Design e semântica: investigação de técnicas estatísticas para auxílio no projeto de produto

• Ano de defesa: 2015
• Autor(a) principal: Holdschip, Rodrigo [UNESP]
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Desenho industrial; Emoções; Estética; Diferencial semantico; Análise fatorial; Industrial design
• Link de acesso: http://hdl.handle.net/11449/136663; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/07-03-2016/000859635.pdf

After nearly half a century under the influence of functionalism whose aesthetic principle was focused on the object (form follows function), the discussion of the post-modern design is focused on the user and the semantic dimension of the product (form follows meaning). Currently, the semantic differential is widely used for the study of human perception towards products. It's application assumes the existence of an underlying structure to the semantic universe that can be explained by a small number of independent concepts that together make up wat is called the semantic space. The scope of this space has been reported in literature as the first step in many studies interested in obtaining subjective or emotional responses. For its definition, the data obtained from the use of semantic differential should be analyzed preferably by statistical methods, among which, the factor analysis is the technique often adopted. The extraction process of the factors in the factor analysis can be done by two methods widely used and know as: (1) common factor analysis and (2) principal component analysis. Althrough these methods are based on similar calculations there is still no consensus on which method is the most appropriate. In this context, in which emotions come to play an important role in the design of new products, this research investigated the influence of methods for factors extraction, used in the factor analysis for the definition of the semantic space. The data collected by five case studies covering different products were submitted to two independent factor analysis to compare the results and to provide practical differences between these two loading and most adopted factorial approaches. The results enable a conclusion that the principal component analysis is more suitable for the reduction and selection of suitable descriptors for the composition of semantic differential scales for the purpose of product evaluation