1. What is the general goal of the feature?
This feature would mean that every time we use select_multiple, ODK would create a new column for each of the possible answers and could code them as 0/1 based on how a person responded to the question.
This feature already exists in KoBoCollect.
2. What are some example use cases for this feature?
I often avoid using select_multiple due to the extra data cleaning required prior to analysis. This means that we end up asking study participants more questions using more select_one and may ultimately lead to longer surveys and perhaps poorer responses to surveys as time goes on.
3. What can you contribute to making this feature a reality?
User rather than developer and so couldn't really contribute to the development of this feature.
Hi @patrick_keating!
How do you export the data that you collect using ODK? For example, do you download your data from ODK Central, export it using ODK Briefcase, or submit your data directly to Google Sheets?
If you're using ODK Briefcase, note that there's an option to "Split select multiples" that's similar to what you describe (see here for more information).
It'd also be great to hear more about your data cleaning process and what the pain points are. What steps do you have to take to transform a select_multiple column into a set of 0/1 columns? Is there an aspect of that process that is complex or time-consuming?
1 Like
Hi Matthew,
Thanks a lot for your reply.
I have previously downloaded directly from the server and also from Briefcase.
I never noticed that option and thanks for pointing it out.
As for the data cleaning process, it's not a major deal, it just involves creating those additional variables with 0/1s. Just an extra step in the process and great to be able to avoid it.
Thanks a lot!
Patrick