Let's say I want to ask a user wether she likes a song or not.

If I ask her to answer:

  • Yes
  • No
  • Don't know this song

I guess there's a risk that she will choose the last option for songs she knows only a bit.

However, if I use grade scale and ask her to answer with one of the following options:

  • Definitely yes!
  • Maybe yes
  • Rather no
  • Definitely not!
  • Don't know

I guess there will be fewer "Don't know" answers, because if a user knows the song only a bit, she will have a chance to "hedge" her answer and answer with maybe/rather answers.

Is there any research supporting my guess?


This is called a Likert Scale.

From MeasuringU:

As a general rule, when measuring a construct that falls on a continuum from low to high (such as satisfaction, ease, and likelihood to recommend), the more points you have in your rating scale, the more reliable (consistent responses) and valid (reflects true attitudes) it generally is. The bluntness of scales with few points reduces correlations and causes the reduction in reliability and validity. We pull much of this from the work of Jum Nunnally, the noted psychometrician who wrote about the reliability of scales in 1967, 1978 and in 1994.

Implied here is that if a user sees an option that may more accurately reflect their feeling, they'll pick that instead of "Don't Know". Keep in mind that the more points, the more cognitive load on the users, which may have a detrimental effect.

  • Thank you. I called it graded scale, because this is how it called in information retrieval where I come from: graded relevance judgements. – dzieciou Sep 17 '20 at 16:56

Obviously the more options you present, the more likely it is that they will use them rather than just selecting "Don't Know" - unless they actually don't know the song of course. So directly answering your question, yes it will help if you have more options.

However, if you are asking a user to rate a song then you should opt for a more familiar approach. Allow the user to select a rating score. For example, let them select from 0-5 stars, or a 1-10 rating, etc. This easily covers the "grey areas" and nearly all users will be familiar with this concept.

With regards to handling the "don't knows" in a rating system. Either allow the user to not select a rating and skip that song, or if you want explicit action to confirm it then just have a button next to the rating control for "I don't know this song".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.