Principal Component Analysis of Customer Churns in Ghanaian Telecommunication Industry
Yarhands Dissou Arthur, Emmanuel Harris, Jonathan Annan

Abstract
This paper presents the major factors that influence clients in the telecommunication network to switch from one telecommunication network to another. Based on principal component analysis (PCA), correlation and descriptive statistical analysis of the factors believed to influence switching behavior of client and therefore clustered into principal component. The sampling test conducted using Bartlett’s test of sphericity and KMO measure of sampling adequacy (MSA) to ascertain the level relationship and the pattern between variables was found to be very significant with MSA (0.737).The result of the analysis found to havefive components displayed eigenvalues greater than 1. Therefore, the entire factor can be put into five principal components which accounted for 73.254% of the total variance. Statistically there was strong interrelationship among the variables in the (PCA) with p-value <0.05 .The principal component are ethical and service failure, knowledge gab, innovation social influence and inconvenience

Full Text: PDF

Copyright © 2014: The Brooklyn Research and Publishing Institute. All Rights Reserved.
Brooklyn, NY 11210, United States