Customer-Satisfaction Measurement

Abstract

A study of more than 400 restaurant patrons allowed a comparison of the validity and reliability of six approaches to measuring customer satisfaction. Four different questionnaires asked respondents about their expectations and the restaurant's performance in conjunction with those expectations. Some of the questionnaires also asked diners to rate the relative importance of certain restaurant attributes. The results indicated that an approach based on measuring “performance only” achieved the most validity and reliability. In contrast, attempting to measure customer satisfaction as a difference score between customers' prior expectations and the restaurant's perceived performance was the least reliable measure of satisfaction. Furthermore, the study identified substantial operational difficulties in administering gap-based surveys in a restaurant-notably, the problem of having to test for expectations before the meal and then test again for performance afterward. The findings of the study also suggest that weighting a performance or disconfirmation score by its importance does not bring about any substantial increase in predictive ability, although the information thus derived may be otherwise useful.