Additional data dimensions¶
About additional data dimensions¶
DHIS2 has the ability to add dimensions to data in addition to what was described in the previous chapter. We will call these dimensions “attribute categories” (ACs). The categories described in the previous chapter we will call “disaggregation categories” (DCs) to differentiate them from ACs.
ACs and DCs are quite similar—they work in much the same way, are accessed through the same part of the maintenance interface, and exist in the same part of the database. The main difference between them is what they are connected to. A DC is attached to a data element; however, an AC is attached to a data set. This means values for all DC options can be entered on the same data entry screen, whereas you must choose the AC option before you begin to enter data.
In setting up a system, you could just use DCs and ignore ACs altogether. However, ACs are a way to simplify data entry screens or reduce the size of the cross-product of category option combos.
Tip
When you’re deciding which categories should be DCs and which should be ACs, here’s a good rubric:
Use DCs when you want to use different combinations of categories on different data elements within a data set
Use DCs when you want to enter all the category option combinations on one data entry screen
Use ACs when you want to use the same combination of categories for all the data in a data set
Use ACs when you want to enter only one category option combination on one data entry screen
While we referred to DCs as part of the what dimension for simplicity in the former chapter, it’s actually more complex. Either DCs or ACs can answer any question about a data element, including what (of course), who, why, how, or even a where or a when beyond the organisation unit and period dimensions.
Create or edit an attribute category and its options¶
The process of creating an attribute category as well as its options and combinations, is discussed in the section Manage categories. As noted there, disaggregation categories are configured by editing a data element and attribute categories are configured by editing a data set.
Data entry with disaggregation categories and attribute categories¶
When entering aggregate data, one must first choose the attribute categories, and then one can enter the data across disaggregation categories on a single page.
For instance, in the graphic below, the attribute categories are Implementing Partner (AIDSRelief Consortium) and Project (Improve access to medicines). The disaggregation categories are gender (male/female/etc.), age (<15, 15-24, 25-49, >49).
Analysis with disaggregation categories and attribute categories¶
In order to do analysis with disaggregation and attribute categories, check the “Data dimension” box in the category editing screen of the Maintenance app, as discussed in Create or edit a category.
Approvals with attribute categories¶
To include attribute categories in approvals, create a category option group that contains the same category options as the attribute category. Then create a category option group set and add that the category option group set as a data approval level.
This is covered in more detail in the section Approving by category option group set and Approving by multiple category option group sets.
Attribute categories and the datavalue table¶
For some, the way attribute categories work with disaggregation categories is clearer when we look at how the data values are stored in DHIS2’s database. If diving into the database internals doesn’t help you understand how the different types of categories work together, please feel free to ignore it.
Each data value is associated with a data element, a period, and an organisation unit, which are represented in this way:
dataelementid | periodid | sourceid |
|
|
|
(Note these are numeric database ids, not DHIS2 uids.)
And of course, each data value has a value, adding the value column to the database:
dataelementid | periodid | sourceid | value |
|
|
Each data value also references the disaggregation category options and the attribute category options assigned to it. For instance, in the example above, the data value entered in the box “Male <15” will have the option “Male” for the disaggregation category gender, and the option “<15” for the disaggregation category age. The combination of these two options is represented in the database by a single category option combination meaning “Male, <15”. The data value references this disaggregation category option combination in the categoryoptioncomboid:
dataelementid | periodid | sourceid | value | categoryoptioncomboid |
|
Likewise, the same data value will have the option “AIDSRelief Consortium” for the category Implementing Partner, and the option “Improve access to medicines” for the category Project. There will also be a database category option combination meaning “Implementing Partner, Improve access to medicines”. The data value references this attribute category option combination in the attributeoptioncomboid:
dataelementid | periodid | sourceid | value | categoryoptioncomboid | attributeoptioncomboid |
Note
The above column list does not include all of the columns in the datavalue table.
If you have not defined a disaggregation category combination for a data element, the categoryoptioncomboid will reference a “default” category option combination, which is defined internally in DHIS2 as the category default with the option default (the only option in the default category). Likewise, if you have not defined an attribute category combination for the dataset in which you enter the data, the attributeoptioncomboid references the same “default” category option combination.
We hope this investigation of how data values are stored in DHIS2’s database makes it clearer how data values can be associated with various groups of category options, both from disaggregation categories and attribute categories.