Does ODK have a provision to warn an enumerator about, but not prevent
them from entering, a value that is likely a data-entry error, but may
simply be a true outlier? We would like enumerators to be told by ODK
that a value appears incorrect, but allow them to enter it anyway, if
it can be confirmed.
For example:
-- enumerator enters # (100) for number of children in a classroom,
this falls outside of a constraint (maximum 40)
-- ODK window pops up saying: the response you entered may be an
error. can you check it and choose one of the following:
a) (if it's an error) enter a different response
b) (if it's correct) keep the response as previously entered
You can, however, add a confirmation question afterward that is relevant
only if the value is an outlier.
That confirmation question can either be (1) an informational read-only
field that the enumerator can blow through if they want, or (2) a required
select1 that defaults to "no", but is constrained to "yes" and therefore
requires the enumerator to select the "yes" in order to keep the outlier,
thereby getting their acknowledgement that this is a valid outlier.
A bit clunky, but gets the job done.
Mitch
···
On Fri, Oct 28, 2011 at 10:15 AM, Edwin Adkins wrote:
Does ODK have a provision to warn an enumerator about, but not prevent
them from entering, a value that is likely a data-entry error, but may
simply be a true outlier? We would like enumerators to be told by ODK
that a value appears incorrect, but allow them to enter it anyway, if
it can be confirmed.
For example:
-- enumerator enters # (100) for number of children in a classroom,
this falls outside of a constraint (maximum 40)
-- ODK window pops up saying: the response you entered may be an
error. can you check it and choose one of the following:
a) (if it's an error) enter a different response
b) (if it's correct) keep the response as previously entered
assuming I take the second approach that you mention (adding a y/n):
the "yes" is clear (the user confirms it's a true outlier) but what if
it's a "no" (the user acknowledges that they've made a data-entry
error). Then what? Can I program ODK to "send" the user back a
question to re-enter the preceding response? I don't know how to do
this...
edwin
···
On Oct 28, 2:43 pm, Mitch Sundt wrote:
> Not as pop-ups like a hard constraint.
>
> You can, however, add a confirmation question afterward that is relevant
> only if the value is an outlier.
>
> That confirmation question can either be (1) an informational read-only
> field that the enumerator can blow through if they want, or (2) a required
> select1 that defaults to "no", but is constrained to "yes" and therefore
> requires the enumerator to select the "yes" in order to keep the outlier,
> thereby getting their acknowledgement that this is a valid outlier.
>
> A bit clunky, but gets the job done.
>
> Mitch
>
> On Fri, Oct 28, 2011 at 10:15 AM, Edwin Adkins wrote:
>
>
>
>
>
>
>
>
>
> > Does ODK have a provision to warn an enumerator about, but not prevent
> > them from entering, a value that is likely a data-entry error, but may
> > simply be a true outlier? We would like enumerators to be told by ODK
> > that a value appears incorrect, but allow them to enter it anyway, if
> > it can be confirmed.
>
> > For example:
> > -- enumerator enters # (100) for number of children in a classroom,
> > this falls outside of a constraint (maximum 40)
> > -- ODK window pops up saying: the response you entered may be an
> > error. can you check it and choose one of the following:
>
> > a) (if it's an error) enter a different response
> > b) (if it's correct) keep the response as previously entered
>
> > is that do-able?
>
> > thanks,
>
> > Edwin Adkins
>
> > --
> > Post: opendatakit@googlegroups.com
> > Unsubscribe: opendatakit+unsubscribe@googlegroups.com
> > Options:http://groups.google.com/group/opendatakit?hl=en
>
> --
> Mitch Sundt
> Software Engineerhttp://www.OpenDataKit.org
> University of Washington
> mitchellsu...@gmail.com
···
On Wed, Nov 2, 2011 at 2:00 PM, Edwin Adkins wrote:
ok, many thanks for the reply.
one follow-up question:
assuming I take the second approach that you mention (adding a y/n):
the "yes" is clear (the user confirms it's a true outlier) but what if
it's a "no" (the user acknowledges that they've made a data-entry
error). Then what? Can I program ODK to "send" the user back a
question to re-enter the preceding response? I don't know how to do
this...
You can, however, add a confirmation question afterward that is relevant
only if the value is an outlier.
That confirmation question can either be (1) an informational read-only
field that the enumerator can blow through if they want, or (2) a
required
select1 that defaults to "no", but is constrained to "yes" and therefore
requires the enumerator to select the "yes" in order to keep the outlier,
thereby getting their acknowledgement that this is a valid outlier.
Does ODK have a provision to warn an enumerator about, but not prevent
them from entering, a value that is likely a data-entry error, but may
simply be a true outlier? We would like enumerators to be told by ODK
that a value appears incorrect, but allow them to enter it anyway, if
it can be confirmed.
For example:
-- enumerator enters # (100) for number of children in a classroom,
this falls outside of a constraint (maximum 40)
-- ODK window pops up saying: the response you entered may be an
error. can you check it and choose one of the following:
a) (if it's an error) enter a different response
b) (if it's correct) keep the response as previously entered
you can either tell the user to swipe back, or you can have them
re-enter the value inside a new widget.
if this is prompt that can be asked another way, one survey design
trick is to ask a related question that helps you figure out if the
first question was correct. so you ask for age, then a few questions
later ask for birth year.
