The National Institute for Japanese Language conducted an International Census on Attitudes toward Languages in 28 countries/areas of the world from 1997 to 1998. This census was planned for the purpose of understanding objectively the actual state of the Japanese language, and focused primarily on how the Japanese language, Japanese people and Japan are viewed internationally in the context of internationalization in Japan. This paper outlines the census and helps you understand the subsequent 3 papers.
Between 1996 and 1998, the National Institute for Japanese Language conducted a survey among 28 nations, the main theme of which was “A General Study of the Japanese Language in Global Society”. Including only the 18 of the 28 nations whose samples were nationwide, the purpose of this report is to focus on the “International Census on Japanese Language Usage”, a sub-theme of the General Study. To do this, the report introduces and summarizes (1) the items which relate to sampling design and interviewee selection, and (2) the items which relate to the survey's conduct, such as the sampling methods used by interviewers. These two items then form the basis for our investigation of data quality. In cases where data containing differing total samples are compared, ideally the collected data are homogeneous. In cases where the samples being surveyed are nations, whose cultures, institutions and environments naturally vary, problems of comparability are inevitable. Data quality is affected by various factors at each stage of the survey. This is especially true when cross-national surveys are conducted where sampling designs and interviewee selection procedures vary, as these are two of the most notable items to be affected. In our survey design, we emphasized to each person in charge of the individual national surveys that data quality would be enhanced by maximizing the homogeneity among nations. However, in the cross-national surveys in our survey field, there are some issues which can rarely satisfy this condition. Here, we present concrete outlines of sampling designs for surveys which we have conducted among several nations in our cross-national studies.
The main objective of this paper is to investigate East & South East Asian peoples' views on Japanese as represented in the response data of the Japanese census 1997-98 by the National Institute for Japanese Language. Firstly, we summarize the historical and social backgrounds of those 10 Asian countries under investigation, concerning their conditions of language learning. Secondly, we analyze their images on Japanese with respect to certain five linguistic dimensions as well as the relationships between their images about Japanese, Japan, and the Japanese. The result shows that these factors are closely related to each other. Thirdly, we show that most Asian countries expect their further advancement in the relationships with Japan in economy, but that this may not necessarily lead to the rapid increase of learners of Japanese. Finally, we suggest some possibility about the near future development of East & South East Asian zone.
The purpose of this paper is to illustrate the utility of Facet Approach developed by L. Guttman and his group by using the example of the questionnaire design and data analysis of the International Census on Attitudes toward the Japanese Language. Facet Approach is composed of three parts : (1) Facet Design. This involves the construction (Structuples) and use of Mapping Sentences for questionnaire design. (2) Facet Analysis. This includes Scalogram Analysis, Partial Order Scalogram Analysis, Smallest Space Analysis, Median Regression Analysis, and so on. (3) Facet Theory. The First Law, The Second Law, and the Laws of Polytone Regressions have been proposed and tested in diverse empirical contexts. I attempt to summarize the basic ideas of Facet Approach by focusing on the above three areas, which I believe are useful for the systematic design of a questionnaire survey. Facet Design can be used to analyze questionnaires to determine the completeness of the information gathered. Some examples of such analyses are shown. Facet Theory guides the data analysis in that it identifies meaningful and interpretable relations which can be explored statistically. In this paper I demonstrate three different kinds of data analysis (Facet Analysis) guided by Facet Theory.
Until now, social surveys (i.e., public opinion surveys) heve been conducted using the face-to-face interview method. This is due to the facts that: (a) resident registrations (or voting registries), which are regarded as a sampling frame which adequately represents total populations, are available in Japan and can relatively easily facilitate obtaining unbiased probability samples, and (b) the reliability of the face-to-face interview method in surveys is regarded as high. Conversely, from the perspective of the spread of household telephones, the telephone survey method has been regarded as one by which it is impossible to obtain probability samples which represent ordinary citizens. Since the 1980s, household telephone coverage has increased (to over 90%). Consequently, the telephone survey method, which is especially useful for its promptness, has been used for voter forecasting surveys, and in the 1990s this method also has been utilized in public opinion surveys. However, not enough research on the telephone survey method has yet been done with respect to the representativeness of its samples or the reliability of survey results. This paper discusses past and current research regarding telephone surveys (specifically the current RDD survey method), and actual survey results, including their errors (except sampling errors). Moreover, by comparing the responses using both the telephone survey method and face-to-face interviews, the differences between the two methods are discussed. The current RDD method is viable; however, we cannot predict this method's future due to developments in communication technology. Finally, because there are many problems to overcome with respect to actual survey methods, further research, e.g., accumulating actual survey results and re-analyzing existing survey data, is necessary.
