[ODK Community] Analysing data from multiple select/ODK aggregate


In Stata, this is simple to resolve: you say "split varname, destring", and
it creates varname1, varname2, etc. variables, to hold the first choice,
second choice, etc. With the "destring" option, it converts the split
variables into numeric variables (since the initial list would have been
imported as a string). Then, when you want everyone who answered 2, you can
say the equivalent of "if varname1==2 || varname2==2 || varname3==2 ||
varname4==2" (for as many options as you have).

I'm sure that SAS and other packages have similar functions, and "split" is
actually a common name for this kind of function across many programming



ยทยทยท On Thu, Apr 4, 2013 at 8:17 PM, wrote:


I wondered if anyone had any tips on analysing data gathered using a
'select multiple' question, with data processed by ODK aggregate

At the moment, for each set of responses it produces a string separated by
spaces in each cell, for example:

1 2
1 3 4
2 4

But the problem is, this makes things tricky for data analysis. Sometimes,
I am interested just in everyone who answered '2' - but it isn't possible
to filter the results to see just the people who answered '2' - because
some people answered '1 2', some '2 4' etc.

I wondered what your tips were on dealing with this kind of data?




Post: opendatakit@googlegroups.com
Unsubscribe: opendatakit+unsubscribe@googlegroups.com
Options: http://groups.google.com/group/opendatakit?hl=en

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 opendatakit+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.