Can the Respondent Have No Opinion?

One of the biggest questions in survey research is whether or not a "no opinion" or "neither agree nor disagree" option is useful. On a basic level, if every respondent was guaranteed to answer each question honestly, leaving the choice would be worthwhile.

But many respondents use those questions as a crutch, choosing to claim they have no opinion either to hurry the survey along or so that they do not have to think critically about their choices. Forcing the user to create an opinion, then, would presumably be more advantageous. By leaving those types of options off of the survey, the respondent is then forced to choose an answer that they may not necessarily agree with.

Yet this can also create its own issues, including: Frustration If the user really does have no opinion, they may get frustrated that the answer is not on the screen. Not every question is controversial, and if there is no N/A or "No Opinion" answer for them to select in order to get out of answering the question, there is a risk that the user will become frustrated and either quit or answer poorly in an effort to move forward. Odd Scales Many people are used to odd number scales, with the middle number representing the mid-point.

Removing the "No Opinion" open may throw off the way people see these scales, since there will now be even numbers, a position most people aren't used to. This can be fixed with proper labeling of scales and consistency throughout the survey, however. Misleading Choices Finally, a forced choice doesn't guarantee an accurate or critically thought out choice.

The same people that would rush through a survey with the no opinion option may still rush through a survey forcing the choice at random. Still, overall, if the user doesn't have an opinion on a topic, having a "no opinion" option is useful. So what you should ask yourself when you create the scales on your survey is just how much the questions in your survey are likely to have an opinion.

Make sure you adjust for the entire survey, however, because you want to keep scale consistency throughout the survey, which would be impossible if some options have a "no opinion" while other questions don't. Still, if you discover that it's possible for your respondents to have no opinion, make sure you keep the question on. If it's not, consider keeping it off and forcing your sample to think critically.