Skip to main content
Announcements
Live today at 11 AM ET. Get your questions about Qlik Connect answered, or just listen in. SIGN UP NOW
cancel
Showing results for 
Search instead for 
Did you mean: 
RutiTaumanRubin
Creator
Creator

Set analysis intersection doesn't work

Hi All,

I want to create measure that in the denominator I want to count all the EMP_KEYs that have ACTION_TYPE "OPENED" AND "DELIVERED".

I try the following calc but it doesn't work:

1.

Avg(Aggr(
count (distinct {<EMP_KEY = P ({<ACTION_TYPE={"OPENED"}>} EMP_KEY ) >*<EMP_KEY = P ({<ACTION_TYPE={"DELIVERED"}>} EMP_KEY ) > } EMP_KEY )
/
count({<ACTION_TYPE={"DELIVERED"}>} Distinct EMP_KEY)
,ISSUE_ID))

 

2.

Avg(Aggr(
count({<EMP_KEY=P({<ACTION_TYPE={"OPENED"}>}EMP_KEY)>*<EMP_KEY=P({<ACTION_TYPE={"DELIVERED"}>}EMP_KEY)>} distinct EMP_KEY)
/
count({<ACTION_TYPE={"DELIVERED"}>} Distinct EMP_KEY)
,ISSUE_ID))

 

3. 

Avg(Aggr(
count({<ACTION_TYPE={"OPENED"}> * <ACTION_TYPE={"DELIVERED"}>} Distinct EMP_KEY)
/
count({<ACTION_TYPE={"DELIVERED"}>} Distinct EMP_KEY)
,ISSUE_ID))

 

Any Idea?

Thanks!!

 

 

 

 

 

 

5 Replies
Vegar
MVP
MVP

Try replacing your bold section with something like this.

count( distinct {<EMP_KEY={"=aggr(count({<ACTION_TYPE={OPENED,DELIVERED}>} distinct ACTION_TYPE),EMP_KEY)= 2"}>}  EMP_KEY)

RutiTaumanRubin
Creator
Creator
Author

Thanks for your response , it doesn't work - it bring wrong value..

marcus_sommer

Are you sure that you have really multiple values for ACTION_TYPE in regard to EMP_KEY and ISSUE_ID and the used dimensions in your object and the current selections. I suggest that you creates a tablebox with all these fields and to check their the real available data.

- Marcus

RutiTaumanRubin
Creator
Creator
Author

Thanks for your response..

Yes , I have data with both ACTION_TYPE, and also this calculation work for me in April version but right now I promote it to other environment with February versio its not working...

marcus_sommer

I would at first checking the data with one/several tableboxes on both sides - quite probably they aren't identically if it worked in one and not in the other. A very slight difference in a single loading/transforming step may cause different (and sometimes not very obvious) results.

- Marcus