Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
Volume 38 , Issue 1
Showing 1-7 articles out of 7 articles from the selected issue
Articles
  • Koichi HIGUCHI
    2011 Volume 38 Issue 1 Pages 1-12
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    This paper investigates whether there is significant association between the content of newspaper articles and social consciousness trends in contemporary Japan, where it is already known that the influence of television and the Internet is pervasive. We examine whether it is possible to explore social consciousness by analyzing the content of newspapers. The data obtained are responses to open-ended questions in a nationwide survey —“JIS2004”— and articles in three newspapers:“Asahi,” “Yomiuri,” and “Mainichi.” The quantitative content analysis indicated a significant association between newspaper content and responses to the open-ended questions. However, this association could not be found in certain abstract themes. Thus, we argue that a content analysis of newspapers can be utilized as a research method for exploring social consciousness, although the focus should remain on concrete themes. The author also presents hypotheses about causal relationships between newspaper content and social consciousness.
    Download PDF (570K)
  • Atsuko KITADA
    2011 Volume 38 Issue 1 Pages 13-32
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    In Japan the “Basic Resident Register (BRR) ” is often used for statistical random sampling in public opinion surveys. The objective of the present paper is to compare two types of area sampling methods, without using the BRR. The first one (RRM) is a random route method, and the second one (DRM) is a method based on a database for a digital residential map. As a result, the percentage of data collected in the case of RRM was very low;and in the case of DRM, there was an uneven distribution in terms of respondents' age (a higher percentage of older respondents). However, when this bias was corrected by according weight to the population proportion, no strong significant difference concerning answer distributions was found. In each survey, the same substantial changes were detected statistically, compared with a past survey data based on the BRR. The findings suggest that these sampling methods are useful when accompanied with a careful analysis in spite of the resulting problems.
    Download PDF (894K)
  • Satoshi USAMI
    2011 Volume 38 Issue 1 Pages 33-50
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    This study investigated measurement problems of essay test data from various perspectives, controlling length of essays. Two sets of essay test data (A:about early introduction of English education, and B:about differences between the sexes in nurturing) were obtained from 303 high school students. Students were divided into 2 groups:one group (N=155) took essays A and B within 400 and 800 words respectively, and vice versa for another group (N=148). 4 raters evaluated 606 (303×2) essays both holistically and analytically (11 or 12 items).
    From factor analysis and covariance structure analysis of analytically-evaluated data, it was statistically confirmed that 2 factor (“linguistic ability factor” and “writing ability factor”) model was valid regardless of length of essays and raters. Reliability of evaluation between raters and within raters varied depending on items, and length of essays showed different effects for different items. From the view of internal consistency, the result based on multivariate generalizability theory indicated, regardless of length of essays and evaluation methods, that increasing the number of tests is more effective than adding raters for improving internal consistency. Propensity score analysis, with analytically-evaluated scores as covariates, showed that “beauty of handwriting” and “direction of opinion” might bias holistic scores.
    Download PDF (780K)
  • Naoki ISOGAWA, Toshiko IKEBE, Wataru SAKAMOTO, Masashi GOTO
    2011 Volume 38 Issue 1 Pages 51-63
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    Since April 2008, Ministry of Health, Labor and Welfare of Japan has carried out Health Checkups and Healthcare Advice with a particular focus on the Metabolic Syndrome which make it obiligatory for person aged 40 through 74 to reduce medical expenses and prevent lifestyle-related diseases. However, Kondo (2004) indicates a lack of foundation for health checkup. And we think that it is important to clarify foundation for effect of health checkups and healthcare advice. In this paper, through consistent statistical data analysis, we suggested a method which evaluate directed effect, attempting to figure the doctor's charactor, based on results of the health checkups carried out in April, 2004. As a result, we confirmed that the doctor classified subjects in the directed group from their body types such as weight and BMI, and in directed effect that it reduced weight and BMI, but it gave increase of TG and decrease of HDL which were likely to develop abnormal lipid metabolism. So, we found that adequate evaluation for effect of healthcare advice after careful consideration about the doctor's charactor leads to suggestion of scientific foundation for health checkups and healthcare advice.
    Download PDF (785K)
  • Minoru ISHIDA, Chizuru NISHIO, Hiroe TSUBAKI
    2011 Volume 38 Issue 1 Pages 65-81
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    Pairwise similarity coefficients are popular measure for binary variables. Many different measures of similarity have been proposed in the literature. Then we are interested in which one is the most effective for classifications. We focus on the fact that almost all measures of similarity are composed of interactions and main effects, and conjecture that the most useful similarity is an interaction because main effect don't play a role of classifications but totally order. All combinations of sixteen similarities coefficients and five clustering method were tested with music CD POS data. The cluster validation were assessed by interpretable, uniform, reproducible, external and internal criteria. As a result, the similarity coefficient which is more correlative with an interaction turns out more useful for classifications. That is, the best similarity is an interaction.
    Download PDF (807K)
Note
  • Takafumi SUZUKI, Kyo KAGEURA
    2011 Volume 38 Issue 1 Pages 83-92
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    This study explores the distributional characteristics of nouns of Japanese prime ministers' Diet addresses for interpreting two political phenomena. Previous content analysis studies focused only on a few content words for interpreting social phenomena and stylistic studies used stylistic characteristics independent on content of the texts with the aim of detecting their authors, genres, and chronological variations. This study focused, instead of these features, on distributional characteristics of nouns for better understanding two political phenomena: (a) the difference between the two types of Diet addresses delivered by Japanese prime ministers, and, (b) the perceived changes made to these addresses by two powerful prime ministers. This study expands the aims and scopes of text analysis by showing the effectiveness of exploring these microscopic textual characteristics for interpreting political phenomena.
    Download PDF (694K)
  • Koji E.KOSUGI, Hiroshi N.SHIMIZU, Takashi X.FUJISAWA, Masanori ISHIMOR ...
    2011 Volume 38 Issue 1 Pages 93-99
    Published: 2011
    Released: July 04, 2011
    JOURNALS FREE ACCESS
    The aim of this study was to construct a scale for family relationships. Essentially, as social relationships are bi-directional and asymmetrical, our data is considered to be three-mode (rater, rating targets, and items) data. A hierarchical factor analysis model was employed to analyze the three-mode data. Ten items that represented the relationships between family members were used. The participants were required to rate the behavior of the other members and vice versa. The model has two types of latent variables—the four factors representing the relationship among the members and a factor of all the items. The model could describe the relationship among the members as well as the characteristics of those relationships. Finally, in order to describe all the family relationships, the data will be constructed on the basis of three-mode data with a four-way combination. Therefore, the confirmation of the item parameters and factor structures in this study will determine the next step that this study should take.
    Download PDF (457K)
feedback
Top