Ask more, know more, do more

Don’t Know?

 → 
 → 
Don’t Know?

Don’t know whether to include a “Don’t Know” (or “Not Applicable”) response option for one of your questions? Don’t worry, you’re not alone. Here are some do’s and don’ts.

The debate about whether to include a “Don’t Know” (DK) option or not starts with a philosophy about valuing “clean” data and valuing an amount of data—both are valid arguments. The latter is grounded in the reality that survey data is expensive to come by in most cases, due to the time required to collect it and the costs of fieldwork. In that light, researchers who employ a DK response may be wasting what could have otherwise been useful responses from people who are attracted to the ease of offering a satisfactory, but ultimately unprovable, response.

But assuming we could spare a few of these satisficers, one of the advantages of a DK response option is that it helps reduce noise caused by people who offer a response to a closed ended response option. That is, people who haven’t thought much about a topic or don’t have the experience or attitude relevant to the question are indistinguishable from people who do because, after all, selecting a radio button in a survey is the same in both cases.

So the answer to the question about whether to include a DK response option is: it depends. It depends on whether you can afford to throw out a bit of data in order to make the part you keep cleaner. There is another factor to help make a decision about whether to include a DK option: the demographics of the people being surveyed and the nature of the question.

For example, if your survey question is “What is the GDP of China?” and your sample is comprised of second graders, you might want to include a DK option. If, on the other hand, the sample is comprised of American adults then a result like “20% of respondents reported ‘Don’t Know’,” then that is a real finding that may have important implications for making a decision about school curriculum. In short, if you have an expectation that the respondents ought to know the answer to a question like “How tall are you?,” then don’t include a DK option. But if not knowing is plausible and even important to know, then include a DK option.

Now you know.

Tags: , , ,

7 thoughts on “Don’t Know?

  1. Fathi Ibrahim says:

    hi dear
    I do not know how I will do the next step

    1. Bennett P says:

      Fathi – The next step is to sign up for a free account here!

  2. I’m having the same problem as Richard had. I can post a link to my survey on my wall, but no survey is embedded on the SurveyMonkey app page I created. Using Internet Explorer, a message briefly appeared saying we are running no surveys at the moment, come back later. But using other browsers that message doesn’t even show up.

    Also please provide instructions for removing the SM app if needed.

    1. Bennett P says:

      Birchandra – email support@surveymonkey.com so they can help. Here’s a link to a PDF to help in the meantime – http://help.surveymonkey.com/euf/assets/docs/pdf/FacebookApp.pdf

  3. Pingback: Don’t know…
  4. Navneet Mathur says:

    Yes this right one of the advantages of a DK response option is that it helps reduce noise caused by people who offer a response to a closed ended response option. But one is you can escape the from a unconfirm answer.

    1. Hanna J says:

      Hi Navneet – You’re so right. Sounds like you have a little bit of survey methodologist in you as well! Thanks for sharing. Have a great day.

Leave a Reply

Your email address will not be published. Required fields are marked *

Inspired? Create your own survey.

Make your own survey

Sign In

Sign up to get started

PRO Sign Up 
Sign Up FREE

Latest Features

Curious about what we’ve been working on?

View New Features

Get expert survey help

Get expert survey help

 

Best practices for planning expert surveys

Planning makes writing your survey easy.

Download eGuide

How many responses do you need?

Use our sample size calculator to find out!

Calculate now

Analyze survey data like a pro

Learn to slice and dice data using the Analyze tool.

Download eGuide