Article Sidebar
Published:
Feb 28, 2025Keywords:
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Main Article Content
Weighted Least Squares Regression on User perception on Internet services by Internet Service Providers (ISP)
Mark Ofori Nketia
Lord Emmanuel Yamoah
Abstract
This study investigates the relationship between user perception and the perceived quality of Internet Service Providers (ISPs) using Weighted Least Squares Regression (WLS). With the rapid advancement of digital technologies and the increasing importance of internet services, understanding customer satisfaction and its determinants is crucial for ISPs. The research employs a quantitative approach, collecting data from 181 students at Kwame Nkrumah University of Science and Technology (KNUST) through a structured questionnaire. The study examines how speed, reliability, and customer assistance influence user perception and, consequently, the overall evaluation of ISP services.
The findings reveal that user perception, weighted by user ratings, has a substantial and statistically significant impact on the perceived quality of ISPs. The high R-Square value (0.807) and significant F- and t-values confirm that user perception is a substantial predictor of ISP service evaluations. Descriptive statistics indicate that respondents generally have a favorable opinion of ISPs, though there is variability in individual assessments. The study also highlights the importance of demographic factors such as gender, age, and educational qualification in shaping user perception, while the profession was insignificant.
Theoretical implications support models like SERVQUAL, emphasising the centrality of customer perception in service evaluation. The study suggests that ISPs should invest in customer care, quality improvements, and data-driven decision-making to enhance customer satisfaction and loyalty. Socially, the findings underscore the importance of reliable internet access for digital inclusion, particularly in underserved populations. The study concludes with recommendations for future research, including using longitudinal data and machine learning techniques to explore user perception and satisfaction with ISPs further.
