I am implementing a household survey using ODK/SurveyCTO. As, I guess, in
almost all surveys I have to provide "do not know", "refuse to answer",
"not applicable" as answers to almost all questions of all types
(select_one, select_multiple, text, integer). Is there any simple solution
provided by ODK to do so? (Like or_other provides a very simple way to add
the possibility of giving a different answer to select_one or
select_multiple questions).
Currently, I am thinking to implement it as follows:
select_one, select_multiple: Add codes for "do not know", "refuse to
answer", "not applicable" to all answer sets
text: Teach enumerators fixed codes for these possibilities
integer, decimal: Let -9, -8 and -7 be codes for these possibilities and
update all constraints on these fields accordingly
This is essentially the paper-based solution. Isn't there something better?
As before, you've done a pretty good job outlining the options as I
understand them.
When it comes to allowing special codes (like -888 or -999), we've tried to
make this easy by including "exceptions" in the Constraint Builder; so,
e.g., you can say that an age has to be between 1 and 150, with -777, -888,
and -999 as allowable exceptions. These can become somewhat complex
constraint expressions, but we try to at least make it easy with the
Builder.
I'm not aware of anything much better than the options you describe,
unfortunately.
Best,
Chris
···
On Sat, Dec 21, 2013 at 8:30 AM, Wouter Gelade wrote:
Hi all!
I am implementing a household survey using ODK/SurveyCTO. As, I guess, in
almost all surveys I have to provide "do not know", "refuse to answer",
"not applicable" as answers to almost all questions of all types
(select_one, select_multiple, text, integer). Is there any simple solution
provided by ODK to do so? (Like or_other provides a very simple way to add
the possibility of giving a different answer to select_one or
select_multiple questions).
Currently, I am thinking to implement it as follows:
select_one, select_multiple: Add codes for "do not know", "refuse to
answer", "not applicable" to all answer sets
text: Teach enumerators fixed codes for these possibilities
integer, decimal: Let -9, -8 and -7 be codes for these possibilities and
update all constraints on these fields accordingly
This is essentially the paper-based solution. Isn't there something better?
Hi Wouter,
Did this ever work? Does -777,-888 and -999 work respectively for "do not
know", "refuse to answer", "not applicable"?
I'm looking for a work around for my select_one options
Thanks!
Amber
···
On Saturday, 21 December 2013 08:30:02 UTC-5, Wouter Gelade wrote:
>
> Hi all!
>
> I am implementing a household survey using ODK/SurveyCTO. As, I guess, in
> almost all surveys I have to provide "do not know", "refuse to answer",
> "not applicable" as answers to almost all questions of all types
> (select_one, select_multiple, text, integer). Is there any simple solution
> provided by ODK to do so? (Like or_other provides a very simple way to add
> the possibility of giving a different answer to select_one or
> select_multiple questions).
>
> Currently, I am thinking to implement it as follows:
> - select_one, select_multiple: Add codes for "do not know", "refuse to
> answer", "not applicable" to all answer sets
> - text: Teach enumerators fixed codes for these possibilities
> - integer, decimal: Let -9, -8 and -7 be codes for these possibilities and
> update all constraints on these fields accordingly
>
> This is essentially the paper-based solution. Isn't there something better?
>
> Thanks!
> Wouter
>
I'm not sure I exactly understand your question. But yes, I used the system
outlined above to deal with special answers and it worked fine. For
instance, I used -7 throughout the questionnaire for "not applicable" in
numeric questions. For select_one, I would have the possible answer "Not
applicable (-7)". Adding the -7 in the select_one answer is not really
needed, but I hoped it would train enumerators unconsciously that "Not
applicable = -7" for numeric questions.
Best,
Wouter
···
On Monday, October 27, 2014 10:15:13 PM UTC+1, Amber Rieder wrote:
>
> Hi Wouter,
> Did this ever work? Does -777,-888 and -999 work respectively for "do not
> know", "refuse to answer", "not applicable"?
