Increasing promoters in the residential broadband service industry: Relationship between customer satisfaction and loyalty using ordinal logistic regression

详细信息   

• 作者：Lartey ; Franklin M.
• 学历：Doctor
• 年：2015
• 关键词：Social sciences ; Communication and the arts ; Custome
• 导师：Hargiss,Kathleen
• 毕业院校：Capella University
• Department：School of Business and Technology
• 专业：Marketing;Management;Information science
• ISBN：9781321561555
• CBH：3682580
• Country：USA
• 语种：English
• FileSize：908398
• Pages：141

文摘

Some studies related to customer satisfaction in the service industry use the SERVQUAL instrument developed by Parasuraman,Zeithaml,and Berry (1991). Similarly,some studies on customer loyalty use the Net Promoter System(TM) (NPS? ) developed by Reichheld (2003). Even though many researchers acknowledge the existence of a relationship between customer satisfaction and loyalty,there is currently no empirical evidence demonstrating such relationship between SERVQUAL and NPS. In that regard,the research problem investigated by this study consisted of identifying the antecedents of customer satisfaction as measured by SERVQUAL that influence customer loyalty as measured by the Net Promoter System(TM) in the context of the residential broadband service industry. To that effect,208 broadband customers in the U.S. Midwest region were surveyed online,out of which 44.2% were males,and 55.8% females. A confirmatory factor analysis validated the latent structure of collected data in regards to SERVQUAL,indicating that it was appropriate to use SERVQUAL's five factors as predictors in the study. An ordinal logistic regression model built on the collected data confirmed the existence of a statistically significant relationship between Empathy (one of the five factors of SERVQUAL) and the customer loyalty status (Wald χ 2 (1) = 4.663,p = .031,95% CI [.010,.212]),suggesting that increasing empathy by one point increased the odds of becoming a promoter by 11.73%. The Pearson test of model fit showed that the overall model was predictive of the relationship,χ2 (401) = 409.022,p = .380. This final model was validated using the Akaike information criterion (AIC),a single sample cross-validation model.