Customer Satisfaction Scale
Satisfaction scales are used to measure customer approval on written or Internet surveys. There are several different ways to set up scales, each with their own pros and cons. It is important to choose an appropriate satisfaction scale for the measurement that is being made.
What is a Satisfaction Scale?
Psychologist Rensis Likert invented the basis of the most common satisfaction scale. Sometimes called Likert scales, these questions offer respondents the opportunity to agree or disagree with a given statement. Not all Likert scales are the same because there are advantages and disadvantages to setting them up in various ways.
When asked to respond to a statement, a satisfaction scale may use
either a digital or a semantic measure. A digital scale uses numbers, usually with lower numbers representing a less positive response and higher numbers indicating a more positive response. A semantic scale uses opposite statements on either end of the scale, such as “strongly agree” versus “strongly disagree.” The midpoint of both scales is usually neutral.
Some people use semantic simply to denote the use of words over numbers. However, a semantic differential question is one where two opposite choices are available. Many Likert scale questionnaires use semantic differentials but not all of them do.
In order to obtain a usable result, at least three options must be available. A typical Likert scale uses five measures and the numbers one through five or a range from “strongly agree” to “strongly disagree” with a neutral position in the middle. Some scales use both numbers and semantic terms to decrease confusion.
How to Set Up a Satisfaction Scale
The range of a scale can be important. Too many options leads to a less meaningful distinction between choices. Too few options may not capture the true opinions of users. Between four and ten options are typical, with five being the most common choice.
Four-point measuring scales are useful in some cases, though. A four-point measuring scale eliminates the neutral, central point and forces survey takers into stating an opinion one way or the other. However, one disadvantage of these “even scales” is that they make it hard to measure evolving customer satisfaction.
Some survey questions are served well by a “don’t know” or “doesn’t concern me” option. These options prevent introducing bias into the survey by forcing customers to state an opinion when the question doesn’t apply to them or is not important to them. These options can take the place of a central, neutral button on a five-point scale.
When Not to Use a Satisfaction Scale
Rating scales are used to measure the respondent’s perception. Questions that have quantitative or true/false answers are best measured in a slightly different way. Although it is often possible to formulate the question for the rating scale setup, the information obtained will often be more useful if it is not.
An example would be when a business wants to know whether people are being greeted at the door. This could be done using a satisfaction scale with a statement like “I was greeted at the door,” and answers “agree,” “disagree,” and “don’t remember.” However, a direction question might work better, here. “Were you greeted at the door?” with a yes/no option for answering makes more sense.
Although it makes more sense to choose a question format, no information is lost by using a three-point satisfaction scale. However, since it is a yes/no question, information would be lost by using a wider scale range. It would be nearly impossible to determine what a person meant when they chose “strongly agree” over “agree,” for example.
The same argument applies even more strongly to quantitative measures. An example of this type of question would be how long a person waited to be seated. An opinion on whether the respondents felt that they waited too long might be useful, but it is not actually answering the same question.