Multiple Correspondence Analysis Technique Used in Analyzing the Categorical Data in Social Sciences

2007 ◽  
Vol 7 (4) ◽  
pp. 585-588 ◽  
Author(s):  
Duygu Akturk ◽  
Sema Gun ◽  
Taner Kumuk
Keyword(s):  
Social Sciences ◽  
Correspondence Analysis ◽  
Categorical Data ◽  
Multiple Correspondence Analysis ◽  
Analysis Technique

Correspondence Analysis: Graphical Representation of Categorical Data in Marketing Research

1986 ◽  
Vol 23 (3) ◽  
pp. 213-227 ◽  
Author(s):  
Donna L. Hoffman ◽  
George R. Franke
Keyword(s):  
Correspondence Analysis ◽  
Categorical Data ◽  
Graphical Representation ◽  
Dimensional Space ◽  
Soft Drink ◽  
Marketing Research ◽  
Data Matrix ◽  
Soft Drink Consumption ◽  
Analysis Technique ◽  
Low Dimensional

Correspondence analysis is an exploratory data analysis technique for the graphical display of contingency tables and multivariate categorical data. Its history can be traced back at least 50 years under a variety of names, but it has received little attention in the marketing literature. Correspondence analysis scales the rows and columns of a rectangular data matrix in corresponding units so that each can be displayed graphically in the same low-dimensional space. The authors present the theory behind the method, illustrate its use and interpretation with an example representing soft drink consumption, and discuss its relationship to other approaches that jointly represent the rows and columns of a rectangular data matrix.

scholarly journals Using multiple correspondence analysis to identify behaviour patterns associated with overweight and obesity in Vanuatu adults

2019 ◽  
Vol 22 (09) ◽  
pp. 1533-1544 ◽  
Author(s):  
Andrew van Horn ◽  
Charles A Weitz ◽  
Kathryn M Olszowy ◽  
Kelsey N Dancause ◽  
Cheng Sun ◽  
...  
Keyword(s):  
Body Fat ◽  
Correspondence Analysis ◽  
Categorical Data ◽  
Multiple Correspondence Analysis ◽  
Overweight And Obesity ◽  
Data Sets ◽  
Pacific Island ◽  
Binary Responses ◽  
Or Groups ◽  
The Us

AbstractObjectiveThe present study evaluates the use of multiple correspondence analysis (MCA), a type of exploratory factor analysis designed to reduce the dimensionality of large categorical data sets, in identifying behaviours associated with measures of overweight/obesity in Vanuatu, a rapidly modernizing Pacific Island country.DesignStarting with seventy-three true/false questions regarding a variety of behaviours, MCA identified twelve most significantly associated with modernization status and transformed the aggregate binary responses of participants to these twelve questions into a linear scale. Using this scale, individuals were separated into three modernization groups (tertiles) among which measures of body fat were compared and OR for overweight/obesity were computed.SettingVanuatu.ParticipantsNi-Vanuatu adults (n 810) aged 20–85 years.ResultsAmong individuals in the tertile characterized by positive responses to most of or all the twelve modernization questions, weight and measures of body fat and the likelihood that measures of body fat were above the US 75th percentile were significantly greater compared with individuals in the tertiles characterized by mostly or partly negative responses.ConclusionsThe study indicates that MCA can be used to identify individuals or groups at risk for overweight/obesity, based on answers to simply-put questions. MCA therefore may be useful in areas where obtaining detailed information about modernization status is constrained by time, money or manpower.

scholarly journals Multiple correspondence analysis as a tool in sociological research

Sociologija ◽  
2021 ◽  
Vol 63 (1) ◽  
pp. 26-49
Author(s):  
Ivica Mladenovic
Keyword(s):  
Social Sciences ◽  
Correspondence Analysis ◽  
Multiple Correspondence Analysis ◽  
Empirical Studies ◽  
Sociological Research ◽  
Jean Paul ◽  
Theoretical Assumptions ◽  
Supplementary Variables ◽  
The 1960S ◽  
Dimensional Graph

Multiple correspondence analysis is a form of a factorial multivariate data analysis which helps us summarize a large quantity of information with assistance of modern statistical programs. In this way, sociological correlations - which are then visually represented on a two-dimensional graph - are established between a certain number of active and supplementary variables, i.e. between the positions, dispositions and position-takings (French: prises de position) by the analyzed agents. It is a research tool initially introduced by a statistician Jean- Paul Benz?cri in the humanities and social sciences in the 1960s. Since the early 1970s, thanks to Pierre Burdieu, this methodological procedure has become an indispensable instrument in the empirical studies of sociologists who were familiar with structuralist approach in sociology. This paper consists of two main sections. The first one sets out the basic theoretical assumptions and methodological properties of multiple correspondence analysis. The aim of the second section is to give a brief recital of one particular research - i.e. its features and results - in order to get the interested sociological public aquaitned with it?s practicle potentials of this tool.

Sign in / Sign up

Export Citation Format

Share Document