Teenage Pregnancy DataSet for Schools

Support - Assistance technique
1

Dear All,
I created a Teenage Pregnancy dataSet whose data should be captured
from schools. Since we have multiple schools per Chiefdom; i created a
data element 'school' in the form of a text string. During data entry,
when a school had been entered for a particular chiefdom, it don't
accept another school again.

Please can someone help.

···

--
Regards,

Gerald

2

Gerald,

I would not recommend using the standard HMIS for school health - it would give you minimal control and reporting capabilities per school.

In South Africa, school health is captured into a separate DHIS2 database where the hierarchy ends with schools and not health facilities (although each school might store a reference health facility, of course). All the ~30,000 schools in the country is part of that orghierarchy, and we generally try to get annual data for the number of learners (pupils) per grade, which you can then use as denominator data to get coverage indicators.

You COULD of course add the schools into the HMIS orghierarchy, but IMHO that is rather messy - better to use a separate database

Regards
Calle

···

On 15 November 2017 at 08:30, gerald thomas gerald17006@gmail.com wrote:

Dear All,

I created a Teenage Pregnancy dataSet whose data should be captured

from schools. Since we have multiple schools per Chiefdom; i created a

data element ‘school’ in the form of a text string. During data entry,

when a school had been entered for a particular chiefdom, it don’t

accept another school again.

Please can someone help.

–

Regards,

Gerald

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

–

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19119

Email: calle.hedberg@gmail.com

Skype: calle_hedberg

3

Dear Calle,
Thanks for your response. That was what i told the partners but their
budget didn't permitted them and they didn't have names for all the
schools. How does comments works on DataSet? Because we can use the
option of comments to tag names to a particular Data Entry because the
name of the schools are for reference purposes.

···

On 11/15/17, Calle Hedberg <calle.hedberg@gmail.com> wrote:

Gerald,

I would not recommend using the standard HMIS for school health - it would
give you minimal control and reporting capabilities per school.

In South Africa, school health is captured into a separate DHIS2 database
where the hierarchy ends with schools and not health facilities (although
each school might store a reference health facility, of course). All the
~30,000 schools in the country is part of that orghierarchy, and we
generally try to get annual data for the number of learners (pupils) per
grade, which you can then use as denominator data to get coverage
indicators.

You COULD of course add the schools into the HMIS orghierarchy, but IMHO
that is rather messy - better to use a separate database

Regards
Calle

On 15 November 2017 at 08:30, gerald thomas <gerald17006@gmail.com> wrote:

Dear All,
I created a Teenage Pregnancy dataSet whose data should be captured
from schools. Since we have multiple schools per Chiefdom; i created a
data element 'school' in the form of a text string. During data entry,
when a school had been entered for a particular chiefdom, it don't
accept another school again.

Please can someone help.

--
Regards,

Gerald

_______________________________________________
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

--

*******************************************

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19119

Email: calle.hedberg@gmail.com

Skype: calle_hedberg

*******************************************

--
Regards,

Gerald

4

Gerald,

OK, so there is no intention or requirement to track and analyze such data per school, then. If the idea is simply to somehow store a list of schools from where the data originates, and assuming that you are collecting it per chiefdom and per month, I would rather change the data element you have already created into a long text field. Capturers can then add school names in there, separated by a semi-colon or similar delimiter.

Just make sure everybody understands that if the data per chiefdom per month has not already been aggregated in the input form, then such aggregation must be done manually during data entry.

Note also that you do have one other possibility, and that is to treat each school report received as an event - or in other words you set this up as an Event Program - without registration, presumably - where the school itself it the entity.

Regards

Calle

···

On 15 November 2017 at 11:45, gerald thomas gerald17006@gmail.com wrote:

Dear Calle,

Thanks for your response. That was what i told the partners but their

budget didn’t permitted them and they didn’t have names for all the

schools. How does comments works on DataSet? Because we can use the

option of comments to tag names to a particular Data Entry because the

name of the schools are for reference purposes.

On 11/15/17, Calle Hedberg calle.hedberg@gmail.com wrote:

Gerald,

I would not recommend using the standard HMIS for school health - it would

give you minimal control and reporting capabilities per school.

In South Africa, school health is captured into a separate DHIS2 database

where the hierarchy ends with schools and not health facilities (although

each school might store a reference health facility, of course). All the

~30,000 schools in the country is part of that orghierarchy, and we

generally try to get annual data for the number of learners (pupils) per

grade, which you can then use as denominator data to get coverage

indicators.

You COULD of course add the schools into the HMIS orghierarchy, but IMHO

that is rather messy - better to use a separate database

Regards

Calle

On 15 November 2017 at 08:30, gerald thomas gerald17006@gmail.com wrote:

Dear All,

I created a Teenage Pregnancy dataSet whose data should be captured

from schools. Since we have multiple schools per Chiefdom; i created a

data element ‘school’ in the form of a text string. During data entry,

when a school had been entered for a particular chiefdom, it don’t

accept another school again.

Please can someone help.

–

Regards,

Gerald

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

–

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19119

Email: calle.hedberg@gmail.com

Skype: calle_hedberg

–

Regards,

Gerald

–

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19119

Email: calle.hedberg@gmail.com

Skype: calle_hedberg

5

Dear Calle,
I had already did as you suggested but i am face with a challenge of
attaching my category combination to the various data element.

Please i need help

···

On 11/15/17, Calle Hedberg <calle.hedberg@gmail.com> wrote:

Gerald,

OK, so there is no intention or requirement to track and analyze such data
per school, then. If the idea is simply to somehow store a list of schools
from where the data originates, and assuming that you are collecting it per
chiefdom and per month, I would rather change the data element you have
already created into a long text field. Capturers can then add school names
in there, separated by a semi-colon or similar delimiter.

