ABSTRACT
As the first of the two goals of this chapter, the Likert scaling method and three other well-known scaling methods (Thurstone, Guttman, and Semantic Differential) are described as a way to provide some historical background on how Likert scaling was viewed before and after it came to be the dominant method of quantifying attributes or constructs in survey research. Rating items are so commonly used in survey research that the majority of the content of this chapter is devoted to discussing issues involved in deciding how to structure rating scales for rating items, including determining how many options to offer and how to label the options. We rely on current consensus among survey researchers to provide our recommendations and a possible path to constructing response or rating scales. There are few hard and fast rules about what the best response options are. However, 15 points of advice are offered to help when deciding on the response scale.
