行動計量学
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
12 巻 , 1 号
選択された号の論文の8件中1~8を表示しています
  • 長谷川 文雄, 嶋崎 伸一
    1984 年 12 巻 1 号 p. 1-10
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    This paper presents the results of analyzing of women's attitude on nuclear power with making a research for the sites of nuclear power station and urban areas only consuming the energy. The skeleton of the tests consists of 9 view points which are safety, confidence, stability, interest, feeling, regional economy, radioactivity, management system and propulsion. After analyzing, in sites of nuclear power, the evaluations about safety, confidence and propulsion are very severe and interest is very high level. Next step, in order to point out the key factors which decide the attitude onnuclear power, we introduce the Hayashi's Theory of Quantification II. Then we find out that safety and regional economy are key factors in site, on the other hand, in urban areas, feeling and radioactivity are key factors.
  • 石井 健一
    1984 年 12 巻 1 号 p. 11-19
    発行日: 1984/09/30
    公開日: 2011/01/25
    ジャーナル フリー
    In this paper representative differential equation models for diffusion process are compared. These models are(1)simple logistic curve, (2)harmonic logistic curve, (3)exponential logistic curve, (4)modified exponential curve, (5)Gompertz curve, (6)Bass's model. Logistic-type models((1), (2), (3))are based on the assumption of the effects of the interactions among members in a group.On the other hand, modified exponential curve model is based on the assumption of the effects of the outside information source. Bass's model synthesizes these two components of diffusion process. The six models are applied to data on consumer durables, music records, and information diffusions. The results show that Bass's model yields the better predictions for most of actual data than any other model.
  • 松原 義弘, 後藤 昌司
    1984 年 12 巻 1 号 p. 20-28
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    In this paper, we describe and discuss some useful graphical approaches in regession analysis from two view points; one is to diagnosis certain structures underlying in fitted regression model and another is to check or evaluate sampling behaviors of the model, including check of outliers and influence observations and so on. Further, we introduce some recent graphical techniques based on computer graphic display. Most of the graphical methods are illustrated on two well-known examples, Hald's cement data and Brownlee's stack loss data.
  • 田崎 武信, 後藤 昌司
    1984 年 12 巻 1 号 p. 29-34
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    This is an expository paper consisting not of new results but of miscellaneous and isolated comments on the probability plotting methods. We also describe two new applications of the probability plotting. One is an application to diagnosis of parallelism among many regression lines. The other is to evaluate normal range of some laboratory findings in clinical laboratory screening. The probability plotting methods, which have graphical diagnosis functions, are usefull to make the corresponding analytical methods vital.
  • 水田 正弘
    1984 年 12 巻 1 号 p. 35-42
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    A graphical representation for three-way data in a three-dimensional Euclidean space is presented in this paper. The method is an extension of Wakimoto and Taguri's Constellation Graphical Method; the projection of the three-dimensional graph on a plane is a constellation graph. Usage of the representation in a three-dimensional space is explained concretely. Its application is illustrated with some actual examples.
  • 永岡 慶三
    1984 年 12 巻 1 号 p. 43-52
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    There are many various data in educational field that should be processed metrically. The subjects using these data are classified into following three: 1)researcher for study, 2)teacher for instruction, 3)learner as the receiver of feedback data from teacher. The method of graphic representation of educational data is discussed in this paper on a viewpoint of improving the instruction of teacher above mentioned(2). Some actual examples are introduced: faces method, S-P table, constellation graph and the other similar methods that are used in the field at present. The characteristics and utilities of these methods and graphs are commented and a few opinions are showed at the conclusion.
  • 有田 清三郎
    1984 年 12 巻 1 号 p. 53-58
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    In this paper the analysis of the spatial pattern by the distance method was applied to the transected nerve fibers in the tissue photographs by an electron microscope. Though this analysis was originally proposed as the test for the randomness of the spatial distribution, here it was used as the method of the classification of the spatial distributions, such as the clustered, random or the regular distribution. The aim of this paper was to show that the process of maturation of regenerated nerve fibers could be described by the term of the spatial patterns. From the analytical results of experimental data by free grafts of common peroneal nerves of rats, it was suggested that the spatial pattern of regenerated nerve fibers was changed from the clustered to the regular through the random approaching to the normal pattern.
  • 渡部 洋, 大 雄作, 鈴木 規夫, 山田 文康
    1984 年 12 巻 1 号 p. 59-80
    発行日: 1984/09/30
    公開日: 2010/06/28
    ジャーナル フリー
    The number of published studies on Exploratory Data Analysis(EDA)has increased at a considerable rate, during the past two decades. This article provides explanatory and bibliographic informations of this area, with reviews of research trends within and across the following categories of research: description of distributions; re-expression; comparison of multiple data sets; resistant smoothing; wandering schematic plot; resistant line; resistant two-way analysis. Our discussions in the present article proceed mainly by comparing the techniques in EDA with the traditional methods used in general. The bibliographies are by no means exhaustive, but they should be enough in order to understand the general features of EDA.
feedback
Top