Multi-correspondence Analysis Research Articles

Association of behavior pattern with overweight and obesity in South Korean adults-A multi correspondence analysis (KNHANES-2018-2020).

The objective was to determine the association between health-related behaviour with overweight and obesity in South Korean adults by using the Korean National Health and Nutritional Examination Survey (KNHANES) 2018-2020. The study participants were 16,784 aged ≥ 20years. The variables were socio-demographic, lifestyle, food habits and metabolic conditions. The logistic regression analysis performed to find the association by the odds ratio (OR, 95% CI). MCA performed to identify risk factors were computed for overweight and obesity. Overweight and obesity were significantly associated with health behaviour, high income (OR = 1.26; 95% CI: 1.15-1.39), smoking(OR = 1.29; 95% CI: 1.08-1.53), low physical activity(OR = 3.23; 95% CI: 1.79-4.69), diabetes(OR = 2.70; 95% CI: 1.62-4.50), high cholesterol and low HDL(OR = 3.98; 95%CI:2.65-5.97). The high discriminant variables of MCA were aged over 60years, lower education, high income, diabetes, lack of physical activity, and high cholesterol. The findings confirm that the OR of obesity and overweight was likely associated with health behaviour patterns. Besides, it indicates the MCA would be very effective to identify the population-based data context than individual data and it may suggest that more research on association between health behaviours and obesity prevention interventions should be developed for each age group for better health outcomes.

Analysis of Clinical Phenotypes through Machine Learning of First-Line H. pylori Treatment in Europe during the Period 2013-2022: Data from the European Registry on H. pylori Management (Hp-EuReg).

The segmentation of patients into homogeneous groups could help to improve eradication therapy effectiveness. Our aim was to determine the most important treatment strategies used in Europe, to evaluate first-line treatment effectiveness according to year and country. Data collection: All first-line empirical treatments registered at AEGREDCap in the European Registry on Helicobacter pylori management (Hp-EuReg) from June 2013 to November 2022. A Boruta method determined the "most important" variables related to treatment effectiveness. Data clustering was performed through multi-correspondence analysis of the resulting six most important variables for every year in the 2013-2022 period. Based on 35,852 patients, the average overall treatment effectiveness increased from 87% in 2013 to 93% in 2022. The lowest effectiveness (80%) was obtained in 2016 in cluster #3 encompassing Slovenia, Lithuania, Latvia, and Russia, treated with 7-day triple therapy with amoxicillin-clarithromycin (92% of cases). The highest effectiveness (95%) was achieved in 2022, mostly in Spain (81%), with the bismuth-quadruple therapy, including the single-capsule (64%) and the concomitant treatment with clarithromycin-amoxicillin-metronidazole/tinidazole (34%) with 10 (69%) and 14 (32%) days. Cluster analysis allowed for the identification of patients in homogeneous treatment groups assessing the effectiveness of different first-line treatments depending on therapy scheme, adherence, country, and prescription year.

Ecological factors shaping the ectoparasite community assembly of the Azara's Grass Mouse, Akodon azarae (Rodentia: Cricetidae).

Parasites are integral members of the global biodiversity. They are useful indicators of environmental stress, food web structure and diversity. Ectoparasites have the potential to transmit vector-borne diseases of public health and veterinary importance and to play an important role in the regulation and evolution of host populations. The interlinkages between hosts, parasites and the environment are complex and challenging to study, leading to controversial results. Most previous studies have been focused on one or two parasite groups, while hosts are often co-infected by different taxa. The present study aims to assess the influence of environmental and host traits on the entire ectoparasite community composition of the rodent Akodon azarae. A total of 278 rodents were examined and mites (Mesostigmata), lice (Phthiraptera), ticks (Ixodida) and fleas (Siphonaptera) were determined. A multi-correspondence analysis was performed in order to analyze interactions within the ectoparasite community and the influence of environmental and host variables on this assembly. We found that environmental variables have a stronger influence on the composition of the ectoparasite community of A. azarae than the host variables analyzed. Minimum temperature was the most influential variable among the studied. In addition, we found evidence of agonistic and antagonistic interactions between ticks and mites, lice and fleas. The present study supports the hypothesis that minimum temperature plays a major role in the dynamics that shape the ectoparasite community of A. azarae, probably through both direct and indirect processes. This finding becomes particularly relevant in a climate change scenario.

