Information theory: structural models for qualitative data by Klaus Krippendorff

By Klaus Krippendorff

Krippendorff introduces social scientists to info idea and explains its program for structural modeling. He discusses key issues akin to: how you can make certain a data conception version; its use in exploratory examine; and the way it compares with different ways equivalent to community research, course research, chi sq. and research of variance. details thought simplifies and clarifies a fancy yet robust statistical process for analysing multivariate qualitative info. it is going to serve either as a textbook and as a sourcebook for researchers in conversation idea, info concept and platforms concept.

Show description

Read Online or Download Information theory: structural models for qualitative data PDF

Similar information theory books

Theory of Information: Fundamentality, Diversity and Unification (World Scientific Series in Information Studies)

This precise quantity provides a brand new method - the overall thought of data - to medical figuring out of data phenomena. in accordance with a radical research of data strategies in nature, know-how, and society, in addition to at the major instructions in info thought, this idea synthesizes present instructions right into a unified method.

Managing Economies, Trade and International Business

The present section of globalization and the elevated interconnectedness of economies via alternate have encouraged the administration and development charges of economies and likewise the aggressive and managerial matters for corporations. This publication specializes in 3 major matters – financial progress and sustainable improvement; exchange, legislations and legislation; and aggressive and managerial matters in overseas company – from a multidisciplinary, transversal and eclectic viewpoint.

Efficient Secure Two-Party Protocols: Techniques and Constructions

The authors current a entire examine of effective protocols and methods for safe two-party computation – either common structures that may be used to safely compute any performance, and protocols for particular difficulties of curiosity. The e-book specializes in concepts for developing effective protocols and proving them safe.

Information Theory and Best Practices in the IT Industry

​​​The significance of benchmarking within the provider quarter is easily well-known because it is helping in non-stop development in items and paintings techniques. via benchmarking, businesses have strived to enforce most sensible practices so that it will stay aggressive within the product- industry during which they function. besides the fact that stories on benchmarking, relatively within the software program improvement quarter, have overlooked utilizing a number of variables and as a result haven't been as entire.

Extra info for Information theory: structural models for qualitative data

Sample text

The preelection process reduced this entropy to nearly 1 bit (characterizing the nearly equally likely success of Carter and Ford before election day on which voters removed the remaining uncertainty). Finally, entropy is also the key to a fundamental law in the cybernetics of regulation. Ashby's (1956) law of requisite variety, which states that "only variety can destroy variety," implies that the survival of a system depends on its ability to generate at least as much variety within its boundaries as exists in the form of threatening disturbances from its environment.

Part of the definition of entropy, and one reason for calling it a measure of observational variety, is that unobserved possibilities do not enter the measure. The average is computed only for categories of observations that do occur at least once. This is reflected in the convention adopted here: 0log20 = 0. Note some of its properties. Entropies are zero or positive and limited by Page 17 where NA is the number of categories in variable A and n is the sample size. The entropy is zero when observational variety is absentthat is, when distinctions are vacuous and all observations are of the same kind, pa = 1 for one category and zero for all others, in which case both 1log21 = 0 and by convention 0log20 = 0.

This places information theory into a framework that most social scientists can readily comprehend and evaluate. Professor Krippendorff's thorough understanding of the theory and use of information theory also takes the careful reader a long way toward competency. This book is particularly successful at making a rather complicated system for analyzing multivariate qualitative data as simple as possible. Krippendorff does this by building the entire presentation around intuitively appealing notions of information, such as the amount of information provided by an answer to a question, the amount transmitted through a noisy channel, and so on, rather than by using the axioms and theorems of information theory.

Download PDF sample

Rated 4.16 of 5 – based on 48 votes