SciELO - Scientific Electronic Library Online

 
vol.27 issue2Electricity prices forecasting through artificial neural networksContribution of electre tri industrial risk classification author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • Have no similar articlesSimilars in SciELO

Share


Investigação Operacional

Print version ISSN 0874-5161

Abstract

MOREIRA, António Carrizo. Comparing the principal components analysis to the CATPCA in the passenger’s satisfaction evaluation of a passenger air carrier. Inv. Op. [online]. 2007, vol.27, n.2, pp.165-178. ISSN 0874-5161.

The main purpose of this article is to confront the results of the Principal Components Analysis (PCA) with the results of the Categorical Principal Components Analysis (CATPCA), in order to assess the feasibility of both techniques. The satisfaction of the service provided by an air passenger carrier was used for that purpose where 250 passengers were interviewed. The conclusions are clear: on the one hand, it is advisable to use the PCA because as a result three components are obtained that explain 70,23% of the total variance of satisfaction explained whereas the CATPCA has a result only one component that explains 60% of the total variance explained of the air carrier passenger satisfaction. On the other hand, the PCA violates two main principles: it uses qualitative variables and those variables do not follow a normal distribution, which is not at stake with CATPCA. The trade off of the two techniques favors PCA vis-à-vis CATPCA due to both the three dimensions obtained and the total variance explained.

Keywords : Principal Components Analysis; Categorical Principal Components Analysis; Satisfaction in Services.

        · abstract in Portuguese     · text in Portuguese     · Portuguese ( pdf )