Closed krpaulson closed 4 years ago
We had a discussion about this at one point where the aggregation function should only return the aggregates since that is more what you'd expect from the function.
Its only one extra line for the user to append on to the original dataset right?
One use case I see is when we have national and subnational data already. But then aggregate the subnational data to the national level data as well just to check that everything adds up.
The user would have to somehow distinguish between the original data and the aggregated subnational data by adding something like an aggregate
column. I was thinking originally the user would just have to figure this out themselves but maybe it is useful to add an option to append back onto the original dataset.
Ok, that sounds fine to make users append.
For example, if you aggregate ages <1, 1, 2, 3, 4 to <5, should the function return the input age groups or just <5? Would be useful to have the option to do either.