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.

Performance Improvement in the Pattern Classification of Nominal Data Sets Applying Multiple Correspondence Analysis

2014 ◽  
Vol 670-671 ◽  
pp. 1482-1487
Author(s):  
Rodrigo Clemente Thom de Souza ◽  
Maria Teresinha Arns Steiner ◽  
Leandro dos Santos Coelho
Keyword(s):  
Correspondence Analysis ◽  
Multiple Correspondence Analysis ◽  
Principal Component ◽  
Knowledge Discovery In Databases ◽  
Data Sets ◽  
Data Types ◽  
Nominal Data ◽  
Linear Discriminant ◽  
Whole Process ◽  
Geometric Data Analysis

Classification is a supervised learning problem used to discriminate data instances in different classes. The solution to this problem is obtained through algorithms (classifiers) that look for patterns of relationships between classes in known cases, using these relationships to classify unknown cases. The performance of the classifiers depends substantially of the data types. In order to give proper treatment to nominal data, this paper shows that the application of previous transformations can substantially improve the performance of classifiers, bringing significant benefits to the result of the whole process of Knowledge Discovery in Databases (KDD). This paper uses three different data sets with nominal data and two well-known classifiers: the Linear Discriminant Analysis (LDA), and the Naïve-Bayes (NB). For data transformation, the paper applies an approach called Geometric Data Analysis (GDA). The GDA techniques compared in this paper are the traditional Principal Component Analysis (PCA) and the underexplored Multiple Correspondence Analysis (MCA). The results confirm the capability of the GDA transformation to improve the classification accuracy and attest the superiority of the MCA in comparison with its precursor, the PCA, when applied to nominal data.

scholarly journals Impact of orange juice consumption on macronutrient and energy intakes and body composition in the US population

2012 ◽  
Vol 15 (12) ◽  
pp. 2220-2227 ◽  
Author(s):  
Ying Wang ◽  
Beate Lloyd ◽  
Meng Yang ◽  
Catherine G Davis ◽  
Sang-Gil Lee ◽  
...  
Keyword(s):  
Body Composition ◽  
Waist Circumference ◽  
Children And Adolescents ◽  
Body Fat ◽  
Orange Juice ◽  
Overweight And Obesity ◽  
Cross Sectional ◽  
Percentage Body Fat ◽  
Significant Difference ◽  
The Us

AbstractObjectiveThe present study evaluated the contribution of 100 % orange juice (OJ) consumption to the intakes of macronutrients and energy and its impact on body composition.DesignA cross-sectional study was conducted. The main exposure was OJ consumption based on two non-consecutive 24 h diet recalls. Macronutrient and energy intakes and body composition parameters were outcome measures. All statistical analyses were carried out using SAS and SUDAAN statistical software packages to allow for multistage sample designs.SettingThe US population and its subgroups.SubjectsThe US population aged ≥4 years (n 13 971) from the National Health and Nutrition Examination Survey 2003–2006, conducted by the National Center for Health Statistics.ResultsIn this US population, OJ consumers had lower BMI and healthier lifestyle behaviours (including lower alcohol consumption and smoking as well as higher exercise level) than non-consumers (P < 0·05). After adjusting for covariates, OJ consumers had higher daily intakes of carbohydrate, total sugar, total fat and energy than non-consumers (P < 0·01). However, these linear trends still remained even after OJ was removed from the food list of items consumed. Adult OJ consumers had lower BMI, waist circumference and percentage body fat than non-consumers (P < 0·01), as well as lower odds ratio for overweight and obesity (P < 0·01). These effects were not seen in children and adolescents, where there was no significant difference in BMI, waist circumference and percentage body fat in OJ consumers compared with non-consumers.ConclusionsOJ consumption was associated with healthier body composition in adults; while there were no significant associations between OJ consumption and body composition in children and adolescents.

Multiple Correspondence Analysis and Geometric Data Analysis

2021 ◽  
pp. 177-184
Author(s):  
Amal Tawfik ◽  
Stephan Davidshofer
Keyword(s):  
Data Analysis ◽  
Correspondence Analysis ◽  
Multiple Correspondence Analysis ◽  
Principal Component ◽  
Categorical Variables ◽  
Large Set ◽  
Geometric Data ◽  
Jean Paul ◽  
Geometric Data Analysis ◽  
Or Groups

This chapter focuses on multiple correspondence analysis (MCA), which is a factor analysis statistical method used to analyse relations between a large set of categorical variables. Developed by Jean-Paul Benzécri in the early 1970s, MCA is one of the principal methods of geometric data analysis (GDA). Three different statistical methods can be identified as GDA: correspondence analysis (CA), which enables the cross-tabulation of two categorical variables; MCA for the analysis of a matrix of individuals and categorical variables; and principal component analysis (PCA), which uses numerical variables. In GDA, data is represented as a cloud of points to allow statistical interpretations. Although MCA is a relational method, it differs from social network analysis (SNA) as it focuses on the objective relations that characterize actors or groups, rather than the effective relations.

Multiple correspondence analysis for “tall” data sets

2009 ◽  
Vol 13 (6) ◽  
pp. 873-885 ◽  
Author(s):  
Angelos Markos ◽  
George Menexes ◽  
Theophilos Papadimitriou
Keyword(s):  
Correspondence Analysis ◽  
Multiple Correspondence Analysis ◽  
Data Sets

Exploring Historical Canadian Foodborne Outbreak Data Sets for Human Illness Attribution

2009 ◽  
Vol 72 (9) ◽  
pp. 1963-1976 ◽  
Author(s):  
A. RAVEL ◽  
J. GREIG ◽  
C. TINGA ◽  
E. TODD ◽  
G. CAMPBELL ◽  
...  
Keyword(s):  
Correspondence Analysis ◽  
Multiple Correspondence Analysis ◽  
Data Sets ◽  
Foodborne Outbreak ◽  
Escherichia Coli Infection ◽  
Foodborne Outbreaks ◽  
Human Illness ◽  
Illness Attribution ◽  
Changes Over Time ◽  
Over Time

Human illness attribution has been recently recognized as an important tool to better inform food safety decisions. Analysis of outbreak data sets has been used for that purpose. This study was conducted to explore the usefulness of three comprehensive Canadian foodborne outbreak data sets covering 30 years for estimating food attribution in cases of gastrointestinal illness, providing Canadian food attribution estimates from a historical perspective. Information concerning the microbiological etiology and food vehicles recorded for each outbreak was standardized between the data sets. The agent–food vehicle combinations were described and analyzed for changes over time by using multiple correspondence analysis. Overall, 6,908 foodborne outbreaks were available for three decades (1976 through 2005), but the agent and the food vehicle were identified in only 2,107 of these outbreaks. Differences between the data sets were found in the distribution of the cause, the vehicle, and the location or size of the outbreaks. Multiple correspondence analysis revealed an association between Clostridium botulinum and wild meat and between C. botulinum and seafood. This analysis also highlighted changes in food attribution over time and generated the most up-to-date food attribution values for salmonellosis (29% of cases associated with produce, 15% with poultry, and 15% with meat other than poultry, pork, and beef), campylobacteriosis (56% of cases associated with poultry and 22% with dairy products other than fluid milk), and Escherichia coli infection (37% of cases associated with beef, 23% with cooked multi-ingredient dishes, and 11% with meat other than beef, poultry, and pork). Because of the inherent limitations of this approach, only the main findings should be considered for policy making. The use of other human illness attribution approaches may provide further clarification.

Sign in / Sign up

Export Citation Format

Share Document