Dynamics of community-acquired meningitis syndrome outbreaks in southern France.

In southern France, cases of community-acquired meningitis syndrome (CAM) are typically clustered as outbreaks with determinants which remain unknown. This 61-month retrospective investigation in Nîmes and Marseille university hospital laboratories, yielded 2,209/20,779 (10.63%) documented CAM cases caused by 62 different micro-organisms, represented by seasonal viral etiologies (78.8%), including Enterovirus, Herpes Simplex Virus (HSV), and Varicella-Zoster Virus (VZV; 1,620/2,209 = 73.4%). Multi correspondence analysis revealed an association of infection with age and sex, with the risk of infection being relatively higher in young men, as confirmed by Fisher's exact test (p < 10-3). Bacterial meningitis accounted for 20% of cases, mostly caused by Streptococcus pneumoniae (27.4% of cases), Neisseria meningitidis (12.5%), and Haemophilus influenzae (9.5%) with bacteria/virus coinfection (0.9%), and only six cases of documented fungal meningitis. In total, 62.6% of cases, of which 88.7% were undocumented, arose from 10 outbreaks. 33.2% of undocumented cases were aged >60 years compared to 19.2% of documented cases (p < 0.001), and viral infection was more common in the summer (87.5%) compared to other seasons (72.3%; p < 0.001). Outbreaks most often started in Nîmes and moved eastward toward Marseille at a speed of ~9 km/day, and these dynamics significantly correlated with atmospheric temperature, especially during summer outbreaks. In particular, the incidence of Enterovirus-driven outbreaks correlated with temperature, revealing correlation coefficients of 0.64 in Nîmes and 0.72 in Marseille, and its occurrence in Marseille lagged that in Nîmes by 1-2 weeks. Tracing the dynamics of CAM outbreak during this retrospective investigation in southern France yielded a speed of displacement that correlated with the variation in temperature between both cities, and these results provide clues for the next occurrence of undocumented outbreaks.

Linking human activity to spatial accumulation of microplastics along mangrove coasts

Microplastics (MP) in mangrove coasts are threating ecological health and seafood safety. However, quantitative evidence on the effects of different coastal human activities on microplastic accumulation in mangrove sediments is lacking, thereby impeding the policy development of evidence-based waste management. In this study, continuous geographical sampling (N = 50) was applied to collect sediments from the largest mangrove coast, namely the Leizhou Peninsula in China. Similar worldwide research data (16 mangrove coasts) were collected from the Science Citation Index Expanded (SCIE) database of the Web of Science. The connections between human drivers and microplastic accumulation were evaluated by spatial comparison, multi-correspondence analysis, and multiple differences analysis. The microplastic abundance fluctuated widely along the mangrove coasts (average value was 51.24, ranged from 6.40 to 255.57 items·kg−1 dry weight; coefficient of variation = 97%) with a globally lower-middle concentration in sediments of the Leizhou Peninsula. Densely populated urban residents and the floating population of tourists largely contributed to the high abundance of microplastics in mangrove sediments, of which large-sized (1–5 mm) white foams were the dominant type. Although suburbs had less crowds, both onshore and offshore fishery production could cause high accumulation of microplastics in neighboring mangrove coasts, which were characterized by small-sized (<1 mm) fragments with fresh color. Small microplastics (80%) with fresh color (44%) were dominant. Weathering may break down more toxic particles in urban areas neighboring mangrove coasts. Larger mangrove patches could partly block ocean-based microplastics; however, coasts surrounded by more geographical barriers had intensified pollutant accumulation. It was suggested that foam packaging of commodities for urban residents and tourists in popular tourism areas should be reduced and restrictions of fishery waste plastics are needed along shores with mangroves, especially in coasts surrounded by more geographic barriers.

Bodies need yoga? No plastic surgery! Naturalistic versus instrumental bodies among professions in the Danish healthcare field

What characterises the similarities and differences in body investments among professions in the Danish healthcare field? This can be important when healthcare professionals relate to each other and to bodies of patients regarding differences in class, gender, age, and disease group. The study takes inspiration from Bourdieu’s sociology and the concept of health capital. We ask whether health capital can explain distinctions in the empirical data? We used the explorative–descriptive method multicorrespondence analysis (MCA) on data from 440 respondents gathered through an online survey. The respondents range from chief physicians to healthcare students. MCA makes it possible to condense frequency statistics and examine patterns of body investments through an examination of variables such as food intake, exercise, and medicine use that are then further related to supplementary variables such as social class and age. The analysis supports the concept of health capital and shows that some groups in the Danish healthcare field have a more instrumental body perception, such as the young age group (e.g. open to surgery, consume medicine, high exercise), while the upper-middle-class and older respondents have a more naturalistic perception of the body (e.g. read often, eat vegetables, practice yoga).

