Hello Community!
I've got a problem with achieving the correct value for all rows using a Fixed LOD. In this snippet, you'll see the calculation I'm using. It works just fine for Northwestern, but the % is not the same for Presence. In this Sheet 15, I need numbers to match the SUM(weight) delta pill. I can't use this simple table calculation in the viz, because I want to be able to filter hospitals in the viz...Hence a LOD. I also show a Max Column Weight which is just WINDOW_SUM(SUM([weight])). I have this to illustrate that it matches Total(SUM(weight)), but again, I can't use this because of the filtering functionality. Ideally, I'd put Window_Max(SUM(Weight)) after the : in the LOD, but due to aggregation, that doesn't work. Can someone please help me solve this issue please?
Best,
Matthew

See attached.
One thing I did was change ATTR(question_desc) on ROWS to be a dimension.
I also added the [Hosp or Sys] dimension to the FIXED SUM calc. And that gives us the same result as simply SUM([Weight]). I continued using the [FIXED SUM Weight...] calc, but eventually realized that it was redundant.
I added a [SUM Weight by datyear] calc. This is a WINDOW_SUM. On 15 (2), It's the default of TABLE(across) and it gives us trash. On 15(3) it's TABLE(down), which adds down each [Datyear] column. There's your number.
Notice that it populates the valus on every row, even if there is no SUM(weight). (You can edit the calc to change that.)
But I also did it with LOD. See [SUM LOD Weight...]. It's not a FIXED, though. Because you want it to obey the filters, I used EXCLUDE, and listed every dimension on the sheet -- except for [datyear]. That will tell tableau to sum for each datyear on the sheet.
In the TEXT editor I put a little label on these last two values so you can see what got displayed for the table calc and what was displayed for the LOD.
Upshot: We really don't need the FIXED SUM calc I made. Sum(weight) is good enough. And we don't need the table calc. The EXCLUDE LOD does the job, and only displays on rows where there is a SUM(weight) in the first place.