This paper presents the results of experimental telephone surveys conducted via two kinds of sampling methods; namely, telephone directory sampling (TD) and random digit dialing sampling (RDD). The TD sample is restricted to those who list their phone number in the directory. This restriction increases the proportion of the aged and the rural sample and causes results which are definitely favorably biased toward a conservative direction when compared with the results of the RDD which includes both the listed and the unlisted samples. Even if we compare TD with RDD limited to the listed respondents, there still remain a number of differences. In order to explain the discrepancies, we propose to divide each sample into two latent classes; one common class which would respond in both survey methods and one unique class which would not be obtained through the alternative method. The results given by a normal mixture model indicate the listed RDD respondents seem contain a moderate size of the unique class.
In this paper, I examined the possible use of random digit dialing sampling (RDD), focusing on the method improved by Waksberg which is often used by the public health professions. Since the telephone possession rate is 90% or more, the probability of selection bias is very low. There is the possibility that individuals who live out of the target area may inadvertently be included in the survey since the object area of the telephone may not coincide with the area of an administrative district. However, this situation can be avoided by confirming the living area at the time of enforcement or by expanding the area of the survey. The numbers called and the times of the survey calls also have an effect on the proportion of survey participants. The importance of making a sufficient number of calls and at the most opportune survey time is recommended for emphasis at the survey design stage.
In this article, we proposed a segmentation method of transaction data by making use of a Latent Class Model. In the analysis, not only a standard Latent Class Model, but also a model that relaxed the assumption of local independence was applied to the transaction data of a convenience store. Consequently, based on the AIC criterion, a model with six classes was adopted. All of these segments were interpreted as having a different purchace-intentions. Additionally, the purchase-intention of a coming-into-the-store visitor and the size of the purchase could be guessed from transaction data. Furthermore, it was shown that the information acquired by analyzing transaction data by our method may translate into useful information for working on a store policy corresponding to the characteristics of the coming-to-the-store visitors of a convenience store from a practical viewpoint.
The Contingent Valuation Method (CVM), an environmental valuation method worthy of attention, has been widely used to estimate the value of an environmental good with a non-market property based on survey data relating to the consumer's Willingness To Pay (WTP) in past decades. Nonsampling errors from measurement instruments in a sampling survey, however, often result in biased quantitative estimates of environmental values. This paper provides a practical discussion on the problems leading to nonsampling errors in CVM, through analysis of the effects of the payment vehicle on WTP in the sampling survey data valuing the surface forest park assumed to be constructed at the reclaimed land located in Tokyo Gulf. In order to concentrate on the measurement error, the CVM survey was carried out using the dichotomous choice (DC) format, and the payment vehicle in the questionnaire includes both tax and fund scenarios. Based on results from both the single-bound DC and the double bound DC, the statistical data analysis has demonstrated that the payment vehicle can affect the estimates of WTP significantly.
The primary focus of this paper is an examination of the construct validity of cognitive age measurement. Through a review of cognitive age research, it is concluded that cognitive age has not been well grounded with respect to construct validity although it can basically enrich one's understanding of older people's self concepts, attitudes, and behaviors. In this research, multitrait-multimethod matrix data consisting of 9 variables (cognitive age, ideal age, and least desired age, as measured by three methods; Staple, Semantic Differential, and Likert scales) was analyzed in the framework of hierarchically nested covariance structure models in order to test the significance of convergent and discriminant validity. A path analytic conceptualization performed well in terms of goodness-of-fit, and the findings provide strong support for construct validity of cognitive age measurement.