When conducting a survey, it’s important to recognize there are many ways to ask a question. The way you ask the question should be determined by the data you want to get back.

Expert researchers often frame questions in the form of ordinal scales. This enables them to collect additional information and better understand the opinions of respondents. An example will explain why experts use scales.

Consider this question about customer satisfaction for the fictional company XYZ Suppliers:

Potential results could look like this:

These results aren’t fantastic, but an XYZ executive could look at them, recognize that a majority of customers were satisfied, and not think anything else of it.

Compare this with the same question that is rephrased in a scale.

If this same question was asked to the same group of people who answered the yes/no question above, the result could look like this:


These results are much more alarming.

As you can see, the scale question revealed some very interesting results that the yes/no question missed. No respondents indicated they were “very satisfied” and 2/3 of respondents who were “satisfied” were only “somewhat satisfied.”

These results also give XYZ executives information that is actionable. For instance, XYZ may want to aim actions at the 53% of respondents who are only “somewhat satisfied” or “somewhat dissatisfied” in order to increase their satisfaction.

This information would have been missed in a yes/no question. This is why experts often use scales for their research. It’s a good idea to re-frame questions, when possible, in a scale format.