dhis2-users team mailing list archive

[Johan Sæbø <johansa@xxxxxxxxxx>]

Tom,

> If this is the wrong forum for this, please just ignore.

This is the perfect forum for this!

> I have the issue, as many must, of different users having different levels of aggregation they could report. For example, one user might have cases by sex, but another user might only have the total number of cases. I could just ask for total cases, but it's nice to have the disaggregation where available. I've come down to two options for dealing with this.
>
> 1. Create two variables where users can report the same cases. If users don't have the disaggregation they just leave them blank. While creating some seeming redundancy in the database and some double entry, it does also provide an opportunity for validation.

One data element for the total, and one with disaggregation (or two 
other data elements, one for each gender)? This would for pratical 
purposes not be different from the proposal below then, if I'm not 
mistaken, just more cumbersome.

> 2. Create one variable with the disaggregating dimensions and have one of the dimensions be "unknown", i.e. "sex unknown". The potential downside of this is one could imagine some lazy users (I've been one) who just fill in unknown rather than bother hunting down the disaggregation.

This is the case for CAREC, covering many countries in the caribbean. 
Some countries will have data disaggregated by sex and age groups, some 
will have for only one of the dimensions, and maybe some will not have 
any disaggregation. In their case, they have the category options "<1, 
1-4 ... Unknown Age", and "Male, Female, Unknown Gender". This will be 
better for analysis as all data will be "within" one data element, for 
all countries regardless of the disaggregation they are using. The 
automatic totalling of the three genders will be comparable without 
having to pick the right data element. As with many things in DHIS2, 
this solution can be mimicked in different ways, but I think one data 
element with three genders is perhaps easiest.

> Does anyone have experience pro or against either method? Is there another method I didn't think of that's better?

Maybe someone from CAREC can chip in here? They've had the three-gender 
model for years (although on paper, excel etc)

