AN APPLICATION OF NONPARAMETRIC STATISTICAL TECHNIQUES IN DETERMINING THE MOST PREFERED NETWORK PROVIDERS IN NIGERIA

Abstract

Over the years, the improvement made by the different network providers in Nigeria has not satisfactorily met the customers' communication needs as there is a barrage of complaints of poor service in different network providers. As a result, the need to evaluate the quality of GSM received signal strength of the different mobile network providers, this study aimed at assessing the most essential and preferred network provider in the country. The data used were collected via student’s opinions at the case study, Kaduna Polytechnic. The Mann – Whitney U test, Kruskal Wallis test, Chi-Square tests of independence, and Correlation analysis were applied for the analysis. The result from Mann – Whitney U test shows that there is no significant difference in the average number of usages of bandwidth types. Thus, from the rank’s column, it was observed that Gigabytes has the highest average rank as an indication to the majority of the population subscribed to Gigabytes than megabytes, the Kruskal Wallis U Test result, indicates that there is a significant difference in populations’ preference of network provider for the bandwidth’s subscription. Similarly, the average rank column of the Kruskal Wallis U Test shows the MTN with the highest mean rank, placed as the most preferred network by the population of the study. Correlation and chi-square test resulted in a positive significant
relationship among the network providers at the national level.