Alternative Indexes for Monitoring Customer Perceptions of Service Quality: A Comparative Evaluation in a Retail Context

Abstract

Customer satisfaction and service quality measures obtained through consumer surveys invariably have skewed distributions. As such, researchers have questioned the appropriateness of the popular approach of using the mean rating to summarize such data. However, no de tailed study on this topic has yet been conducted. In two independent studies, the relative validity of the various indexes that can be used to summarize consumers’ service quality ratings (e.g., mean, median, mode, kurtosis, skewness, top/bottom-tail percentiles) are examined. In Study 1, using typical commercial survey data from a fast-food/ convenience retail chain, both the mean and top-box per centiles are found to be the best indicators of service quality, based on their correlation with customer-driven business performance measures. In Study 2, the results are further confirmed by an extensive simulation that varies factors such as the shape of the underlying distribution of customer ratings and the strength of the relationship be tween customer ratings and business performance mea sures. The article concludes with a discussion of the findings and implications for future research.