# Question about "Prefer not to answer" but then providing answers!

**URL:** <https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961>\
**Category:** Questions\
**Created:** [July 9, 2024, 2:55pm UTC](https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961 "2024-07-09T14:55:29Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![EmberM](https://community.weallcount.com/letter_avatar_proxy/v4/letter/e/3ec8ea/32.png) [@EmberM](https://community.weallcount.com/u/EmberM)\
**Post date:** [July 9, 2024, 2:55pm UTC](https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961/1 "2024-07-09T14:55:30Z")

</div>

Hi folks!

I am working with the largest data set of my career ( 😅) and have a question for you all!

When asking about social identity data, we provided options as well as a “prefer not to answer” option. At the end of the demographics section of the survey, we had an open ended question that said something similar to “Please tell us anything else you’d like to share about your background.”

So for example, a respondent put “prefer not to answer” for the race/ethnicity question, but then in the open-ended response about their background, they put “Mexican native American”.

Our team is wondering if we should recode their race/ethnicity response based on what they provided (So “Latino/Hispanic” and “Native American/Alaska Native”) or should we leave the “prefer not to answer” and add “Mexican native American” to a new “Race/Ethnicity2” variable we’ve already created for other use cases?

---

<div class="post-metadata">

**Author:** ![DaraMurray](https://community.weallcount.com/user_avatar/community.weallcount.com/daramurray/32/396_2.png) [@DaraMurray](https://community.weallcount.com/u/DaraMurray)\
**Post date:** [July 10, 2024, 12:55pm UTC](https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961/2 "2024-07-10T12:55:36Z")

</div>

Respondents always find new ways of answering our questions, don’t they? haha. Two quick questions: In your identity question, did you also include an option of “Prefer to self describe” or even “other”? Could the question be skipped? Asking because I think that might help inform my advice or suggestion to your team.

---

<div class="post-metadata">

**Author:** ![EmberM](https://community.weallcount.com/letter_avatar_proxy/v4/letter/e/3ec8ea/32.png) [@EmberM](https://community.weallcount.com/u/EmberM)\
**Post date:** [July 10, 2024, 4:19pm UTC](https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961/3 "2024-07-10T16:19:30Z")

</div>

Ha! Yes indeed! So I took a screen shot to show how the question would have appeared to the respondent. The question could not be skipped.

 ![Screenshot 2024-07-10 at 11.20.47 AM](https://community.weallcount.com/uploads/default/original/1X/49dc84cb64666948605e996f51dbe5b76f37d93d.png)

---

<div class="post-metadata">

**Author:** ![Jadevieira\_Bristol](https://community.weallcount.com/letter_avatar_proxy/v4/letter/j/6bbea6/32.png) [@Jadevieira\_Bristol](https://community.weallcount.com/u/Jadevieira_Bristol)\
**Post date:** [April 12, 2025, 10:21pm UTC](https://community.weallcount.com/t/question-about-prefer-not-to-answer-but-then-providing-answers/2961/4 "2025-04-12T22:21:40Z")

</div>

Hello I would not recode their answers. They choose prefer not to answer and thus that’s how I would code it. They may not realize their other question provided the answer to the 1st questions.