···
On Thu, Nov 3, 2011 at 03:49, Mitch Sundt wrote:
> I don't think that's possible. Yaw - thoughts?
>
> Mitch
>
> On Wed, Nov 2, 2011 at 2:00 PM, Edwin Adkins wrote:
>>
>> ok, many thanks for the reply.
>>
>> one follow-up question:
>>
>> assuming I take the second approach that you mention (adding a y/n):
>> the "yes" is clear (the user confirms it's a true outlier) but what if
>> it's a "no" (the user acknowledges that they've made a data-entry
>> error). Then what? Can I program ODK to "send" the user back a
>> question to re-enter the preceding response? I don't know how to do
>> this...
>>
>> edwin
>>
>> On Oct 28, 2:43 pm, Mitch Sundt wrote:
>> > Not as pop-ups like a hard constraint.
>> >
>> > You can, however, add a confirmation question afterward that is relevant
>> > only if the value is an outlier.
>> >
>> > That confirmation question can either be (1) an informational read-only
>> > field that the enumerator can blow through if they want, or (2) a
>> > required
>> > select1 that defaults to "no", but is constrained to "yes" and therefore
>> > requires the enumerator to select the "yes" in order to keep the
>> > outlier,
>> > thereby getting their acknowledgement that this is a valid outlier.
>> >
>> > A bit clunky, but gets the job done.
>> >
>> > Mitch
>> >
>> > On Fri, Oct 28, 2011 at 10:15 AM, Edwin Adkins wrote:
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > > Does ODK have a provision to warn an enumerator about, but not prevent
>> > > them from entering, a value that is likely a data-entry error, but may
>> > > simply be a true outlier? We would like enumerators to be told by ODK
>> > > that a value appears incorrect, but allow them to enter it anyway, if
>> > > it can be confirmed.
>> >
>> > > For example:
>> > > -- enumerator enters # (100) for number of children in a classroom,
>> > > this falls outside of a constraint (maximum 40)
>> > > -- ODK window pops up saying: the response you entered may be an
>> > > error. can you check it and choose one of the following:
>> >
>> > > a) (if it's an error) enter a different response
>> > > b) (if it's correct) keep the response as previously entered
>> >
>> > > is that do-able?
>> >
>> > > thanks,
>> >
>> > > Edwin Adkins
>> >
>> > > --
>> > > Post: opendatakit@googlegroups.com
>> > > Unsubscribe: opendatakit+unsubscribe@googlegroups.com
>> > > Options:http://groups.google.com/group/opendatakit?hl=en
>> >
>> > --
>> > Mitch Sundt
>> > Software Engineerhttp://www.OpenDataKit.org
>> > University of Washington
>> > mitchellsu...@gmail.com
>>
>> --
>> Post: opendatakit@googlegroups.com
>> Unsubscribe: opendatakit+unsubscribe@googlegroups.com
>> Options: http://groups.google.com/group/opendatakit?hl=en
>
>
>
> --
> Mitch Sundt
> Software Engineer
> http://www.OpenDataKit.org
> University of Washington
> mitchellsundt@gmail.com
>
> --
> Post: opendatakit@googlegroups.com
> Unsubscribe: opendatakit+unsubscribe@googlegroups.com
> Options: http://groups.google.com/group/opendatakit?hl=en
>
You could constrain the second question to only accept an answer of "yes"
and make the constraint text something along the lines of, "please go to
the previous question and change your answer", but you can't explicitly
jump backwards to a certain question.
···
On Thu, Nov 3, 2011 at 5:46 AM, Yaw Anokwa wrote:
you can either tell the user to swipe back, or you can have them
re-enter the value inside a new widget.
if this is prompt that can be asked another way, one survey design
trick is to ask a related question that helps you figure out if the
first question was correct. so you ask for age, then a few questions
later ask for birth year.
assuming I take the second approach that you mention (adding a y/n):
the "yes" is clear (the user confirms it's a true outlier) but what if
it's a "no" (the user acknowledges that they've made a data-entry
error). Then what? Can I program ODK to "send" the user back a
question to re-enter the preceding response? I don't know how to do
this...
You can, however, add a confirmation question afterward that is
relevant
only if the value is an outlier.
That confirmation question can either be (1) an informational
read-only
field that the enumerator can blow through if they want, or (2) a
required
select1 that defaults to "no", but is constrained to "yes" and
therefore
requires the enumerator to select the "yes" in order to keep the
outlier,
thereby getting their acknowledgement that this is a valid outlier.
Does ODK have a provision to warn an enumerator about, but not
prevent
them from entering, a value that is likely a data-entry error, but
may
simply be a true outlier? We would like enumerators to be told by
ODK
that a value appears incorrect, but allow them to enter it anyway,
if
it can be confirmed.
For example:
-- enumerator enters # (100) for number of children in a classroom,
this falls outside of a constraint (maximum 40)
-- ODK window pops up saying: the response you entered may be an
error. can you check it and choose one of the following:
a) (if it's an error) enter a different response
b) (if it's correct) keep the response as previously entered