Kobo Toolbox integration for survey creation / management


#1

@mdc ok please modify, here’s a starting point…

Our Sci Kit is unique in it’s ability to integrate sensor, survey, and API data in a flexible way. However, we recognize that there is a large amount of work done to make the survey component easy to use and easy to view the results. We hope to utilize existing open source options which have been dealing with survey creation and data display for a long time.

Kobo Toolbox seems like a great fit. It’s well maintained, easy to use, and has some key features that we will need. Specifically:

  1. Survey creation, editing, and rights management
  2. Survey deployment (versioning, archiving, etc.).
  3. Nice GUI for creating surveys and viewing basic results.
  4. “Library” feature for utilizing groups of questions or individual questions across many surveys.

Additional feature needs:

  1. We want a public library feature which cuts across all users. This should include whole surveys, question groups, and individual questions. Ideally, it would also show the number of data points in other projects using that survey/question set/question and perhaps the largest users to help motivate reuse (“if Harvards using this survey and there are 10,000 comparable data points, we should definitely use it!”). We want to direct users first to replicate others’ whole work, replicate parts of their work, and finally (if all else fails) to create something new. The goal is to increase comparability of data by making re-use of others’ work the easiest path. We may also need some explicit “comparable” tag which two creators of surveys applies manually, as even using the same survey does not automatically imply the data is comparable.

  2. We would need to integrate our sensor question-type into Kobo’s interface.

There are a few ways that we could use Kobo…

  1. Have a Kobo docker container
  2. have Kobo maintain an image for us which is modified / branded. That way we don’t have to worry about database management which may be trickier than just copy/pasting a docker image.

There may be some ways Kobo and Our Sci collaborate…

  1. We could help integrate some of the sensor work we’ve done into their platform and the broader ODK community. This would add an asset they could sell to development organizations in terms of utilizing sensors in their deployments, or accessing APIs, or delivering more flexible results.
  2. They could help build/maintain an Kobo survey creation instance for us which is tailed to our application and user base which is more sensor / science focused.
  3. There may be some collaboration possible on the Android app side.