Johan

> Thanks,
>
> Tom
>
> -----Original Message-----
> From: dhis2-users-bounces+hiattt=wpro.who.int@xxxxxxxxxxxxxxxxxxx [mailto:dhis2-users-bounces+hiattt=wpro.who.int@xxxxxxxxxxxxxxxxxxx] On Behalf Of Johan Sæbø
> Sent: 02 May 2012 17:42
> To: dhis2-users@xxxxxxxxxxxxxxxxxxx
> Subject: Re: [Dhis2-users] [Dhis2-devs] DHIS version 2.8 released
>
> Hi Tom,
>
> for calculating the overlapping categories, you can define them as
> indicators.
>
> For example; Data Element "Population", with catcombo age+sex, would be
> used for M/F<1, 1-4, 5-14, 15-25 etc
>
> Then for any combo of these, like<5 or<14, make an indicator with the
> expression: "Population<1"+"Population 1-4" + etc. You can add whatever
> you want, to construct your indicators<5 Male,<5 Female,<5 Total etc.
> Then you have these indicators to use whenever you would like to,
> without the need to calculate them each time.
>
> Johan
>
> On 02.05.2012 11:03, Hiatt, Mr Tom (WPRO) wrote:
> > Thanks, Ola. That's very helpful to know what to consider when
> > structuring this. I'm using UN population data which up to now they've
> > made aggregate nicely. I have dropped categories that overlap figuring I
> > can calculate them later (e.g. children 0-14 years). So I think it
> > should be okay.
> >
> > Tom
> >
> > *From: * olatitle@xxxxxxxxx [mailto:olatitle@xxxxxxxxx] *On Behalf Of
> > *Ola Hodne Titlestad
> > *Sent:* 02 May 2012 16:16
> > *To:* Lars Helge Øverland
> > *Cc:* Hiatt, Mr Tom (WPRO); dhis2-users@xxxxxxxxxxxxxxxxxxx
> > *Subject:* Re: [Dhis2-users] [Dhis2-devs] DHIS version 2.8 released
> >
> > Hi Tom,
> >
> > If I understand you correctly you have a data element called
> > "Population" and a categorycombination with Age+Sex linked to this data
> > element?
> >
> > If so, you should make sure that the total of the category option
> > combinations (All ages and sex combinations) add up to the total
> > population.
> >
> > The total for the data element is used in many places in DHIS 2 and if
> > this total is not the correct total population you might confuse the
> > users and also make indicator and report creation more complicated than
> > it has to be.
> >
> > Typically the age groups for population is overlapping and not complete
> > sets (e.g.<1,<5, female 15-49 (WRA), total), that is why I ask.
> >
> > If you have a complete set of disaggregated population data then
> > categories will work fine.
> >
> > If not, I would use separate data elements for the different population
> > groups.
> >
> > Ola
> >
> > -------
> >
> > ----------------------------------
> > Ola Hodne Titlestad (Mr)
> > HISP
> > Department of Informatics
> > University of Oslo
> >
> > Mobile : +47 48069736
> > Home address: Vetlandsvn. 95B, 0685 Oslo , Norway . Googlemaps link
> > <http://maps.google.com/maps?f=q&source=s_q&hl=en&geocode=&q=Vetlandsvn.+95B,+0685+Oslo,+Norway>
> >
> > On 1 May 2012 14:20, Lars Helge Øverland<larshelge@xxxxxxxxx
> > <mailto:larshelge@xxxxxxxxx>>  wrote:
> >
> > On Tue, May 1, 2012 at 2:15 PM, Hiatt, Mr Tom (WPRO)
> > <hiattt@xxxxxxxxxxxx<mailto:hiattt@xxxxxxxxxxxx>>  wrote:
> >
> > Ah yes, of course. Thanks!
> >
> > For me, I have individual codes for these data elements (population by
> > age and sex), but previously I had them stored as individual variables
> > rather than dimensions of the population variable. This time around I'm
> > trying to store them in the "right" way as dimensions, which seems more
> > logical. (I hope I don't get bit down the road for this.) For my use so
> > far, this is a one-off data import exercise, but I could imagine
> > scenarios where one would like to import more frequently from a database
> > that stores them as individual variables as mine did. However, seeing as
> > how these data would probably come from different systems that use
> > different coding schemes, perhaps a mapping exercise would be necessary
> > anyway. I'll let you know if I run into that hurdle.
> >
> > Tom
> >
> > Great. Let us know how it goes.
> >
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~dhis2-users
> > Post to : dhis2-users@xxxxxxxxxxxxxxxxxxx
> > <mailto:dhis2-users@xxxxxxxxxxxxxxxxxxx>
> > Unsubscribe : https://launchpad.net/~dhis2-users
> > More help : https://help.launchpad.net/ListHelp
> >
> > This message was scanned for viruses with Trend Micro ScanMail, GFI
> > MailSecurity and GFI MailEssentials by the World Health Organization
> > Regional Office for the Western Pacific. However, the recipient is
> > advised to scan this e-mail and any attached files for viruses.
> >
> > Disclaimer:
> >
> > This e-mail, together with any attachments, is intended for the named
> > recipients only and is confidential. It may also be privileged or
> > otherwise protected by law.
> >
> > If you have received it in error, please notify the sender immediately
> > by reply e-mail and delete it and any attachments from your system. You
> > may not copy or disclose its contents to anyone.
> >
> >
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~dhis2-users
> > Post to     : dhis2-users@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~dhis2-users
> > More help   : https://help.launchpad.net/ListHelp
>
>
> _______________________________________________
> Mailing list: https://launchpad.net/~dhis2-users
> Post to     : dhis2-users@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~dhis2-users
> More help   : https://help.launchpad.net/ListHelp
>
>
>
>
> ====================================================
> This message was scanned for viruses with Trend Micro ScanMail, GFI MailSecurity and  GFI MailEssentials by the World Health Organization Regional Office for the Western Pacific.  However, the recipient is advised to scan this e-mail and any attached files for viruses.
>
> Disclaimer:
>
> This e-mail, together with any attachments, is intended for the named recipients only and is confidential. It may also be privileged or otherwise protected by law.
>
> If you have received it in error, please notify the sender immediately by reply e-mail and delete it and any attachments from your system. You may not copy or disclose its contents to anyone.