>
> I'm looking for a work around for my select_one options :)
>
> Thanks!
> Amber
>
>
> On Saturday, 21 December 2013 08:30:02 UTC-5, Wouter Gelade wrote:
>>
>> Hi all!
>>
>> I am implementing a household survey using ODK/SurveyCTO. As, I guess,
>> in almost all surveys I have to provide "do not know", "refuse to answer",
>> "not applicable" as answers to almost all questions of all types
>> (select_one, select_multiple, text, integer). Is there any simple solution
>> provided by ODK to do so? (Like or_other provides a very simple way to add
>> the possibility of giving a different answer to select_one or
>> select_multiple questions).
>>
>> Currently, I am thinking to implement it as follows:
>> - select_one, select_multiple: Add codes for "do not know", "refuse to
>> answer", "not applicable" to all answer sets
>> - text: Teach enumerators fixed codes for these possibilities
>> - integer, decimal: Let -9, -8 and -7 be codes for these possibilities
>> and update all constraints on these fields accordingly
>>
>> This is essentially the paper-based solution. Isn't there something
>> better?
>>
>> Thanks!
>> Wouter
>>
>
See how that works? If a value is >=0, it uses the value, otherwise it uses
0 in place of the value. You're going to have to do something like that to
implement your desire to ignore negative numbers when summing. I don't
think that there's any way for the software to do what you want totally
automatically.
Best,
Chris
···
On Mon, Oct 27, 2014 at 5:15 PM, Amber Rieder wrote:
Hi Wouter,
Did this ever work? Does -777,-888 and -999 work respectively for "do not
know", "refuse to answer", "not applicable"?
I'm looking for a work around for my select_one options
Thanks!
Amber
On Saturday, 21 December 2013 08:30:02 UTC-5, Wouter Gelade wrote:
Hi all!
I am implementing a household survey using ODK/SurveyCTO. As, I guess,
in almost all surveys I have to provide "do not know", "refuse to answer",
"not applicable" as answers to almost all questions of all types
(select_one, select_multiple, text, integer). Is there any simple solution
provided by ODK to do so? (Like or_other provides a very simple way to add
the possibility of giving a different answer to select_one or
select_multiple questions).
Currently, I am thinking to implement it as follows:
select_one, select_multiple: Add codes for "do not know", "refuse to
answer", "not applicable" to all answer sets
text: Teach enumerators fixed codes for these possibilities
integer, decimal: Let -9, -8 and -7 be codes for these possibilities
and update all constraints on these fields accordingly
This is essentially the paper-based solution. Isn't there something
better?
See how that works? If a value is >=0, it uses the value, otherwise it
uses 0 in place of the value. You're going to have to do something like
that to implement your desire to ignore negative numbers when summing. I
don't think that there's any way for the software to do what you want
totally automatically.
Hi Wouter,
Did this ever work? Does -777,-888 and -999 work respectively for "do
not know", "refuse to answer", "not applicable"?
I'm looking for a work around for my select_one options
Thanks!
Amber
On Saturday, 21 December 2013 08:30:02 UTC-5, Wouter Gelade wrote:
Hi all!
I am implementing a household survey using ODK/SurveyCTO. As, I guess,
in almost all surveys I have to provide "do not know", "refuse to answer",
"not applicable" as answers to almost all questions of all types
(select_one, select_multiple, text, integer). Is there any simple solution
provided by ODK to do so? (Like or_other provides a very simple way to add
the possibility of giving a different answer to select_one or
select_multiple questions).
Currently, I am thinking to implement it as follows:
select_one, select_multiple: Add codes for "do not know", "refuse to
answer", "not applicable" to all answer sets
text: Teach enumerators fixed codes for these possibilities
integer, decimal: Let -9, -8 and -7 be codes for these possibilities
and update all constraints on these fields accordingly
This is essentially the paper-based solution. Isn't there something
better?
