Todd Hollander Market Research

Between my clients at Todd Hollander Market Research and my students in the Priniciples of Marketing Research course, I am constantly asked “What is the best scale to use for customer satisfaction surveys?” The short answer is “There isn’t one.”

The question about which scale is best usually refers to noncomparative scales. In noncomparative scaling (the most common scaling technique for market research), each object is rated on its own scale independently of the others in the stimulus set. For example, respondents may be asked to rate a product attribute on a 1 to 7 scale where a rating of “1″ = “Poor” and a rating of “7″ = “Excellent.” This is the type of rating that begs the question about which scale is best. The question is usually posed along the lines of “Is a 7-point scale better than a 5-point scale?” or “Is a 10-point scale better than a 7-point scale?”

In reality, there simply isn’t one rating scale (5-point, 7-point, etc.) that is preferred over all others for satisfaction surveys (or most other types of surveys). Why not? Because the ideal scale depends on a combination of:

• The question being asked
• How knowledgeable respondents are about the subject of the question
• The number of completed surveys anticipated
• How the results will be analyzed
• How the results will be used in decision-making

Using these factors to construct effective rating scales takes considerable skill and experience. It is also important to know how to transition respondents from one rating scale (e.g. 5-point) to another (e.g. 10-point) without causing confusion. When inappropriate scales are used, or when multiple scales are used in a single survey, respondents may perceive the survey as illogical and/or unprofessional.

The bottom line: There is no “best” rating scale for satisfaction surveys. Knowing which scales to use and how to use them effectively takes a considerable amount of experience.