dhis2-devs team mailing list archive

[Bob Jolliffe <bobjolliffe@xxxxxxxxx>]

2011/10/3 Lars Helge Øverland <larshelge@xxxxxxxxx>:
> Hi,
>
> I think your data element/category definitions make sense in this
> case. One should clearly not join the four category options you
> mention into the same category here.
>
> In Kenya we discovered that 90 % of the forms fit with the classic
> data-elements-on-the-rows-and-categories-on-the-columns pattern,
> meaning the auto-form function was not evil but rather quite handy.

Sure.  alternatively not settling concretely on the form design before
designing the dataelements could work here.

> When designing data elements it's also important to make sure it
> becomes easy to produce analysis - ideally what you typically want to
> look for in reports should act as data elements. Your approach here
> might work fine but trying to define what sort of reports are required
> before settling completely on the data elements might be smart.

Sound advice.  We should look at that.

>
> Re the age categories issue I am not sure. It might work fine. Again
> it might become a bit unwieldy when creating reports - you now need to
> "remember" for which age category options this data is collected for
> in order to avoid "blank cells" in reports/pivots.

Yes the "remembering" is a pain.  With a concept/category hierarchy
this will become easier, but for short term workaround it won't be too
bad and will be easier to convert existing data to updated design
later.  Mind you, the pivot tables should be much better, not worse,
in terms of blank cells.

>Anyway we will
> improve this model soon.
>
> Hope things went well in Rwanda

Thanks for the input
Bob
.
>
> cheers
>
> Lars
>