Doria, IsabelSousa, ÁureaBacelar-Nicolau, HelenaLe Calvé, Georges2014-02-052014-02-052013Doria, Isabel; Sousa, Áurea; Bacelar-Nicolau, Helena; Le Calvé; Georges (2013). "Comparison of Modal Variables Using Multivariate Analysis". In: Advances in Regression, Survival Analysis, Extreme Values, Markov Processes and Other Statistical Applications. Studies in Theoretical and Applied Statistics. J. Lita da Silva et al. (eds.) Berlin, Heidelberg: Springer-Verlag, pp 363-370. ISBN: 978-3-642-34903-4 (Print) 978-3-642-34904-1 (Online), DOI 10.1007/978-3-642-34904-1_38.978-3-642-34903-4 (Print)978-3-642-34904-1 (Online)DOI 10.1007/978-3-642-34904-1_38http://hdl.handle.net/10400.3/2711Copyright © Springer-Verlag Berlin Heidelberg 2013.Domiciliary palliative care satisfaction and quality were estimated by caregivers via five perception scales with partly ordered answering modalities. The perception scales were codified as symbolic modal variables and analyzed using two multivariate approaches based on complex (symbolic) data to compare modal variables. This study compares the outcomes of previous work by Doria (Representações euclidianas de dados: Uma abordagem para variáveis heterogéneas. Tese de doutoramento, Universidade de Lisboa, Lisboa, 2008), Doria et al. (Livro de Resumos da XI Conferencia Española de Biometria e Primer Encuentro Iberoamericano de Biometria (CEIB2007)101–102,2007) and Bacelar-Nicolau et al. (Revista Portuguesa de Filosofia 66(2):427–460,2010). In particular, it focuses on the differences and similarities of the results obtained with principal component analysis and ascendant hierarchical cluster analysis, directly applied to the similarity matrix SLC and to the generalized affinity matrix, adapted to the comparison of modal variables.engBiomatemáticaEstatísticaProbabilidadeHierarchical Cluster AnalysisSymbolic VariablesComparison of Modal Variables Using Multivariate Analysisbook part