A carregar...
Publicação
Requisitos para a qualidade de um executive information system
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 3.12 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
OS EIS pretendem ser a solução para o acesso dos gestores às bases de dados, por
intermédio de um interface especializado. Este estudo pretende testar a veracidade
dessa relação. Assim, foram seleccionados 107 gestores, com idades compreendidas
entre os 20 e os 63 anos, repartidos em 3 grupos: 2 grupos de 37 elementos com
experiência prévia nestes sistemas (grupo I e II) e outro com 33 elementos, sem contacto
com EIS. A experiência foi conduzida em 2 fases, uma com relatórios em conformidade
com as orientações da literatura (Modo OK) e outra com relatórios contrários às mesmas
(Modo NOK). Assim, ao grupo I foi proposta a sequência Problema-OK-NOK, ao grupo II
e III a sequência Problema-NOK-OK. Ao grupo III foi ainda dada formação específica
para a utilização do sistema, que consistiu em aceder via web ao modo NOK e ao modo
OK através de uma aplicação, também via web, que reunia as condições para a
experiência. Em sintonia com a literatura, foram consideradas como variáveis
dependentes a satisfação, quer com a compreensão do problema, quer com a solução
obtida. Como variáveis independentes, as características próprias de um EIS e a atitude
face aos sistemas informáticos. Foram feitos testes de diferença de médias de satisfação
entre os dois modos (teste t para amostras emparelhadas), bem como regressões
lineares múltiplas, No final, foi possível sustentar que o modo OK produziu maior
satisfação, mas não se encontraram provas convincentes da sua relação com as
características EIS. A análise dos dados, bem como das correlações entre as respostas,
sugere aliás que é a concepção gráfica, mais até do que a possibilidade da manipulação
OLAP, a responsável pela diferença de satisfação encontrada.
The EIS intend to be the solution for managers to access large databases, by means of a specific type of interface. This study wishes to find if we can in fact relate one to the other. Therefore, 107 managers were selected, aging from 20 to 63 years old, divided in 3 groups: 2 of this groups with 37 elements with previous experience with EIS (group I and II), and another with 33 with no experience in this field (group III). The test was conducted in 2 phases, one with reports taylored to the EIS features (mode OK), and another with reports made to be exactly the opposite (mode NOK). The group I was confronted with the sequence Problem-OK-NOK, the groups II and III with the sequence Problem-NOK-OK. Group III also took specific trainning for this purpose. This was mainly the access through the web to a set of reports in mode NOK and to mode OK through an web-based application specially selected to support the test. In accordance with the literature, satisfaction was considered the dependent variable, in what concerns the understanding of the problem and the solution adopted. As independent variables, the features of an EIS and the altitude towards computerized systems. A t-test for paired samples was applied to the satisfaction of both modes, as well as multiple linear regressions. In the end, proof was given that satisfaction improves with the mode OK (ElS-like), but the relationship between the EIS features and this improved satisfaction fell short. The analyses of the data suggests, however, that it is the graphic design, more so than the OLAP capabilities, that it is responsible for the increased satisfaction.
The EIS intend to be the solution for managers to access large databases, by means of a specific type of interface. This study wishes to find if we can in fact relate one to the other. Therefore, 107 managers were selected, aging from 20 to 63 years old, divided in 3 groups: 2 of this groups with 37 elements with previous experience with EIS (group I and II), and another with 33 with no experience in this field (group III). The test was conducted in 2 phases, one with reports taylored to the EIS features (mode OK), and another with reports made to be exactly the opposite (mode NOK). The group I was confronted with the sequence Problem-OK-NOK, the groups II and III with the sequence Problem-NOK-OK. Group III also took specific trainning for this purpose. This was mainly the access through the web to a set of reports in mode NOK and to mode OK through an web-based application specially selected to support the test. In accordance with the literature, satisfaction was considered the dependent variable, in what concerns the understanding of the problem and the solution adopted. As independent variables, the features of an EIS and the altitude towards computerized systems. A t-test for paired samples was applied to the satisfaction of both modes, as well as multiple linear regressions. In the end, proof was given that satisfaction improves with the mode OK (ElS-like), but the relationship between the EIS features and this improved satisfaction fell short. The analyses of the data suggests, however, that it is the graphic design, more so than the OLAP capabilities, that it is responsible for the increased satisfaction.
Descrição
Instituto Superior de Economia e Gestão.
Palavras-chave
Contexto Educativo
Citação
Carreiro, André Ricou Tavares, (2006)." Requisitos para a qualidade de um executive information system". Dissertação de Mestrado. Universidade de Lisboa, Instituto Superior de Economia e Gestão.