Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
Volume 36, Issue 1
Displaying 1-5 of 5 articles from this issue
Articles
  • Takuya OKAMOTO, Yoshifumi HAYASHI, Takehiro FUJIHARA
    2009 Volume 36 Issue 1 Pages 1-14
    Published: 2009
    Released on J-STAGE: August 13, 2009
    JOURNAL FREE ACCESS
    This study aimed to measure social identities to belonging university using the Photo Projective Method (PPM) and the social identity scale. PPM is a new technique based on a projective method that uses photographs to capture perceived environments. First, we developed the social identity scale by referring to previous researches on social identity scales and a sense of community scales. Second, 96 university students were provided with a camera and requested to photograph their daily lives in their universities for a week. The Correspondence Analysis showed that the pictures represented some new aspects of the social identities that could not be measured by ordinal social identity scales based on verbal responses. Further, the significant correlations between social identity measured by PPM and social identity measured by verbal responses suggested the validity of PPM. We discussed the possibilities and disadvantages of the PPM as a tool of social research.
    Download PDF (593K)
  • Yiping ZHANG
    2009 Volume 36 Issue 1 Pages 15-24
    Published: 2009
    Released on J-STAGE: August 13, 2009
    JOURNAL FREE ACCESS
    This research compared the two-parameter and three-parameter item-response-models based on the data and theory, and showed the tendencies of those two models. Two-parameter model overestimated the information of the subjects whose ability value were estimated to low, and underestimates the information of the subjects whose ability value were estimated to high. Three-parameter model overestimated the information of the subjects whose ability value were estimated to high, and underestimates the information of the subjects whose ability value were estimated to low. Moreover, because the three-parameter model cannot distinguish whether the correct answer is carried out by the item unique necessary ability or guessing, in an item with low discrimination parameter and difficulty parameter, the guessing parameter becomes bigger. Furthermore, the mutual influence among the guessing parameter, the discrimination parameter and the difficulty parameter were also examined.
    Download PDF (587K)
Note
  • Manabu IWASAKI, Kasumi DAIDOJI
    2009 Volume 36 Issue 1 Pages 25-34
    Published: 2009
    Released on J-STAGE: August 13, 2009
    JOURNAL FREE ACCESS
    Statistical analysis of the number of correct answers of a test consisting of n items will be considered in this article. A beta-binomial distribution is assumed to the model of the distribution of test scores in this article. For such test scores, the situation of extra zeros sometimes arises in practical applications. This can happen in WEB testing, for example, when examinees miss-operate their PC that may make their test scores zero. To model such extra zeros we consider zero-inflated beta-binomial (ZIBB) distribution, which is a mixture of a beta-binomial distribution and a distribution that has mass at zero. For the estimation of the ZIBB model parameters we adopt the method of moments that is easily implemented in PC. As an extension of the ZIBB model it is also considered a mixture model of a beta-binomial distribution and a binomial distribution with parameters known, abbreviated as BB+B. This model can be used for a particular test procedure. An estimation procedure of the parameters of BB+B model is provided. It will be shown that such models fit well to datasets from practical applications.
    Download PDF (456K)
Data
  • Hideo AIZAKI
    2009 Volume 36 Issue 1 Pages 35-46
    Published: 2009
    Released on J-STAGE: August 13, 2009
    JOURNAL FREE ACCESS
    This paper aims to develop an application program for the design and statistical analysis of choice experiments with R. Microsoft Excel is used as the user interface, and RExcel, which is included in the R-(D)COM Server, connects MS Excel and R. The AlgDesign and the survival package in R are used for designing the choice sets and analyzing a data set respectively. In the Japanese environment, this program has several advantages in comparison with MS Excel-based application programs or using R via the command prompt mode. First, it can design the choice sets via the GUI mode. Second, it automatically generates the data set for a conditional logit model. Finally, the accuracy of the estimation result is improved. However, since this program is designed for novice users of the choice experiments, the upper bound of the number of attributes, levels, alternatives per choice set, and choice sets is imposed, and an analytical model is limited to the conditional logit model.
    Download PDF (1124K)
  • Masami ASAKAWA
    2009 Volume 36 Issue 1 Pages 47-61
    Published: 2009
    Released on J-STAGE: August 13, 2009
    JOURNAL FREE ACCESS
    In the empirical analysis for grasping “multi-dimension of consumer perception of TV commercials,” measurement items varied among previous studies, and the study targets included a broad range of television commercials, which advertise various products and brands. Therefore, each study extracted arbitrary dimensions. In this circumstance, this study aimed to extract the robust “consumer perception of TV commercials” characteristics, with reference to the results of previous studies. In detail, 156 female students underwent an audiovisual experiment to evaluate 20 items regarding “consumer perception of TV commercials” on a 7-point scale for 30 television commercials of food products. Then, I conducted the factor analysis and extracted five factors: “stimulation,” “conveyance,” “sensation,” “utility,” and “dignity.” Furthermore, I checked the reliability, the robustness and the appropriateness of the structures of these five factors.
    Download PDF (417K)
feedback
Top