rojamila Evolution of Angiosperms(flowering plant)

 How do I set up a rating scale question correctly?

If you incorrectly set up a rating scale question, this could confuse your respondents, muddy your data results and waste your time. There are several things to note when setting up your rating scale question:

1. Identify your scale points

The scale’s points should map as closely as possible to the underlying idea (construct) of the scale. Also, the scale should give enough points to differentiate respondents from one another as much as validly possible

It should be easy to interpret the meaning of each scale point - i.e. the meaning of scale points should be interpreted identically by all respondents.

2. Using the right rating scale

The breakdown of each rating scale type above should give you an idea of how you can incorporate these into your survey (see the bolded text for a quick reference).

However, if you’re not sure about using one type over another, consider using them both in a test first. If your results show you what you were hoping to collect, then you know which test you should.

In numerical scales, ‘1’ must always represent the lowest scale end, and ‘5’ must always represent the highest. This is because your scale will run from left to right on a horizontal axis, so as you move to the right, the numbers should get bigger.

3. Gain more information with an extra field

When you include a rating scale, you may want to include a free-text answer box underneath to ask your respondent to expand on their earlier answer. This can give you the reasons why a person answered in the way they did, which could be compared with other similar results for verification.

In addition, you can add on another field to the scale, for people who are genuinely not sure - for example, a ‘Not applicable’, ‘Other’ or ‘Don’t know’ tickbox.

Usually, software will keep this separate from the scale points, to make sure it doesn’t confuse the respondent. This will also prevent a respondent from picking a random answer, and affecting your data results, from the pressure of not having an alternative action.