China 
Africa 
Explore chapters and articles related to this topic
The US airlines relative positioning based on attributes of service quality
Published in Thomas C. Lawton, Strategic Management in Aviation, 2017
Gursoy a Dogan, Chen b * Ming-Hsiang, Kim c Hyun Jeong
Correspondence analysis is often used in positioning and image studies where the researcher wants to explore the relationship between brands, between attributes, and between brands and attributes (Whitlark & Smith, 2001) because it represents graphically the row and column categories and allows for a comparison of their correspondences or associations at a category level. Correspondence analysis has several features that enable researchers to have a better understanding of the relationships among variables. The most important feature of correspondence analysis is its multivariate nature, which enables multivariate treatment of multiple categorical data simultaneously. The multivariate nature of correspondence analysis can reveal relationships that would not be detected in a series of pairwise comparisons of variables. Correspondence analysis also helps to show how variables are related, not just that the relationship exists. The joint graphical display obtained from correspondence analysis can help in detecting structural relationships among the variable categories (Hoffman & Franke, 1986).
Applied Multivariate Statistics
Published in Nick Zacharov, Sensory Evaluation of Sound, 2018
As you have hopefully understood, correspondence analysis is the multidimensional method dedicated to the analysis of the dependence between two categorical variables, from the point of view of their categories. In other words, CA is dedicated to the analysis of the correspondence between the categories of one categorical variable and the categories of another. To apply this method on our contingency table, I will use the CA function of the FactoMineR package, then I will use the plot.CA function in order to represent the rows of our data set, i.e. the occupations.
Regional Observational Studies: Assembling and Exploring Data
Published in Susan B. Norton, Susan M. Cormier, Glenn W. Suter, Ecological Causal Assessment, 2014
Jeroen Gerritsen, Lester L. Yuan, Patricia Shaw-Allen, David Farrar
PCA is not effective for use with species composition data, because species are often distributed unimodally along environmental gradients. Also, species-by-site matrices often have large numbers of empty cells, representing sites where a given species was not found. Absence falsely contributes to similarity in PCA because the analysis uses correlation as the measure of similarity. For species composition data, we recommend NMS or correspondence analysis.
The spatial feature and use pattern of external space in Chongqing traditional urban settlement
Published in Journal of Asian Architecture and Building Engineering, 2023
Lin Chen, Kai Fang, Xinpeng Wang, Wenda Zhang, Guoqing Zhu, Zhehan Zhang, Nobuaki Furuya
Correspondence analysis is a multivariate statistical analytical method developed on the basis of R-type and Q-type factor analyses; therefore, it is also known as R-Q-type factor analysis. Correspondence analysis can reveal the difference across various categories of the same variable and the relationships between categories of different variables. It is a visualized analytical method, which can display several groups of data that seem to have no connection through a visually intuitive positioning map (Clausen 1998). Based on the spatial features and use patterns of the 143 samples collected from the spatial division and statistics described above, a cross-tabulation of the spatial types and use patterns at the three micro-levels (i.e., upper, side, and bottom) was constructed. Then, the SPSS software was employed to perform a correspondence analysis of the three levels of spatial features and use patterns. The purpose of this analysis was to obtain a corresponding map that clearly and intuitively shows the degree of correlation between different categories in a certain spatial element and the five utilization patterns.
Differentiating between fatal and non-fatal mining accidents using artificial intelligence techniques
Published in International Journal of Mining, Reclamation and Environment, 2020
Saki Gerassis, Ángeles Saavedra, Javier Taboada, Elena Alonso, Fernando G. Bastante
Correspondence analysis is a statistical technique that is used to analyse association relationships between factors from a graphical point of view. The objective is to reduce, with the least possible loss of information, a large amount of data to a small number of dimensions, usually two. When the number of factors analysed is greater than two, the technique is called multiple correspondence analysis (MCA). Although based on complex algebraic methods, it is a very intuitive technique, since it creates a map of the relative position of the factors that reflects the degree of association between them. While MCA is very similar to principal components analysis (PCA), they differ in that the Euclidean distance between observations is considered in PCA, whereas the distance , based on state frequencies, is considered in MCA. Measuring the distance consists of counting the frequencies of the connections between all possible states of the different factors.
Modern Psychometrics With R
Published in Technometrics, 2020
Chapter 7 considers correspondence analysis (CA) conducted on contingency tables by Chi-square components of the standardized residuals with the SVD applied to them, and the results obtained and displayed in plots built in the ca, anacor, ved, tm, plot3D, or ggtern packages. For a higher dimension data the multiple CA is performed using ca and anacor, and the configural frequency analysis (KFA) can be assessed with the cfa package, also useful for log-linear modeling. Chapter 8 of Gifi methods deals with assignment of numeric values to the Likert scale levels in the optimal scaling (OS) also called analysis levels. Gifi system uses a set of binary variables for each level of the originally categorical variables, and multivariate methods based on the binary sets include PCA (Princals), multiple CA (Homals), multiple regression (Morals), conjoint analysis (Addals), canonical correlation analysis (Canals), multiblock canonical correlation (Overals), discriminant analysis (Criminals), and linearized relations (Lineals). These techniques are implemented in the packages of Gifi, homals, and aspect. It is interesting to note that “Gifi is a pen name for a group of Dutch researchers, named after Francis Galton’s manservant Albert Gifi” (p. 233), and it seems the author of this book has been belonging to that group as well.