Missing values in market research with questfoxPosted: 8. March 2021
The phenomenon is not recent. People do not answer to questions. If someone does not answer to a question we are talking about so called missing values.
Also see the Wikipedia article: https://en.wikipedia.org/wiki/Missing_data
questfox knows 3 different kinds of missing values.
Missing value -99
someone clicked explicitly on “no answer”
Missing value -77
the question was not even shown to the repsondent (due to skipping/hiding procedures)
Missing value -88
classic missing value meaning the question was shown, but the answer was not mandatory so the respondent could click through
A good idea is to make someone click on each question, but this means that you will have to add the option “no answer / do not want to tell” to every answer where you require a mandatory click. Even if you do some of the questions or items will not be shown and therefore create a Missing value (-77).
Why is it negative with two digits in questfox?
The answer is pretty simple. If you use a scale from 1 to 5 and your default missing value is 9 (as some of us learned in their statistics class), you may not realize that you have an issue with missing values in your report. In the questfox default you would easily see that something is wrong in your definition of missing values.
These three default values are assigned to every question type and can be changed under “Define Missing Values” in the edit answer section.
When using SPSS these values can be defined as discrete missing values.
Now it is up to you if you can force your population into answering all the questions.