Multi-correspondence analysis of the status and related factors of chronic diseases multimorbidity in middle-aged and elderly people in China in 2009

To study the status of chronic disease multimorbidity of middle-aged and elderly people over 45 years old in China, and to analyze the influencing factors. Using the data of the"China Health and Nutrition Survey(CHNS)", this survey has been carried out in 10 rounds nationwide since 1989. Multi-stage cluster random sampling was used to investigate rural and urban areas in 9 provinces and cities in China. In this study, the general socio-demographic characteristics, disease history, living habits and other information in the 2009 database were used to analyze the subjects who were collected blood samples. The prevalence of 8 common chronic diseases and multimorbidity such as hypertension, mixed-hyperlipidemia and hyperuricemia were described respectively. The Venn diagram in the R software package was used to calculate the multimorbidity of the disease. The χ~2 test and multiple correspondence analysis were used to explore the influencing factors of chronic disease multimorbidity in the middle-aged and elderly people in China. Among the 5316 subjects, the highest prevalence among 8 chronic diseases was hypertension(2143, 40. 3%). The lowest prevalence was myocardial infarction(87, 1. 4%). There were all together 1498 patients who had two kinds or more than two kinds of chronic diseases with a detection rate of 28. 18%. The most common multimorbidity were hypertension+hyperuricemia(199, 13. 30%), followed by hypertension + mixed-hyperlipidemia(191, 12. 77%). Hypertension+hyperuricemia+mixed-hyperlipidemia was the most common combination of the three chronic diseases(103, 6. 89%). Age(Kendull=0. 158, P&lt;0. 001), region(χ~2=30. 129, P&lt;0. 001), BMI(Kendull=0. 344, P&lt;0. 001) and marital status(χ~2=21. 923, P&lt;0. 001) were associated with the number of multimorbidities. Correspondence analysis showed that subjects aged 65 to 74, living in cities and sleeping less than 7 hours were more likely to have multimorbidity. The prevalence of chronic disease multimorbidity among middle-aged and elderly residents in China is high. Older age, unmarried, overweight and obesity, too little or too much sleep may increase the risk of multimorbidity.

Towards an objective assessment of climate multi-model ensembles – a case study: the Senegalo-Mauritanian upwelling region

Abstract. Climate simulations require very complex numerical models. Unfortunately, they typically present biases due to parameterizations, choices of numerical schemes, and the complexity of many physical processes. Beyond improving the models themselves, a way to improve the performance of the modeled climate is to consider multi-model combinations. In the present study, we propose a method to select the models that yield a multi-model ensemble combination that efficiently reproduces target features of the observations. We used a neural classifier (self-organizing maps), associated with a multi-correspondence analysis to identify the models that best represent some target climate property. We can thereby determine an efficient multi-model ensemble. We illustrated the methodology with results focusing on the mean sea surface temperature seasonal cycle in the Senegalo-Mauritanian region. We compared 47 CMIP5 model configurations to available observations. The method allows us to identify a subset of CMIP5 models able to form an efficient multi-model ensemble. The future decrease in the Senegalo-Mauritanian upwelling proposed in recent studies is then revisited using this multi-model selection.

Behavioural Differences among Educated Young Consumers in the Czech Republic: The Case of Organic Cheese Consumption

This study aims to identify behavioural differences between consumers who are driven by different sets of attitudes toward information related to food in their purchasing, including consumers who display rational behaviour. Although this study investigated “spill over” changes in attitude or behaviour (e.g. the use of information about food ingredients), the survey data were used to calculate the probability of behavioural differences between consumers. The survey involving 909 respondents from chosen Czech universities. Data were analysed by Multi Correspondence Analysis to investigate the association between several attitudes towards information on labelling, and consumer behaviour. In the second step, the ordered probit model probability of consumer behaviour was processed. There are two clusters of consumers: The first consumer segment was called “rationality involvement consumer”. They have a certain tendency to need to know what a product contains, the nutritional value of a product, and how to maintain the biological value of a product. The second consumer segment was called “non-rationality involvement consumer”. They have a certain tendency to not need to know what a product contains, the nutrition value of product, or how to maintain the biological value of a product.

