This study addresses three issues on proactive behavior (PB) in occupational/organizational context, by using four samples gathered from Japan and China. The first issue is to develop the dimensional structure model of PB. A four-factor model was found out by exploratory factor analysis using Japanese student samples (n=384), and validated by multi-group confirmatory factor analysis using Japanese (n=1009) and Chinese (n=308) employee samples. The second issue is to develop a 12 item-Scale (PBS) based on item response theory. PBS was supported by two construct-related validity evidences across the two countries' employee samples: 1) the configural- & metric-invariance of the four-factor model measured by PBS; and 2) the reproducibility of both positive & negative correlation pattern among PBS score and some work-related variables. Finally, the third issue is to clarify the similarities/differences between Japanese and Chinese employees. The result revealed two similarities, which were referred in the second issue, and the following three differences: 1) the average hurdle for taking PB in the two countries' organizations; 2) the average level of PB; 3) what aspect (process or outcome) of work has strong relevance to PB. One explanation for the three differences is the distinction in occupational/organizational mobility due to performance-based system between Japan and China.
TUCKER3 (Tucker, 1966) is one of the methods of principal component analysis (PCA) for “quantitative” three-mode data. In this paper,we extend TUCKER3 to “nonmetric” three-mode PCA by deriving quantification scores. We also present an alternating least squares algorithm for finding optimal solution of nonmetric three-mode PCA by updating parameters such as quantification scores, loading matrices and core matrix successively. Calculating quantification scores of qualitative variables allows us to exclude arbitrariness which is included in questionnaire, such as the coding of nominal variables and the equality of intervals of ordinal variables. Two real data examples are given to illustrate the effectiveness of the proposed procedure.
Multidimensional item response theory model (MIRT) is a sub-model of Item Response Theory (IRT). It assumes that multidimensional latent traits influence test item responses. Equating or linking, which calibrate the parameters of two or more tests, are critical to the application of IRT, because they allow for comparisons between test scores. However, little research has been conducted for MIRT equating, especially on the common examinee design. The present study describes three equating methods for MIRT, which utilizes rotation methods in factor analysis focusing on factor scores, by employing the common examinee design, and analyzes the characteristics of each method through computer simulations. The results of the simulations reveal that the Weighted Procrustes method is effective when correlations between equated scores of each dimension are expected to approximate the target ones.