A rating scale is a method that requires the respondent to assign a value, sometimes numeric, to the rated object.
This guide will teach you:
- How to set it up correctly
- How many points should your scale have?
- Different types of rating scales
- Best-practice tips
1. How to set it up correctly
- It should be easy to interpret the meaning of each point
- The meaning of the points should be interpreted identically by all respondents
- You should include enough points to allow for differentiation between respondents as much as validly possible
- Responses should be reliable, meaning that if we asked the same question again, each respondent should provide the same answer
- The points should map as closely as possible to the underlying idea (construct) of the scale
2. How many points should your scale have?
The number of points depends on what sort of question you’re asking.
If you’re asking something that ranges from positive to negative (also known as bi-polar constructs) then you’re going to want a 7-point that includes a middle or neutral point. In practice, this means the response options for a satisfaction question should look like this:
If you’re asking a question that ranges from zero to positive (also known as unipolar constructs) then you’ll go with a 5-point. The response options for this kind of question would look like this:
The goal is to make sure respondents can answer in a way that allows them to differentiate themselves as much as is validly possible without providing so many points that the scale becomes unreliable. Even on an 11-point (0-10), respondents start to have difficulty reliably placing themselves: for example, 3 isn’t so different from 4.
3. Types of rating scales
- Numeric rating scale
- Graphic rating scale
- Descriptive graphic rating scale: A descriptive graphic rating scale has two defined endpoints. A line connects these points. Descriptive phrases identify different points along the continuum.
Learn how to use the Rating question in Survey Anyplace.
4. Best-practice tips
- Make sure to balance the answers by having an equal number of positive and negative categories.
- Research has shown that respondents are generally biased towards the left-hand side of a scale that is bi-polar (including negative and positive options). For example, if the answers starts with “strongly agree” on the left and goes to “strongly disagree” on the right, the results would come out differently than if “strongly disagree” would be placed on the left. If you’re asking a few questions using a rating scale, you can minimize this bias by sometimes starting with positive categories on the left, and sometimes starting with negative categories.
- Think carefully about the wording at each end of your answers. If the wording sounds too extreme, this may be off-putting to respondents, as generally, people prefer not to appear as extremists. Consider using “very good” as an end-point in place of “terrific” or “very bad” instead of “dreadful”.
- You can “force” respondents to choose an option by not including a category such as “unsure” or “undecided”, however, this may lead to respondents having to choose an option that they don’t really agree with, or they haven’t yet fully decided about.