Epidemiology of human exposure to rabies in Nunavik: incidence, the role of dog bites and their context, and victim profiles

BackgroundIn Nunavik, Arctic rabies is still endemic due to a spillover from wildlife to dogs. The prevention of human exposure and the management of potential exposure is a significant public health concern in this region.MethodsThis study retrospectively describes cases of potential exposure to rabies in humans as reported to the Nunavik Public Health Board through their registry of reported cases. We used multi-correspondence analysis as well as univariable and multivariable regression models to test for differences between children and adults in reported cases, and to examine the contexts of exposure to dogs and dog attacks.ResultsFrom 2008 to 2017, 320 cases of potential exposure to rabies were reported, 92% of which were linked to dogs. The annual incidence rate was 2.5 per 1000 people. The incidence increased significantly during the study period, although the reasons for this are unclear. Fifteen cases of exposure were with rabid animals, mostly dogs (9 of 15). No human cases of rabies occurred thanks to adequate medical case management. Two specific profiles for potential exposure to rabies were identified based on age and gender. The first was children (< 15 y/o), male or female, who were more likely to be exposed through playing with dogs and were more often injured in the head and/or neck. The second was young male adults (aged 15 to 34 y/o), who were more involved with wildlife than other age groups and mostly injured in the upper limbs and as a result of a reaction by the animal.ConclusionRabies is a real public health threat in Nunavik. Potential human exposure needs to be prevented, and prevention measures should be tailored to the two risk profiles identified based on age, gender and animal species involved.

Revisiting the Trivers-Willard theory on birth sex ratio bias: Role of paternal condition in a Malagasy primate.

Within current theories on potential adaptive manipulation of offspring sex ratio, giving birth to a male or to a female is assumed to depend on the capacity of the mother to invest in offspring to maximize her fitness. The active role of the father in sex ratio bias at birth has been neglected until recently. The human sex ratio at birth is biased towards sons, although in occidental populations, the ratio has decreased regularly for 30 years and could be the consequence of the adverse effects of environmental chemicals on male hormones. In a Malagasy primate, the lesser mouse lemur, the potential effect of paternal testosterone levels on sex ratio bias at birth was tested on 130 litters (278 babies) produced in 52 mixed-sex groups. For each group, social dominance among males was characterized based on aggressive interactions and sexual behaviours. Using a multi correspondence analysis, high testosterone levels in grouped males, particularly those of the dominant male, were significantly correlated with more infants produced in male-biased litters, independent of the female condition. According to these results, predictions for sex ratio bias towards one sex or the other in mouse lemurs were discussed considering the influence of both parents.

How do universities implement the Health Promoting University concept?

The Health Promoting University (HPU) concept encourages universities to incorporate health into the university culture, processes and policies in an effort to promote the health of the university community. Universities worldwide have adopted the approach and a framework for action has been developed to guide universities to become a HPU. However, information on how universities translate the framework into actions is scarce. This study explored the way in which 54 universities from 25 countries across the world implemented the HPU framework. An online questionnaire was used to assess the action areas and items of work addressed by the universities and to determine their adherence to the components of the HPU framework: use of the whole systems approach; multiservice collaboration; recognition by the university authorities; funding availability; membership of a HPU network and evaluation of the initiative. The results showed that these components were addressed by most universities. A Multi Correspondence and cluster analysis identified four types of universities based on the implementation of the components: 'emerging' HPUs that are not recognized by the university authorities and tend to not apply the whole systems approach or evaluation of the initiative, and 'established' HPUs that are recognized by the authorities, apply the whole systems approach and evaluate the initiative but that differ with regard to funding and membership of a HPU network. These results demonstrate that universities implement the HPU framework for action differently in order to become a Health Promoting University.

