[Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

image

Weekly

image

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Inline image 1

Weekly

image

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

···


Thanks
Preeti Ajgaonkar

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

image

image

image

···

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Weekly

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Weekly

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.


Thanks
Preeti Ajgaonkar

Thanks
Preeti Ajgaonkar

Hello Preeti,

We are facing the same thing in 2.28 where Week 48 (the week that intersects between Nov and dec) is actually aggregated/added in both Nov and Dec.

Maybe the dev team has some insight on this?

From our side, when the period of collection of data is weekly, we never compare monthly to weekly and only report either monthly or weekly but never the two at the same time; even though solving this could be extremely helpful into not repeating the intersecting week’s data in 2 months which would cause errors. This is not the case when the data is collected on a daily basis.

Looking forward to a possible answer/solution.

All the best,

Bernard

cid:image002.png@01D36F50.572585A0

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

image

image

image

image

···

From: Dhis2-users [mailto:dhis2-users-bounces+lib.ima=pu-ami.org@lists.launchpad.net] On Behalf Of Preeti Ajgaonkar
Sent: Friday, March 2, 2018 8:43 AM
To: DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net
Cc: Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Sonia Diverres sonia.diverres@paris.msf.org
Subject: Re: [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Inline image 1

Weekly

Inline image 3

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Inline image 1

Weekly

Inline image 4

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

Thanks

Preeti Ajgaonkar

Thanks

Preeti Ajgaonkar


Virus-free. www.avast.com

Hello all,

this was reported in Jira by Victor https://jira.dhis2.org/browse/DHIS2-2769

I see the issue is open but the pull request is merged (for 2.28 at least, not sure is was backported)… So I would suggest to check the most recent built.

Cheers,

Marta

image

image

image

image

image

···

On 2 March 2018 at 08:18, Bernard Sadaka (IM Advisor) lib.ima@pu-ami.org wrote:

Hello Preeti,

We are facing the same thing in 2.28 where Week 48 (the week that intersects between Nov and dec) is actually aggregated/added in both Nov and Dec.

Maybe the dev team has some insight on this?

From our side, when the period of collection of data is weekly, we never compare monthly to weekly and only report either monthly or weekly but never the two at the same time; even though solving this could be extremely helpful into not repeating the intersecting week’s data in 2 months which would cause errors. This is not the case when the data is collected on a daily basis.

Looking forward to a possible answer/solution.

All the best,

Bernard

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

From: Dhis2-users [mailto:dhis2-users-bounces+lib.ima=pu-ami.org@lists.launchpad.net] On Behalf Of Preeti Ajgaonkar
Sent: Friday, March 2, 2018 8:43 AM
To: DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net
Cc: Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Sonia Diverres sonia.diverres@paris.msf.org
Subject: Re: [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Weekly

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Weekly

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

Thanks

Preeti Ajgaonkar

Thanks

Preeti Ajgaonkar


Virus-free. www.avast.com


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

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

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

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

Yes indeed Marta,

The current build does solve that problem.

Thank you for highlighting this.

All the best,

Bernard

cid:image002.png@01D36F50.572585A0

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

image

image

image

image

image

···

From: Marta Vila [mailto:martavila@gmail.com]
Sent: Friday, March 2, 2018 10:26 AM
To: Bernard Sadaka (IM Advisor) lib.ima@pu-ami.org
Cc: Preeti Ajgaonkar apreeti@thoughtworks.com; DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net; Sonia Diverres sonia.diverres@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org
Subject: Re: [Dhis2-devs] [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Hello all,

this was reported in Jira by Victor https://jira.dhis2.org/browse/DHIS2-2769

I see the issue is open but the pull request is merged (for 2.28 at least, not sure is was backported)… So I would suggest to check the most recent built.

Cheers,

Marta

On 2 March 2018 at 08:18, Bernard Sadaka (IM Advisor) lib.ima@pu-ami.org wrote:

Hello Preeti,

We are facing the same thing in 2.28 where Week 48 (the week that intersects between Nov and dec) is actually aggregated/added in both Nov and Dec.

Maybe the dev team has some insight on this?

From our side, when the period of collection of data is weekly, we never compare monthly to weekly and only report either monthly or weekly but never the two at the same time; even though solving this could be extremely helpful into not repeating the intersecting week’s data in 2 months which would cause errors. This is not the case when the data is collected on a daily basis.

Looking forward to a possible answer/solution.

All the best,

Bernard

cid:image002.png@01D36F50.572585A0

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

From: Dhis2-users [mailto:dhis2-users-bounces+lib.ima=pu-ami.org@lists.launchpad.net] On Behalf Of Preeti Ajgaonkar
Sent: Friday, March 2, 2018 8:43 AM
To: DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net
Cc: Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Sonia Diverres sonia.diverres@paris.msf.org
Subject: Re: [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Inline image 1

Weekly

Inline image 3

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Inline image 1

Weekly

Inline image 4

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

Thanks

Preeti Ajgaonkar

Thanks

Preeti Ajgaonkar

Image removed by sender.

Virus-free. www.avast.com


Mailing list: https://launchpad.net/~dhis2-devs
Post to : dhis2-devs@lists.launchpad.net
Unsubscribe : https://launchpad.net/~dhis2-devs
More help : https://help.launchpad.net/ListHelp

Thanks a lot Marta for the update. The PR is merged in 2.28, but not backported to 2.27. It would be great if we can have the fix in 2.27 as well.

image

image

image

image

image

···

On Fri, Mar 2, 2018 at 1:56 PM, Marta Vila martavila@gmail.com wrote:

Hello all,

this was reported in Jira by Victor https://jira.dhis2.org/browse/DHIS2-2769

I see the issue is open but the pull request is merged (for 2.28 at least, not sure is was backported)… So I would suggest to check the most recent built.

