Dealing with Updated/Expired data elements/indicators

Hi,

Hope all are well.

I can envisage a situation where a given indicator/data element is changed in some way (perhaps a changed calculation or definition). For historical reporting purposes we would want to keep the old version of the indicator/data element but would prefer that it is not apparent for standard users that are building new datasets.

Does anyone have any experience of this that they could share - how they went about dealing with it?

Have thought about a group for active/inactive items or perhaps changing the name to indicate it is inactive, but not sure what would be the best approach.

Would be good to hear your experiences.

Cheers

Craig

Hi Craig,

In South Africa we have a Indicator dataset review process every 2 years and the revision is released officially to all users so it is largely a transparent process

The outcome usually results in a mixture of the following:

  1. Indicators and their related elements being dropped off - these are moved to an inactive group and ensure that they are removed from used datasets. The indicators simply don’t calculate anymore so that is not really a problem. You can also control access to these by restricting sharing settings but that might affect access to historic reporting.

  2. Indicators and their related data elements name changes. These are simply matched with the old name/renamed and data collection tools adapted. Since the finalised review document is shared this is a transparent process.

  3. New indicators and data elements are simply commenced

  4. Changed indicator formulas, we don’t do that often but have a few such scenarios and the solution depends on the change E g we had an indicator with a constant factor of 1.07 in a population element pre-inserted and now want to change it to 1.05 but don’t want to affect the previous indicator values. What we doing is to create a data element and pre-inserting the value x 1.07 until period xx and x 1.05 after period xx. For some other indicators we rename them to (old) but it can be tricky if some of the data elements are still being collected.

Hope this helps.

Elmarie Claasen

HISP-SA

···

Hi,

Hope all are well.

I can envisage a situation where a given indicator/data element is changed in some way (perhaps a changed calculation or definition). For historical reporting purposes we would want to keep the old version of the indicator/data element but would prefer that it is not apparent for standard users that are building new datasets.

Does anyone have any experience of this that they could share - how they went about dealing with it?

Have thought about a group for active/inactive items or perhaps changing the name to indicate it is inactive, but not sure what would be the best approach.

Would be good to hear your experiences.

Cheers

Craig

Hi Craig,

Why not use the “sharing” feature? Set the indicator’s Public sharing to “None”. It would still be present, but not visible to anyone. Then you could create a user group “Legacy indicators user group” and add users to this group, and set the sharing of the indicator to allow this group to see it.

Regards,

Jason

···

On Tue, Nov 17, 2015 at 4:17 AM, Craig Hollingsworth craig.hollingsworth@nrc.no wrote:


Mailing list: https://launchpad.net/~dhis2-users

Post to : dhis2-users@lists.launchpad.net

Unsubscribe : https://launchpad.net/~dhis2-users

More help : https://help.launchpad.net/ListHelp

Hi,

Hope all are well.

I can envisage a situation where a given indicator/data element is changed in some way (perhaps a changed calculation or definition). For historical reporting purposes we would want to keep the old version of the indicator/data element but would prefer that it is not apparent for standard users that are building new datasets.

Does anyone have any experience of this that they could share - how they went about dealing with it?

Have thought about a group for active/inactive items or perhaps changing the name to indicate it is inactive, but not sure what would be the best approach.

Would be good to hear your experiences.

Cheers

Craig

Jason P. Pickering
email: jason.p.pickering@gmail.com
tel:+46764147049