Cultural Openness as an Emerging Form of Cultural Capital in Contemporary France

This article explores the changing pattern of cultural privilege in contemporary France. Using French data on cultural practices, including variables on ‘highbrow’ culture, mass culture and cosmopolitan culture, we apply a multi-correspondence analysis (MCA). The findings first show that cultural privilege among French social and educational elites remains primarily a matter of cultural capital endowment, with a structuring contrast between ‘legitimate’ and ‘mass’ culture. The MCA also shows an additional divide between local and global culture underpinned by a strong age gradient. Yet the emergence of a changing pattern of cultural privilege among the youngest cohorts does not imply any clear reduction in cultural inequalities. Rather, it suggests a growing cultural distinctiveness of French elites. Finally, these tendencies should not be easily extrapolated to other contexts as they reflect strong French specificities related to the evolution of social and educational structures during the second half of the 20th century.

Indian Software Capital: Sociography of a New Entrepreneurial Elite

The aim of this article is to study Indian software capital considered as the nexus of IT companies and their top managers. For this purpose, we have conducted a Multi Correspondence Analysis of a set of data collected for the first hundred IT firms listed on the National Stock Exchange in Mumbai in 2012. We address three main research questions. Firstly, what is the socio-demographic profile of these entrepreneurs, secondly, what is the shareholding pattern of these companies and their position on the IT market and, lastly, what are the differences between these IT companies and traditional business groups? Although the managers are highly educated and share the same technical culture, the results show a clear division of software capital in 3 clusters. First, the Multinational companies, which are the oldest firms and the biggest in terms of manpower and global revenue, are opposed to a second group of IT companies founded or managed by executives who are mainly graduates, and whose activities are oriented towards the domestic market; lastly, a third group of companies run by highly-qualified managers (with MBAs) combined characters of clusters 1 and 2, but are smaller companies more dependent on the stock-market. In the software industry, the family-business model, which is not specifically related to the merchant high castes, is well represented, except for the Indian MNCs, but with some qualifications.

Interrelationships in the Variability of Root Canal Anatomy among the Permanent Teeth: A Full-Mouth Approach by Cone-Beam CT.

ObjectivesIn endodontic practice, clinicians should be aware of possible root canal anatomic variations. The aim of this study was to assess using CBCT acquisitions regarding whether one root canal anatomy of a tooth is associated with a specific anatomy of another tooth.MethodsA total of 106 CBCT acquisitions were obtained using a CBCT scanner with 200μm voxel size. Numbers of roots and canals of the entire dentition were described. Bivariate analyses and logistic regressions were conducted to explore root canal anatomy on one tooth according to age, gender, jaw, side and the others teeth. Multiple correspondence analysis (MCA) was performed to correlate the different numbers of canals profiles.ResultsA total of 2424 teeth were analyzed. Independently from the other variables, the presence of an additional root canal on a mandibular incisor increases the risk of having an additional root canal on a mandibular premolar (OR [95%] 3.7 [1.0;13.2]). The mandibular molar variability increases in women compared to men (OR [95%] 0.4 [0.1; 0.9]). MCA showed correspondence between 2-canals maxillary incisor and canines and 5-canals maxillary molars, and some correlation between additional canal on maxillary and mandibular premolars.ConclusionsAlthough CBCT examinations are conducted in the first intention of making a diagnosis or prognostic evaluation, medium FOV acquisitions could be used as an initial database thus furnishing preliminary evaluations and information. In endodontic practice, clinicians should be aware of possible root canal anatomic variations. The visualization of all canals is considered essential in endodontic therapy. The use of multi-correspondence analysis for statistics in endodontic research is a new approach as a prognostic tool.

An Analysis for the Questionnaire Investigation about the High School Teachers Workload

High School teachers in Japan are sending very busy days on their daily works including teaching, support for the club activities and deskwork. Among them, they share a lot of time for managing the club actives of students compared with other countries. In that area, professionals can make instruction much better than teachers for the special sports like Judo and Kendo (Japanese fencing) etc. School Social Worker can coordinate the professionals out of school and can help teachers by decreasing their burden on that area. There are few related papers concerning the support of club activities by utilizing the professionals outside. In this paper, a questionnaire investigation is executed to the four High Schools at Iwate Prefecture in Japan in order to clarify their current condition and their consciousness, and to seek the possibility of utilizing school social worker for their support. Fundamental statistical analysis, Factor Analysis and Multi Correspondence Analysis are performed. Some interesting and instructive results were obtained.

