Metric scaling : correspondence analysis /
Three methods of metric scaling - correspondence analysis, principal components analysis and multiple dimensional preference scaling - are explored in detail for their strengths and weaknesses over a wide range of data types and research situations.
| Основен автор: | Weller, Susan C. |
|---|---|
| Други автори: | Romney, A. Kimball |
| Формат: | Електронна книга |
| Език: | English |
| Публикувано: |
Newbury Park, Calif. :
Sage Publications,
℗♭1990.
|
| Серия: |
Quantitative applications in the social sciences ;
no. 07-075. |
| Предмети: | |
| Онлайн достъп: |
http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=24745 |
| Подобни документи: |
Print version::
Metric scaling. |
Съдържание:
- 1. Introduction
- Some sample results
- Some background comments
- The central unifying theme of the monograph
- 2. The basic structure of a data matrix
- Basic structure of a matrix
- Transformations
- 3. Principal components analysis
- Single factor example
- Multifactor example
- 4. Multidimensional preference scaling
- 5. Correspondence analysis of contingency tables
- The mechanics of correspondence analysis
- The reconstruction of expected and observed data
- Another perspective: The case approach with indicator variables
- 6. Correspondence analysis of nonfrequency data
- Correspondence analysis of rank-order data
- Correspondence analysis of proximities
- 7. Ordination, seriation, and Guttman scaling
- The horseshoe effect
- Guttman scaling as a special case of correspondence analysis
- An invariance property of multiple-way indicator matrices
- 8. Multiple correspondence analysis
- Multiple comparisons using "stacked" matrices
- A few final words.