Validação do modelo de satisfação ACSI modificado: um exame empírico com equações estruturais

• Ano de defesa: 2004
• Autor(a) principal: Andreia Cassia de Moura
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Telefonia móvel; Relacionamento entre variáveis; Modelagem de equações estruturais; Modelo de satisfação ACSI; Validação de modelo; Administração; Satisfação do consumidor; Marketing
• Link de acesso: http://hdl.handle.net/1843/BUBD-99PHZP

This work aims to validate a satisfaction model, the American Customer Satisfaction Index - ACSI on the mobile telecommunication sector in Brazil. The ACSI model indicates the relationship between Satisfaction and its antecedents (Expectations, Quality and Value) and also consequents (Complaints and Loyalty). In this research, the original ACSI model was modified due the inclusion of a construct: Convenience. The research had a conclusivedescriptive character involving two phases: exploratory and descriptive. The exploratory one was conducted in order to adapt the model to the researched sector; the technique used was focus group. Later, the research entered in a quantitative phase in order to validate the model. To achieve this objective, a sample of 606 users of cell phone was interviewed. Research respondents are clients of the three mobile telecommunication companies of Minas Gerais, inBrazil. Data statistical treatment was supported by multivariate techniques, including Structural Equations Modeling to analyze plausibility of new proposed model. The results indicated proposed model validity with few changes suggested by both qualitative andquantitative analysis. The relationships of following variables were supported by data analysis: Quality Value, Quality Satisfaction, Value Satisfaction and Satisfaction Loyalty. The relationship between Convenience and correlated constructs wasnt supported bydata analysis. Finally, theoretical and managerial implications are presented, as well as inherent limitations of the study.