In NRC we are going to try to import some historic data from excel into our DHIS2 instance (global data for 2013-2015). We have matching data element codes but we don’t have the same disaggregation in our historic data. For example we only used (in pre DHIS times) to capture kg’s of food distributed. Now we disaggregate food into the various types of food.
I am wondering if anyone has any tips on the best way to import historic data in such a situation. I have a couple of ideas in mind:
The ideal would be to be able to upload data onto the prime UiD of the data element (without the “.categoryoptionuid). Such that if I do pivot on any of the disaggregations I don’t get back the historic data, which lacks this information, but without breaking it down I do get the comparative totals.
E.g. like this:
···
Kg’s of food distributed
2015
15 kg
2016
22 kg
With disaggregation:
Kg’s of food distributed
Potatoes
Strawberries
2015
2016
20 kg
2 kg
Does this work and is there any potential issues in doing this?
If the above doesn’t work. I’m considering to add a temporary category option of “historic” into each category. But preferably I do not want this to ever show up in any future data set form or be added data to later. Does the “start date” and “end date” serve this purpose?
my suggestion here would be to simply import the data without specifying any category option combo at all in the JSON/XML. The system will then assign the data to the “default” option combo (meaning no disaggregation). This should give you the behavior you outline in alternative 1.
In NRC we are going to try to import some historic data from excel into our DHIS2 instance (global data for 2013-2015). We have matching data element codes but we don’t have the same disaggregation in our historic data. For example we only used (in pre DHIS times) to capture kg’s of food distributed. Now we disaggregate food into the various types of food.
I am wondering if anyone has any tips on the best way to import historic data in such a situation. I have a couple of ideas in mind:
The ideal would be to be able to upload data onto the prime UiD of the data element (without the “.categoryoptionuid). Such that if I do pivot on any of the disaggregations I don’t get back the historic data, which lacks this information, but without breaking it down I do get the comparative totals.
E.g. like this:
Kg’s of food distributed
2015
15 kg
2016
22 kg
With disaggregation:
Kg’s of food distributed
Potatoes
Strawberries
2015
2016
20 kg
2 kg
Does this work and is there any potential issues in doing this?
If the above doesn’t work. I’m considering to add a temporary category option of “historic” into each category. But preferably I do not want this to ever show up in any future data set form or be added data to later. Does the “start date” and “end date” serve this purpose?
my suggestion here would be to simply import the data without specifying any category option combo at all in the JSON/XML. The system will then assign the data to the “default” option combo (meaning no disaggregation). This should give you the behavior you outline in alternative 1.
In NRC we are going to try to import some historic data from excel into our DHIS2 instance (global data for 2013-2015). We have matching data element codes but we don’t have the same disaggregation in our historic data. For example we only used (in pre DHIS times) to capture kg’s of food distributed. Now we disaggregate food into the various types of food.
I am wondering if anyone has any tips on the best way to import historic data in such a situation. I have a couple of ideas in mind:
The ideal would be to be able to upload data onto the prime UiD of the data element (without the “.categoryoptionuid). Such that if I do pivot on any of the disaggregations I don’t get back the historic data, which lacks this information, but without breaking it down I do get the comparative totals.
E.g. like this:
Kg’s of food distributed
2015
15 kg
2016
22 kg
With disaggregation:
Kg’s of food distributed
Potatoes
Strawberries
2015
2016
20 kg
2 kg
Does this work and is there any potential issues in doing this?
If the above doesn’t work. I’m considering to add a temporary category option of “historic” into each category. But preferably I do not want this to ever show up in any future data set form or be added data to later. Does the “start date” and “end date” serve this purpose?