Just make sure everybody understands that if the data per chiefdom per
month has not already been aggregated in the input form, then such
aggregation must be done manually during data entry.

Note also that you do have one other possibility, and that is to treat each
school report received as an event - or in other words you set this up as
an Event Program - without registration, presumably - where the school
itself it the entity.

Regards
Calle

On 15 November 2017 at 11:45, gerald thomas <gerald17006@gmail.com> wrote:

Dear Calle,
Thanks for your response. That was what i told the partners but their
budget didn't permitted them and they didn't have names for all the
schools. How does comments works on DataSet? Because we can use the
option of comments to tag names to a particular Data Entry because the
name of the schools are for reference purposes.

On 11/15/17, Calle Hedberg <calle.hedberg@gmail.com> wrote:
> Gerald,
>
> I would not recommend using the standard HMIS for school health - it
would
> give you minimal control and reporting capabilities per school.
>
> In South Africa, school health is captured into a separate DHIS2
> database
> where the hierarchy ends with schools and not health facilities
> (although
> each school might store a reference health facility, of course). All
> the
> ~30,000 schools in the country is part of that orghierarchy, and we
> generally try to get annual data for the number of learners (pupils)
> per
> grade, which you can then use as denominator data to get coverage
> indicators.
>
> You COULD of course add the schools into the HMIS orghierarchy, but
> IMHO
> that is rather messy - better to use a separate database
>
> Regards
> Calle
>
> On 15 November 2017 at 08:30, gerald thomas <gerald17006@gmail.com> >> wrote:
>
>> Dear All,
>> I created a Teenage Pregnancy dataSet whose data should be captured
>> from schools. Since we have multiple schools per Chiefdom; i created a
>> data element 'school' in the form of a text string. During data entry,
>> when a school had been entered for a particular chiefdom, it don't
>> accept another school again.
>>
>> Please can someone help.
>>
>> --
>> Regards,
>>
>> Gerald
>>
>> _______________________________________________
>> 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
>>
>
>
>
> --
>
> *******************************************
>
> Calle Hedberg
>
> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>
> Tel/fax (home): +27-21-685-6472
>
> Cell: +27-82-853-5352
>
> Iridium SatPhone: +8816-315-19119
>
> Email: calle.hedberg@gmail.com
>
> Skype: calle_hedberg
>
> *******************************************
>

--
Regards,

Gerald

--

*******************************************

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19119

Email: calle.hedberg@gmail.com

Skype: calle_hedberg

*******************************************

--
Regards,

Gerald

6

Hi @Calle_Hedberg,

I see the post was three years ago.

I am interested to hear your input for my use case. I have built a tracker program in which data must be captured per school. My org unit hierarchy now goes from country to health facilities, i.e, country, province, district and health facility). On paper forms we are capturing data in the following fields: District (this is already in the org unit hierarchy) - administrative post - village - schools . As mentioned, in the tree we have just four levels; we do not have administrative posts - villages - schools and this is applicable to two districts only and not to the entire org unit. My question is: how should map these extra levels for the two districts knowing that data entry and analysis will be per school. I was thinking of creating administrative posts - villages - schools as attributes and assign them to the program but I do not know how this approach can affect data analysis.
Please provide advice on the best possible solution to my problem.

Looking forward to you sharing your expertise .

Thanks

Hernandez

7

@hernandezmachava ,

They way you describe the orghierarchy, I would have considered using a separate instance with the hierarchy Country, province, district, administrative post, village, school (i.e. 6 levels). BUT if you also want the pregnancy data to somehow be linked to the nearest health facility, it changes things.

  1. You say Tracker program - just confirm that the Tracked Entity Type is PERSON (i.e. the pregnant teenager).
  2. A village can be small - are you sure that each school only have learners from a single village? In most countries that would not be the case, one school would typically have learners from several villages. If yes, then the village you are interested in is basically part of the teenager’s home address.
  3. What is the relevance of the “administrative posts” here? What do they have to do with Teenage pregnancies, i.e. why do you want to analyse the data using those posts?
  4. How does health facilities relate to these pregnancies - do you want to register which health facility that is responsible for providing reproductive health / contraceptives to these teenagers, and/or which health facility that are responsible for their care during the pregnancy (or possibly Termination of the Pregnancy, if relevant/allowed)?

What I am basically saying is that you need to determine the importance of both health facility (facilities) and the school in terms of analysis and actions/interventions. Because you can use an orghierarchy ending with the school or the one ending with the health facility - and then add the other “geographical” relationship as one or more attributes.

Analysis based on the orghierarchy is always easier, since it is a clear-cut hierarchical and well-defined data set. But you can run analysis based on one or more attributes too, like admin posts & village & school, it’s just a bit more cumbersome. One advantage of having both village (aka home address) and school as just attributes (possible option sets if you have authoritative lists of both) is that you can have several villages for every school but also several schools per village, as indicated above. I do not even know which country this is and how your villages and schools are linked, so difficult for me to comment further…

My 2c worth
Calle

8

Hi @Calle_Hedberg,

Now having a separate instance may involve additional/ unexpected costs.
We have already a tracked entity type: person.

Regarding the text below, In our country one school has learners from several villages.

Administrative posts have no relevance to data analysis in our case.

Concerning this part

The scenario is this: health providers from the closest Health Facilities trained on Adolescent and youth sexual and reproductive health (AYSRH) provide counseling and services (including provision of contraceptives) at the school health corners. The information about these sessions is collected on paper forms and then must be entered in a tracker program in the DHIS 2. On paper forms apart from the village and school, what we also register is the health provider responsible for providing reproductive health/ contraceptives to teenagers, so the analysis is based on school and the health provider.

In the view of the situation just described above, what would be a pragmatic approach?

Thanks again.