Hayashiâs quantification methods are multivariate analysis techniques dealing with qualitative data, which were developed by Dr. Chikio Hayashi, former Director-General of the Institute of Statistical Mathematics (Tokei-Suri Kenkyuzyo), in the late 1940s and 1950s. The methods have been, and are, widely used for academic and business fields including marketing research in Japan, hence their inclusion here. The methods traditionally employ the following terminology: 1. âexternal criterionâ for dependent variable; 2. âexplanatory variablesâ for independent variables; 3. âitemsâ for question items in the questionnaire; and 4. âcategoriesâ for alternative answers to the question.
A chapter such as this can only scratch the surface in terms of informing the reader what techniques are available, what they do, how they do it, and what are the pitfalls, so it must be seen as purely introductory. Nevertheless, it attempts to introduce the subject in as undemanding a way as possible, using verbal rather than mathematical descriptions of the techniques wherever possible. Complex mathematical descriptions are, we hope, kept to a minimum. Multivariate analysis is broadly concerned with the relationships between a set of variables. How similar are they? Are they correlated? Can they be summarised effectively? Are they predictive of outcomes in any sense? Are there groups of respondents with similar behavioural or attitudinal patterns as measured by these variables? Multivariate techniques attempt to answer questions such as these.
Factor Analysis (like the similar technique of Principal Components Analysis) is a multivariate statistical method used primarily for data reduction. Itis usually employed to reduce the columns of our data matrix, that is the variables measured for each respondent. The reduction Is done by grouping together those variables which are intercorrelated as measured by the coefficient of correlation.
In our data matrix we have one variable of particular interest and we wish to see how it is affected by movements in a single explanatory variable. For example, the variable we might be interested in is Sales in £mn and we may wish to know how it is affected by advertising expenditure in £000's. The rows of our data matrix may consist of the last seven years of data for these two variables.
A second generation of multivariate methods is now available for marketing research. This article: shows the specific characteristics and possible links of the second generation methods of multivariate analysis with the traditional first generation approaches, puts the stress on the main features of each of the new methods, examines their impact and respective contributions for theory testing in marketing research. The impact of second generation methods in terms of methodology for theory testing and measurement validity is of prime importance. They are not only more powerful than earlier multivariate techniques, they also bring a new perspective to overall research methodology.
The purpose of this communication is to expose the basic principles of causal modelling, and to examine its conditions of application to marketing research. It will be showed that these models are part of a "second generation" of multivariate analysis. The switch from an exploratory approach to a confirmatory one contributes to fill the gap, and reinforce the links, between data analysis and model building.
The purpose of this communication is to expose the basic principles of causal modelling, and to examine its conditions of application to marketing research. It will be showed that these models are part of a "second generation" of multivariate analysis. The switch from an exploratory approach to a confirmatory one contributes to fill the gap, and reinforce the links, between data analysis and model building.
The paper considers the current state of multivariate analysis when the data available only take the form of binary and classified variables. This restriction has meant that researchers have had to develop new techniques to deal with problems of defining and segmenting markets. An encouraging amount of work has been done in this field in the last twenty years and now there are statistical methods for handling the various types of dependence and interdependence analyses. The paper reviews some of the literature that shows how market analysts proceed with model building when the data are not normally distributed measured variables. Among the techniques cited are those under the general umbrella of Log Linear Model!ing, which enable regression and correlation type procedures to be employed and new ways of clustering respondents. The paper concludes that this growing area will continue to flourish especially as computer programs are available for many of the techniques.
Consumption patterns and motivations are constantly changing with the complex developments in society as a whole. In the present study recent trends in the consumption of tobacco and alcohol are analysed. Data from surveys measuring use and non- use of tobacco and alcohol in Norway in 1979 and 1981 are used for describing changes in the patterns of consumption. Through bi- and multivariate analyses relationships between a set of demographic variables and use of the stimulants are assessed. An evaluation of the changes in the demographic structure of users and non-users is used for analysing possible future trends in the consumption.
The purpose of this communications is to review the main applications of multivariate methods to a preeminent area of marketing research consumer behavior. Three areas of application are distinguished, based on a paradigm of consumer behavioral system: 1. market segmentation; 2. consumer perceptions and product positioning; 3. preferences and choices. For each area, available methods are briefly presented in a comparative framework. The conclusive part aims to identify the main trends of evolution of multivariate data analysis.
Only a flexible, sophisticated and scientifically based system can give a satisfactory answer all sides can agree to and use as rational guidelines for the future. Colgate- Palmolive utilizes a MMIS to analyse the effect of its marketing-mix and to give medium and long-range sales forecasts. In addition to data processing the main role of this MMIS is to process and evaluate market information. The structure of this MMIS, which has become known in Europe under the name of MARKET, and some of its possible applications will be discussed in the paper "FORSYS/MAVIS and MAR- KET/E. I.S. - forecasting-oriented information systems for use on the operational and strategic levels of decision making" by J.M. Becher. Here, the structure of a multivariate market model is to be described, which Colgate-Palmolive uses to explain and forecast the market for nappies in an European country. This model was constructed with the above mentioned system and is being continuously up-dated.
In the marketing research literature, the term 'multi- variate analysis' has been broadly used to include all statistical procedures that are concerned with the simultaneous analysis of the relationship among several variables. The emphasis in the presentation will be on international applications of one or more of the multivariate procedures.