Multivariate Analysis Results on a Questionnaire Investigation for Rare Sugars

&lt;p&gt;The Rare Sugars exist naturally and have many kinds (more than 50). They have good effect for health such as prevention of increasing the blood-sugar level after eating, suppression of fat accumulation, suppression of increasing the blood pressure, and anti-oxidative effect etc. It is in the spotlight for many people especially for those who are in the metabolic syndrome. There are few related papers concerning the marketing research and its utilization of this matter. In this paper, a questionnaire investigation is executed in order to clarify consumers’ current condition and their consciousness, and to seek the possibility of utilizing the Rare Sugars. Such multivariate analysis as Factor Analysis and Multi Correspondence Analysis are executed based on that. Some interesting and instructive results were obtained.&lt;/p&gt;

Measuring multimorbidity in a working population: the effect on incident sickness absence.

PurposeMultimorbidity research typically focuses on chronic and common diseases in patient and/or older populations. We propose a multidimensional multimorbidity score (MDMS) which incorporates chronic conditions, symptoms, and health behaviors for use in younger, presumably healthier, working populations.MethodsCross-sectional study of 372,370 Spanish workers who underwent a standardized medical evaluation in 2006. We computed a MDMS (range 0–100) based on the sex-specific results of a multicorrespondence analysis (MCA). We then used Cox regression models to assess the predictive validity of this MDMS on incident sickness absence (SA) episodes.ResultsTwo dimensions in the MCA explained about 80 % of the variability in both sexes: (1) chronic cardiovascular conditions and health behaviors, and (2) pain symptoms, in addition to sleep disturbances in women. More men than women had at least one condition (40 vs 15 %) and two or more (i.e., multimorbidity) (12 vs 2 %). The MDMS among those with multimorbidity ranged from 16.8 (SD 2.4) to 51.7 (SD 9.9) in men and 18.5 (SD 5.8) to 43.8 (SD 7.8) in women. We found that the greater the number of health conditions, the higher the risk of SA. A higher MDMS was also a risk factor for incident SA, even after adjusting for prior SA and other covariates. In women, this trend was less evident.ConclusionsA score incorporating chronic health conditions, behaviors, and symptoms provides a more holistic approach to multimorbidity and may be useful for defining health status in working populations and for predicting key occupational outcomes.Electronic supplementary materialThe online version of this article (doi:10.1007/s00420-015-1104-4) contains supplementary material, which is available to authorized users.

Analysis of accidents in biogas production and upgrading

In the last 10 years biogas production tripled, resulting in an increasing number of related facilities. The present study addresses safety issues of such activity on the basis of past accident analysis. A database of accidents related to biogas supply chain was created and data on 169 accidents were collected from different literature sources. Trends, causes of accidents, scenarios, and consequences were analysed also using Multi Correspondence Analysis to obtain correlations between causes of accidents and scenarios. The study showed that almost 12% of the past accident analysed can be classified as major accidents. The number of accidents is growing faster than biogas production. A risk assessment was carried, based on the analysis of accident frequencies and consequences. A non-negligible risk profile, typical of ALARP zone, was obtained for this industrial sector, revealing an emerging risk issue. The main lessons learnt are the need of developing and adopting specific safety standards and of improving the safety culture and risk awareness in the biogas production sector.

Archetypal personalities of software engineers and their work preferences: a new perspective for empirical studies

As the area of Software Engineering (SE) matures the role of human factors in software development is commonly recognized as important. Increasingly we see empirical studies that investigate the connection between, for example, personalities and preferences, attitudes or performances of software engineers. Statistical analysis holds a key role by providing the means for uncovering associations between various facets of human factors and behavioral effects on projects and outcomes. Traditional statistical techniques tend to explore and interpret the multidimensional personality and behavioral data from an "average-point" perspective, targeting central trends. This paper introduces a methodology with statistical tools that can provide a new and different perspective for this type of SE data. It seeks the boundaries of a psychometric dataset and discovers reference or "benchmark" personalities, the archetypal personalities. Then, the method examines the placement of all individuals in the dataset in relation to the archetypes. Furthermore, the SE preference characteristics, or generally, any other types of behavioral SE data, are analyzed with respect to the archetypes. As a case to exemplify the methodology we analyze personality and project preference data from 276 master level SE students and compare to previous "average-point" statistical analysis of the same data. We also discuss how Archetypal Analysis, the heart of the proposed methodology, combined with multi-correspondence analysis might be of general use in empirical SE.