Hi all, I’m new to the forum and a relative beginner on Kobo (always used FormBuilder only). Sorry for my non-technical language!
I’m designing a 2-part survey, for women before and after giving birth (pre/post survey). Each woman will be interviewed twice at a distance of several months, and I would like a fool-proof way to match the 2 parts for the same woman - obviously simply asking to enter names manually will create a lot of inconsistencies. There is no pre-existing database to pull names/IDs from initially, as the names are collected as part of the pre survey. All data will be collected by “enumerators” (i.e. field staff not particularly tech-savvy) through the KoboCollect app, mostly working offline.
I see 2 options, but not sure if either is feasible or how to implement:
- There is a single form divided in 2 parts, the enumerator chooses the right part through skip logic, first time fills out the pre-survey, second time goes back and fills out the post. This requires a way to go back and search previous submissions in an easy way (eg by name) and make edits – is that even feasible on KoboCollect once you upload a form to the server?
- The 2 forms are separate, the pre-survey form dynamically populates a database with names/IDs, and the post-survey form then pulls from that database to provide a list of names for the enumerator to select for the post survey. The matching will then need to be done externally, but the match will be guaranteed. I know there is a pull data function that can be programmed using xls forms, I haven’t dived into that yet, but am not sure if it can work with a csv that is not static but updates as more submissions come in from the pre survey?
Any other ideas or pointers on how to best handle this scenario? It can’t be the first time someone designs a longitudinal survey… Again, this all needs to work on KoboCollect in a context where enumerators are mostly offline (but occasional synchronizations can happen).
Thanks in advance for your suggestions and help!!
Cecilia
