Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
Volume 2 , Issue 1
Showing 1-7 articles out of 7 articles from the selected issue
  • Masao IKUZAWA
    1974 Volume 2 Issue 1 Pages 1-10
    Published: March 30, 1974
    Released: June 28, 2010
    JOURNALS FREE ACCESS
    Green's solution of latent clas sanalysis is a well-devised method of computation. H owever, it generates sometimes unreasonable solutions, when data violates the conditions required for the valid solutions. Therefore, some improvements in the algorithm were tried in order to obtain valid solution, regardless of the response pattern of the original data. 1. Positive manifoldness in solution was attained by introducing the rotation of axes after Tarjan and Miller's gradient method(1971)with some modifications. 2. Postive. responses and negative It was found that the improved algorithm works fairly well. Numerical examples were also presented.
    Download PDF (1006K)
  • Kimiko KOZAWA
    1974 Volume 2 Issue 1 Pages 11-21
    Published: March 30, 1974
    Released: June 28, 2010
    JOURNALS FREE ACCESS
    This issue describes the development of the model which evaluates the citizen consicousness quantitatively by using the hierarchical worth program structure and worth function. This model isuseful to assist the decision maker in giving as much information as possible in generating administrative policy and planning. The basis of this model is Hiroshima city and this model contributes to optimize the societal benefit for citizens by means of improving these two objects; (1)to set up policies and goals. (2)to evaluate practical plans and set societal priorities.
    Download PDF (1482K)
  • Bien Tsujioka, Takayoshi Higashimura
    1974 Volume 2 Issue 1 Pages 22-32
    Published: March 30, 1974
    Released: June 28, 2010
    JOURNALS FREE ACCESS
    Download PDF (1314K)
  • Susumu YAMAKAGE, Nobuo Jo, Hiroaki YOSHII
    1974 Volume 2 Issue 1 Pages 33-40
    Published: March 30, 1974
    Released: January 25, 2011
    JOURNALS FREE ACCESS
    The technique is applicable to dyadic interaction data already measured, and is toclarify the interaction pattern of an actor and/or the structure of a system from the viewpoint of intensity and skewness of interaction. To treat a square matrix R, a simple techniques is developed as follows: New indices are defined, i. e. Rs=rijsdef=γij+γji(γijs=γsji)Ra=fijadef=γij-γji(γija=-γjia). Expressed in a different way, R=Ra+Rs. this, two unilateral relations γij and γij are transformed in to the dyadic interaction in terms of its intensity(γijs)and skewness(γija). Physical meanings of Rs and Ra are also discussed on the basis of physical interpretation of quantification theoryIV. As an example the Relative Acceptance(RA)index is chosen as measure of γij and this technique is applied successfully to transaction flows in international society.
    Download PDF (1063K)
  • Shigenobu AOKI, Shousuke SUZUKI, Haruo YANAI
    1974 Volume 2 Issue 1 Pages 41-53
    Published: March 30, 1974
    Released: June 28, 2010
    JOURNALS FREE ACCESS
    It was attempted to construct a new form of health and personality inventory, the Todai Health and Personality Inventory(THPI), which will be usefull for an aid of medical diagnosis. Twelve scales concerning health and personality were constructed on the basis of the result of factor analysis of191items, most of which were made referring to the described self complaints of 1044 male adults. On the process of scale construction, an attempt was made to eliminate the acquiescence tendency defined in terms of the total response score over items for each subiect. By making use of these scales, satisfactorily good result was obtained for computer asisted diagnosis between normals and psychosomatic disease patients based upon Fisher's discriminant analysis.
    Download PDF (1861K)
  • Shuichi IWATSUBO
    1974 Volume 2 Issue 1 Pages 54-65
    Published: March 30, 1974
    Released: January 25, 2011
    JOURNALS FREE ACCESS
    Two new classification methods are proposed to find out clusters from 3-way discrete data. They are quantification methods using the criteria maximizing correlation ratio and three-dimensional correlation coefficient, which lead to solving eigenequations. Threedimensional correlation coefficient is defined by E?{(X-E(X))(Y-E(Y))(Z-E(Z))}/√E(X-E(X)) 2E(Y-E(Y)) 2E(Z-E(Z)) 2, which is convenient to estimate the degree of joint relation among three variables of 3-way data. Some properties of three-dimensional correlation coefficient are given. Results of applying the methods to small data are also presented.
    Download PDF (1144K)
  • Eiichi KIMURA
    1974 Volume 2 Issue 1 Pages 66-71
    Published: March 30, 1974
    Released: June 28, 2010
    JOURNALS FREE ACCESS
    Although the first report on numerical diagnostics appeared more than 10 years agoin Japan, its use for practical medicine is only limited even now. According to the author's view, several reasons can be considered for this obstruction:(1)entrance of new diagnostic entities, such as sick sinus syndrome, (2)not suffciently reliable measurements, such as cardiothoracic ratio, (3)diffculties for quantification of two or three dimensional data, such as X-ray figures, blood cells, or biopsy specimen, (4)those of the data containing parameter of time, such as motor disturbances, and(5)those of subjective symptoms, such as chest pain. In order to realize the numerical diagnostics in clinical practice, it is necessary to develop the systems for numerical processing of these data, before making the programs for diagnosis of the diseases.
    Download PDF (993K)
feedback
Top