Halo in customer satisfaction measures: The role of purpose of rating, number of attributes and customer involvement

Abstract

Firms usually measure customer satisfaction on an attribute-by-attribute basis in order to identify and improve potential weaknesses, and to fortify their strengths in service delivery. However, research has shown that halo can threaten the interpretability of such data. Also, halo is particularly acute in satisfaction measurement of services with a high degree of ambiguous and credence attributes. This paper examines three halo-reducing methods developed in psychology and organizational behavior in the context of customer satisfaction. The perceived purpose of evaluation (evaluative vs developmental) and the number of attributes measured (few vs many) were examined in an experimental design, and the level of product involvement (low vs high) was examined using a quasi-experimental design. The data showed reduced halo when the respondents were presented with a developmental rather than evaluative purpose, when more rather than fewer attributes were measured, and when subjects were highly involved with the service.