Forum Discussion
DISTINCTCOUNT column Filtered by TRUE or FALSE
Hi All,
Column [Order] = Original Data Set, contains duplicates, type whole number.
Column [TECO] = type TRUE/FALSE, added in the query.
I need to FILTER column [TECO] for TRUE and then DISTINCTCOUNT column [Order] so that I have a unique count of [Order] = TRUE... no count of duplicate [Order].
Order TECO
| 401103218 | TRUE |
| 401103220 | FALSE |
| 401103222 | FALSE |
| 401103227 | TRUE |
| 401103471 | TRUE |
| 401103471 | TRUE |
This is the formula that I wrote:
I thought this was working but after checking it is returning a much higher number than it should. I have tried and continue to try various combinations of functions but it either returns an incorrect number or it is an illegal combination.
Any help appreciated.
- Anonymous7 years ago
OK... I apologise if you are looking into this issue for me... it's taken me all day but it was a mistake in my relationship model. Sorry!!!!
4 Replies
- AnonymousNot applicable
OK... I apologise if you are looking into this issue for me... it's taken me all day but it was a mistake in my relationship model. Sorry!!!!
- Gordonlilj
Solution Sage
Ah, no worries. I'm just glad you solved it
- Gordonlilj
Solution Sage
Hi,
Are you creating a measure or a calculated column?
If it's a calculated column, make it a measure instead
- AnonymousNot applicable
Hi Gordonlilj,
It is a measure not a column.
Regards,
Moby