What is the ideal number of points on an evaluation scale?

The use of scales facilitates the process of collecting customer data, allowing information about the order or magnitude of the variables to be measured. The scales must be one-dimensional, that is, involve a single dimension which, in practical terms, means that the scale must only measure one attribute.

But what should be the size or number of points on the scale?

All measuring instruments must have a sufficient resolution. This principle is applied to X-ray machines, cameras, but also to the scales used in the questionnaires. Otherwise, it will not be possible to find statistically significant differences between different segments, which degrades the accuracy of the results and limits the methods of data analysis that can be used.

In this way, a binary scale (0/1) is completely excluded as it does not allow any discrimination. But the same goes for scales with five and seven points, which, however, are often used in customer experience studies.

Indeed, in a relatively competitive market where it is easy to find alternatives, dissatisfied customers leave the company. As a result, customer responses to a satisfaction survey do not have a normal distribution. Thus, on a scale of 1 to 5, values ​​1, 2 and even 3 will have a much lower frequency than values ​​4 and 5, which poses the problem of discrimination since there will be a large concentration of responses in these two values. The seven-point scale reduces this problem but does not solve it.

In addition, these scales have an odd number of points. On these scales, the intermediate point can be used by respondents who do not want to take a position or also by those who have no opinion, substitute the answer "I don't know" or "I have no experience" for the midpoint of the scale. On a scale with an even number of points, the respondent has to take a stand, so the answer is more accurate.

In theory, the greater the number of points on the scale, the better it would be, since it would allow greater discrimination. However, from a certain number of points, the interviewee has difficulty discriminating, making the questionnaire difficult to answer.

Bearing in mind, the aforementioned 10-point scale is the one that best answers the questions raised, which is why it is used in the main international studies of customer satisfaction and experience. This is also the scale favored by Qmetrics.