Search

The present study tested the factorial validity of the 9-item Bergen Burnout Inventory (BBI-9) 1) . The BBI-9 is comprised of three core dimensions: (1) exhaustion at work; (2) cynicism toward the meaning of work; and (3) sense of inadequacy at work. The study further investigated whether the three-factor structure of the BBI-9 remains the same across different organizations (group invariance) and measurement time points (time invariance). The factorial group invariance was tested using a cross-sectional design with data pertaining to managers (n=742), and employees working in a bank (n=162), an engineering office (n=236), a public sector organization divided into three service areas: administration (n=102), education and culture (n=581), and social affairs and health (n=1,505). Factorial time invariance was tested using longitudinal data pertaining to manag- ers, with three measurements over a four-year follow-up period. The confirmatory factor analysis revealed that the three-factor structure of the BBI-9 was invariant across cross-sectional samples. The factorial invariance was also supported across measurement times. To conclude, the factorial structure of the BBI-9 was found to remain the same regardless of the sample properties and mea- surement times.

The purpose of this study was to confirm the causal structure model of muscle, motor and living functions utilizing structural equation modeling (SEM). As subjects, 103 community-dwelling older men and women, aged 65.7±6.9 years of age, participated in the study to measure muscle cross-sectional area, maximum voluntary contractions, muscle power, 4 physical performance tests, and 16 questionnaires regarding ability of activities of daily living. The causal structure model of muscle, motor and living functions was hypothesized to be a hierarchical causal structure. The causal structure model of muscle function was hypothesized to be a hierarchical causal structure consisting of 3 sub-domains of muscle mass, muscle strength, and muscle power. Data analysis procedures were as follows : a) testing of construct validity of muscle function variables using confirmatory factor analysis (CFA) in SEM ; b) testing of causal structure using SEM ; c) testing of factor invariance using multi-group analysis for gender. The highest goodness of fit indices was obtained in the causal structure model of muscle, motor and living functions (NFI=.928, CFI=.978, RMSEA =.061). The causal coefficient of muscle function to motor function was. 98 (p

The purpose of this study was to confirm the causal effect model of strength on walking ability development as a result of exercise participation among elderly people in a community, utilizing the latent curve model (LCM) in structural equation modeling (SEM). Twenty-six male and 57 female subjects, 83 in total, aged 67.8±5.7, 63.9±7.1 and 65.1±6.9 in a pooled sample participated in the exercise program which lasted for two years. Grip strength and sit-ups used in the Japan Fitness Test were measured for muscular strength, 10-m hurdle walk and 6-min walk for walking ability, and the fitness test score for physical ability. The data analysis procedures were as follows : a) analysis of test-retest reliability and construct validity of measurement items, b) analysis of causal structure model of aging, muscular strength and walking ability, c) analysis of variance for repeated measurement of walking performance by sex, age and year, d) analysis of LCM for walking performance development. The highest goodness-of-fit indices of SEM were obtained in the LCM of 10-m hurdle walk performance development (GFI=0.989, AGFI=0.920, CFI=0.998, RMSEA=0.038). The path coefficient of sit-ups at pre-test effect on the intercept of 10-m hurdle walk performance development was significant (p

The purpose of this paper is to review the structural equation model using the terms of factor analysis, to discuss the distinction between the exploratory factor analysis and the hypothesis testing factor analysis, to review the LISREL's basic formulations, and to demonstrate the potentiality of the RAM algebraic treatment for the structural equation modeling. The factor analysis using the maximum likelihood for the parameter estimation is useful for the hypothesis testing situation, and is the part of the structural equation models. The problem of factor inavriance is almost solved by this hypothgsis testing methods, especially using multi-sample simultaneous analysis. Therefore, traditional factor analysis and related methods using the least square method for the parameter estimation are recommended for the purpose to accumulate the knowledge in the specific research fields and to constract the hypothesis. After representing the RAM logic, the simple.algebraic treatment provided this logic is applied to find out the parameter matrices of the LISREL model. The parameter matrices of the common factor model and the primary factor model with second order factors are also treated with the RAM logic., 本稿では，構造方程式モデルを因子分析の銀点からレビューし，探索的因子分析と仮説検証的因子分析との差異を検討した。さらにLISRELの基本式を整理し，そして構造方程式モデルに対するRAMの可能性を代数的に展開した。パラメータの最尤推定を採用している因子分析は，仮説検定をおこなう状況においては有用であり，この因子分析は構造方程式モデルの一つである。そして因子不変性の問題は，この仮説検証の方法によって，特に多群分析法を使用した場合，ほとんど解決された。最小二乗法による伝統的因子分析や関連手法はそれぞれの研究分野における知識の蓄積と仮説の構成には推奨される。RAMの論理を紹介した後で，LISRELモデルのパラメータ行列を見出すために，この論理によって与えられる単純な代数展開が適用された。また，共通因子分析モデルや一次因子と二次因子とから構成されるモデルのパラメータ行列もRAMで展開された。, 尾崎康夫教授古稀記念号

多群間の因子的不変性を確認するための総合システムが，因子分析における確認的観点からの数学的公式化とそれに伴なうコンピュータプログラムの両方から開発された。多群から得られた資料の場合について，Kristof & WingerskyとTen Bergeによって公式化された因子的不変性の目的関数を最大化するため，必要条件と十分条件が論究された。Ten Bergeによって提案されたStrong Necessary Conditionに基づく手続きは，因子的不変性に到達するためのもっともましな方法であると考えられる。しかし，Kristof & Wingerskyによって提案された Weak Necessary Conditionに基づく手続きは，経験諸科学における実際の使用に際して効果的な方法であると考えられる。帰納と演繹の2種の基準が，因子間相関の固定を伴なう単純構造のために提案された。フォートラン・プログラムが，この目的のために書かれ，そして5群の高校生男子から得られたYG性格検査の因子負荷行列に対して数値的検証がなされた。, A synthetic system to confirm factorial invariance among multiple groups was developed by means of both mathematical formulations and also of computer programs from the confirmatory point of view in factor analysis. The necessary and sufficient conditions were discussed for maximizing the objective function for factorial invariance formulated by both Kristof & Wingersky and Ten Berge in case of the data obtained from multiple groups. The procedure under the strong necessary condition suggested by Ten Berge is considered as the better method for obtaining the factorial invariance. But the procedure under the weak necessary condition suggested by Kristof & Wingersky is considered as the effective method for a practical use in empirical sciences. Two kinds of criteria, inductive and deductive were proposed for simple structure with constraint of factor correlations. Fortran programs were written for this purpose and checked numerically for the factor loading matrices of the YG personality inventory obtained from 5 groups of highschool boys. (author abstract)