4
Measurement scales
This chapter considers the problem of measurement scales, which is an essential issue in every statistical analysis, in particular also in the analysis of customer satisfaction surveys. We begin by presenting the problem of scale construction and the different types of scales commonly used in customer satisfaction surveys, that is, the nominal, ordinal, interval and ratio scales. Then we consider the problem of obtaining an interval scale from an ordinal one; the resulting scale has a more informative characterization. In this context scaling methods are taken into account to obtain a unifying scale for a set of homogeneous items in a questionnaire.
4.1 Scale construction
Let 
be the responses given by n individuals to a generic item in a questionnaire. Each response is chosen from set 
of alternatives – work positions, gender, the date on which a service relationship began, the number of contacts that have taken place over a specified period of time, or degrees of agreement or satisfaction, the latter commonly measured on five- or seven-point Likert scales, where the minimum integer value represents ‘complete dissatisfaction’ or ‘completely disagree’ while the maximum value stands for ‘complete satisfaction’ or ‘completely ...
