A. Kimball Romney - Böcker

Presents a set of closely related techniques that facilitate the exploration and display of a wide variety of multivariate data, both categorical and continuous. Three methods of metric scaling, correspondence analysis, principal components analysis, and multiple dimensional preference scaling are explored in detail for strengths and weaknesses over a wide range of data types and research situations. "The introduction illustrates the methods with a small dataset. This approach is effective--in a few minutes, with no mathematical requirement, the reader can understand the capabilities, similarities, and differences of the methods. . . . Numerical examples facilitate learning. The authors use several examples with small datasets that illustrate very well the links and the differences between the methods. . . . we find this text very good and recommend it for graduate students and social science researchers, especially those who are interested in applying some of these methods and in knowing the relationship among them." --Journal of Marketing Research "Illustrate[s] the service Sage provides by making high-quality works on research methods available at modest prices. . . . The authors use several interesting examples of practical applications on data sets, ranging from contraception preferences, to pottery shards from archeological digs, to durable consumer goods from market research. These examples indicate the broad range of possible applications of the method to social science data." --Contemporary Sociology "The book is a bargain; it is clearly written." --Journal of Classification

Based on a conference held in honor of Professor Tarow Indow, this volume is organized into three major topics concerning the use of geometry in perception: * space -- referring to attempts to represent the subjective space within which we locate ourselves and perceive objects to reside; * color -- dealing with attempts to represent the structure of color percepts as revealed by various experimental procedures; and * scaling -- focusing on the organization of various bodies of data -- in this case perceptual -- through scaling techniques, primarily multidimensional ones. These topics provide a natural organization of the work in the field, as well as one that corresponds to the major aspects of Indow's contributions. This book's goal is to provide the reader with an overview of the issues in each of the areas, and to present current results from the laboratories of leading researchers in these areas.