accomplished with the Amos program. For the comparison with the Lisrel and easy understanding for the method of Bagozzi, Yi (1989), we used the same illustrative examples as Bagozzi, Yi (1989). The noticeable difference in MANOVA or MANCOVA analysis is that a pseudo variable is present to estimate means in the Lisrel while means are estimated automatically by the option Estimate means and intercepts by which the model with the Amos is simpler than the one with the Lisrel. Model equation formulas, assumptions for parameters and appropriate data forms for a given model specification in the Amos were explained by comparing with the Lisrel. This paper showed all the procedures for one-way MANOVAs suggested by Bagozzi, Yi (1989) could be performed easily by using the Amos while another software PLS had some limitation for application of the procedures (Bagozzi, Yi, 1991).

목차

1. Introduction

2. MANOVA

3. MANCOVA

4. Example

5. Summary and Discussion

References

키워드

latent variable MANOVA (MANCOVA) dummy variable approach multiple-group approach step-down analysis