Special Values for text questions

Add “Special Value” button for text questions as in numeric questions.
Often text questions with patterns are used in place of numerical questions. Special values would facilitate answers in many situations.

Example 1:
Q: “Enter Code:” (pattern “~~####”)
Special value: “Not applicable”
Special value: “Unknown”

Example 2:
Q: “Notes:”
Special value: “None”

2 Likes

Klaus, agree, this is a good idea. I will add it to the next sprint feature list.

Thanks much,

1 Like

Also would be good to add this to a date question. It is very common that e.g. the date of birth is not known, but currently this requires a screening question “Do you know the exact date of birth?”.

1 Like

Yes, agree, I will added it.

Dear Survey Solutions team,

I just wanted to ask if this feature was included. I couldn’t find it in Designer, and found a later question in the forum (Special Values on Text Questions), where the answer was to use two questions (which suggests the feature wasn’t included).

If not possible yet, I would find this feature useful too.

Thank you,

Sebastiaan

You are right, Sebastiaan, the string questions do not have any special values, to date.

Thanks for confirming Sergiy. Is it still a planned feature or have you decided against it?

Cheers,

Sebastiaan

It is not going to fly well in the context of multilingual surveys. For numeric questions what is stored is language-neutral (a code) but for text questions one will have to store the actual text. It is not impossible, but also is not a priority. What is your situation in the questionnaire?

Thanks for the explanation Sergiy, and sorry for the delay in replying (I wasn’t on the forum for a few days and didn’t get an email notification, I guess because I didn’t start the topic).

I understand that would be tricky. I wanted to create a question such as the one below. It was mainly intended for the pre-test, with closed options later. I just thought that it would be less of a change for enumerators if I could already have the pre-test question in the same format as the final one (instead of using a primer YES/NO question in the pre-test). (A similar example I would find useful would be to have a NONE option in an open question). But it is not a big deal, it would just streamline the questionnaire a bit, and can easily be dealt with in two questions.

Q1. Do you think there are benefits to population growth? If YES, what are they?

NO…0
DON’T KNOW…88

OPEN

Cheers,

Sebastiaan

Q1. [SINGLE SELECT] Do you think there are benefits to population growth?
YES…1
NO…2
DON’T KNOW…88

Q2: [TEXT] What are they ? (if Q1==1)


I’d be totally against keeping the two questions together as it, imho, implies strongly that there are benefits of the population growth and the only question is whether the respondent can mention them or not.

Still other respondents may be reacting to the need to mention the reasons and prefer the NO answer, simply because they don’t want to elaborate. By separating the questions you get the initial belief more objectively. Whether you will or will not get clarification is an entirely different matter.

You can also use instructions to clarify difficult situations:
Write a dash “-” if the respondent can’t mention any benefits.
This will be a specific sign that the interviewer did ask the question and received no response, rather than forgot to ask the question.

Thanks Sergiy,

I agree it is a little bit of a confounding question, but the enumerators would be trained to ask the second part only with a yes answer, so the interviewee would not be primed by the question. But it can indeed be done in two questions quite simply.

The dash is a good suggestion.

Thanks,

Sebastiaan