Cheers,

Marta

On 2 March 2018 at 08:18, Bernard Sadaka (IM Advisor) lib.ima@pu-ami.org wrote:

Hello Preeti,

We are facing the same thing in 2.28 where Week 48 (the week that intersects between Nov and dec) is actually aggregated/added in both Nov and Dec.

Maybe the dev team has some insight on this?

From our side, when the period of collection of data is weekly, we never compare monthly to weekly and only report either monthly or weekly but never the two at the same time; even though solving this could be extremely helpful into not repeating the intersecting week’s data in 2 months which would cause errors. This is not the case when the data is collected on a daily basis.

Looking forward to a possible answer/solution.

All the best,

Bernard

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

From: Dhis2-users [mailto:dhis2-users-bounces+lib.ima=pu-ami.org@lists.launchpad.net] On Behalf Of Preeti Ajgaonkar
Sent: Friday, March 2, 2018 8:43 AM
To: DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net
Cc: Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Sonia Diverres sonia.diverres@paris.msf.org
Subject: Re: [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Weekly

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Weekly

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

Thanks

Preeti Ajgaonkar

Thanks

Preeti Ajgaonkar


Virus-free. www.avast.com


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

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

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

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

Thanks
Preeti Ajgaonkar

pfa

image

image

image

image

image

···

On Fri, Mar 2, 2018 at 3:12 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Thanks a lot Marta for the update. The PR is merged in 2.28, but not backported to 2.27. It would be great if we can have the fix in 2.27 as well.


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

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

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

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

On Fri, Mar 2, 2018 at 1:56 PM, Marta Vila martavila@gmail.com wrote:

Hello all,

this was reported in Jira by Victor https://jira.dhis2.org/browse/DHIS2-2769

I see the issue is open but the pull request is merged (for 2.28 at least, not sure is was backported)… So I would suggest to check the most recent built.

Cheers,

Marta

Thanks
Preeti Ajgaonkar

On 2 March 2018 at 08:18, Bernard Sadaka (IM Advisor) lib.ima@pu-ami.org wrote:

Hello Preeti,

We are facing the same thing in 2.28 where Week 48 (the week that intersects between Nov and dec) is actually aggregated/added in both Nov and Dec.

Maybe the dev team has some insight on this?

From our side, when the period of collection of data is weekly, we never compare monthly to weekly and only report either monthly or weekly but never the two at the same time; even though solving this could be extremely helpful into not repeating the intersecting week’s data in 2 months which would cause errors. This is not the case when the data is collected on a daily basis.

Looking forward to a possible answer/solution.

All the best,

Bernard

Bernard Sadaka

Information Management Advisor

Première Urgence - Aide Médicale Internationale

Nehmeh Building, Sanine Street, Chiyah, 3rd Floor

Ain el Remmaneh, Beirut, Lebanon

E-mail: lib.ima@pu-ami.org

Cell: +961 71 004 881 (208)

Skype: imapuami

Website: https://www.pu-ami.org

From: Dhis2-users [mailto:dhis2-users-bounces+lib.ima=pu-ami.org@lists.launchpad.net] On Behalf Of Preeti Ajgaonkar
Sent: Friday, March 2, 2018 8:43 AM
To: DHIS 2 Developers list dhis2-devs@lists.launchpad.net; DHIS 2 Users list dhis2-users@lists.launchpad.net
Cc: Evgenia ZELIKOVA evgenia.zelikova@paris.msf.org; twmsf twmsf@thoughtworks.com; Anceline DUPOUY Anceline.DUPOUY@paris.msf.org; Sonia Diverres sonia.diverres@paris.msf.org
Subject: Re: [Dhis2-users] [Dhis-devs][DHIS 2.27]Discrepancy in distribution of data in monthly reports as per EPI weeks

Apologies, the figures for expected and actual values for the months of Nov and Dec 2017 should be as follows

Expected

November 2017 - 495

December 2017 - 361

Actual

November 2017 - 398

December 2017 - 458

Thanks

Preeti A

On Fri, Mar 2, 2018 at 12:08 PM, Preeti Ajgaonkar apreeti@thoughtworks.com wrote:

Hi team,

Ever since we have upgraded to DHIS 2.27 and started using EPI weeks for distribution of data in any reports, we have noticed a discrepancy in the data for the months starting on a Friday.

For e.g. data for W44 to W48 should be attributed to Nov 2017 as per the EPI week rule, but W48 data it is attributed to Dec 2017 (which starts on a Friday). See screenshots below

Monthly

Weekly

Expected

November 2017 - 398

December 2017 - 458

Actual

November 2017 - 495

December 2017 - 361

Similar issue is observed from W31 - W35 where the data should for W35 should be counted as part of Aug 2017 but it appears in Sep 2017(which also starts on a Friday).

Monthly

Weekly

Expected

August 2017 - 614

September 2017 - 452

Actual

August 2017 - 499

September 2017 - 567

I am expecting the same thing to happen in Apr-May 2018.

Any leads on this would be appreciated.

Thanks

Preeti Ajgaonkar

Thanks

Preeti Ajgaonkar


Virus-free. www.avast.com


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

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

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

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

Navod Neranjan
B.Sc. (Honours) in Management and Information Technology
Department of Industrial management
University of Kelaniya
m:
+94717661555 skype : neranjan1992@hotmail.com
a:
No 188/1,Abeyaraja Mawatha,Matugama
e:
neranjanthi@gmail.com