dhis2-devs team mailing list archive

[Maheed Ramadan <maheed.ramadan@xxxxxx>]

Hi Bob,

Regarding point 1, you're right. What I meant is that currently the actual and target data elements are not related in a way or another. So what is missing is some way to link them.

I am also with you that the target value could be either date bound and space bound. But I was just looking at it from a simple case/perspective.

Maybe your point to define target values for indicators is logical. But if this is done so that we can select a target data element for an indicator, I don't know how simple it is to implement technically as both indicators and target data elements are subject to time and space.

At the end, I think what is mostly needed is to show/compare the actual & target values next to each other easily in pivot tables and/or reports.

Thanks for your feedback. Looking forward to other people's comments.


Maheed



From: Bob Jolliffe [mailto:bobjolliffe@xxxxxxxxx]
Sent: Tuesday, September 02, 2014 11:59 AM
To: Maheed Ramadan
Cc: DHIS 2 Developers list; dhis2-users
Subject: Re: [Dhis2-devs] Requirement related to indicators - Target vs achieved

Hi Maheed

I am sure you will get much more informed feedback on this question because I think its something many folk grapple with.  Just a quick observation regarding implementation.


I don't know if anybody have similar issues or requirements like this one.
I think 2 thinks are missing here:

1-      Cannot define the target value for a data element within the data element itself. It should be defined using another data element

2-      The target value are not time-bound. This means we cannot set that the target value should be reached starting date X and ending date Y.
These two seem to be contradictory.  Associating a target value with a data element sounds sensible enough except that it would then lose the sense of target time.  So you need to use another dataelement. ( I am not sure whether that is in fact what you are saying with 1 or you are pointing out what is missing).  Note the same time boundedness can also be true of space.  It is possible that different countries, provinces, districts set different target values.

Do you think it makes more sense to focus on target values for indicators than dataelements?  (Oddly enough I imagine the measured value should be an indicator and the target value should be a dataelement because you'd want to enter these targets directly).

So maybe what is missing is the ability to assign a target dataelement to an indicator definition.  So an indicator can have (an optional) dataelement associated with it for its target values.  Of course you can do this "manually" but, as you say, the reports become laborious.  Having an explicit association might make it easier to design smart reports.

Look forward to hear what other people's experience is.