The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
Hi powerbi community
I'm trying to do count on columns, in the case i get a match on certain criteria, but my measure is ignoring the global context (as in, if i filter on the criteria itself on the case where i should get no values, my result stays the same instead of 0.
I've been able to recreate the problem with smaller data:
data | id |
red | 1 |
orange | 2 |
red | 3 |
orange | 4 |
red | 5 |
yellow | 6 |
blue | 7 |
orange | 8 |
blue | 9 |
orange | 10 |
Here "Data" and "id" are read as text.
The measures are :
The expected result is that if I create a slicer on "ID" and select anything but "3", i should get 0.
But here is the result i get :
I noticed that "Measure 3" returns 1 whenever i choose an id with "Data" value "red".
Could someone explain me this behaviour and how to get the expected result please ?
Thank you
Solved! Go to Solution.
@Anonymous , Try like
Measure 3 = CALCULATE([Measure2], filter(aTest, aTest[id] = "3"))
refer if needed
http://dataap.org/blog/2019/04/22/difference-between-calculate-with-and-without-filter-expression/
@Anonymous , Try like
Measure 3 = CALCULATE([Measure2], filter(aTest, aTest[id] = "3"))
refer if needed
http://dataap.org/blog/2019/04/22/difference-between-calculate-with-and-without-filter-expression/
User | Count |
---|---|
80 | |
74 | |
41 | |
30 | |
28 |
User | Count |
---|---|
107 | |
96 | |
53 | |
47 | |
47 |