We have a peculiar aggregation need here. I don’t know if any other country uses it or it is only Ethiopia but here it is.
Consider this data element which is collected monthly:
**Number of adults and children who are currently on ART - ****First Line Regimen **
When analyzing data for a year, the last month’s data shall always be taken rather than the summation or average of the data values.
There are a few data elements like this we capture every month.
However, the current aggregation operators cannot handle this. The max and the min operators may be the closest ones to handle this (for data that increases and decreases over time respectively).
Would it be possible to add one more aggregation operator that always takes the last period’s value when the aggregation is across periods and when it is hierarchically it sums the last period’s values of each captured data.
Hi Seid,
This request has come up a few times, in slightly different forms. Currently, there is no such operator, although it would be useful. How we have handled it in other implementations is through the use of a custom script which would populate an additional data element (for instance **Number of adults and children who are currently on ART - ****First Line Regimen (Last observed) ) **and populate this with the last observed data for a given time period. The script could be automated to run regularly, so that the data was always up to date with the last observed value So, if you are looking at yearly data, you would look back over the last year, and determine which value had been observed “last”, and use this to populate your new data element (which would have a period type of year).
You would probably need to use a combination of SQL queries combined with some scripting language (Python, NodeJS, R, etc) to pull the required data and then submit it back through the API.
We have a peculiar aggregation need here. I don’t know if any other country uses it or it is only Ethiopia but here it is.
Consider this data element which is collected monthly:
**Number of adults and children who are currently on ART - ****First Line Regimen **
When analyzing data for a year, the last month’s data shall always be taken rather than the summation or average of the data values.
There are a few data elements like this we capture every month.
However, the current aggregation operators cannot handle this. The max and the min operators may be the closest ones to handle this (for data that increases and decreases over time respectively).
Would it be possible to add one more aggregation operator that always takes the last period’s value when the aggregation is across periods and when it is hierarchically it sums the last period’s values of each captured data.