ODK Survey, in public Alpha (and suitable only for pilot studies where you
are comfortable with loosing any or all of your data), has some of this
If you can define a full set of all values you will be collecting, and if
you do not need repeat groups (e.g., limit your survey to collect
information on the first N people living in a dwelling), then you can write
multiple forms that manipulate and access data from that one full set of
We're just beginning work on a pilot study using ODK Survey consisting of:
(1) initial screening interview
(2) detailed follow-up interview
(3) post-follow-up #1
(4) post-follow-up #2
By consolidating the answer fields for all the questions from all of these
screenings into a single table, you can develop a different form for each
of the 4 questionnaires, all of which read from and write to that single
table of values. The completed results from each of these questionnaires
would be published up to ODK Aggregate as a separate stream of submissions
tied to whichever form is used to do the upload.
The two example forms "Initial Refrigerator Information" and "Refrigerator
information update form" demonstrate this usage. I used the first to
create a new refrigerator record. I then chose to submit that to ODK
Aggregate (this is optional, as the data is held on the phone), then I used
the 2nd form to update the status of the fridge and give a stocking level.
I then uploaded that data to ODK Aggregate (this hits a bug in the way the
ODK Survey alpha works, so it requires a work-around -- long-press the
"Toggle All" button, and select "Sent and Unsent Forms". You can then
select the updated record and send that to ODK Aggregate). You can see the
two different result streams on the ODK 2 server:
Because we don't have ODK Tables available yet, and that can handle adding
and removing data rows from a table (to re-allocate them to different
devices / case workers), there are operational challenges to this pilot
(a) either the phone must retain all the information from all the
interviews for the duration of the study (in order for any follow-up
interviewer to be able to see and update data from the previous visit) or
extra steps would be required to manage and reassign cases across the
phones. This is doable, as all the data is stored in a SQLite database,
and you would be able to merge data across all phones in some sort of
manual process to share all cases or re-distribute cases.
(b) theft or loss of the phones means that having all the information
available on all devices is likely not a good idea, especially since the
period of the study is several months long.
On Thu, Jan 31, 2013 at 1:24 AM, Victoria Trinies email@example.com:
If this is a one-time survey, and you don't need the software to refer to
the respondent's information at a subsequent visit, I think ODK can handle
what you're asking for. You would need to set up groups of questions with
constraints so that the right set of questions show up based on the
responses to your gender/age group questions.
On Wednesday, 30 January 2013 13:14:55 UTC, Asen wrote:
To whom it may concern,
We have taken an interest in ODK and are considering making use of it for
an upcoming survey. I have a few questions, maybe you can point me in the
Is it currently possible to gather data from several respondents then
later (on the same questionnaire), retrieve the gathered data from the
different respondents in order to apply more information?
I hope I am being clear, what we want to do is gather details of 2/3
people in the household, then once we have done so, go back to each
gathered member, that meets a certain criteria and obtain more information,
based on their gender or age groups for example.
Looking forward to your kind feedback.
You received this message because you are subscribed to the Google Groups
"ODK Community" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to firstname.lastname@example.org.
For more options, visit https://groups.google.com/groups/opt_out.
University of Washington