Forum Discussion

91asma2's avatar
91asma2
Helper I
5 years ago
Solved

Distinct Count Two Measures

  • Anonymous's avatar
    Anonymous
    5 years ago

    Here's what the measures should look like:

     

    [Follow up] =
    COUNTROWS(
        FILTER(
            VALUES( Assessment[User ID] ),
            CALCULATE(
                DISTINCTCOUNT( Assessment[assessment_id] ) >= 2
            )
        )
    )
    
     
    [Total Users Training] =
    CALCULATE(
        DISTINCTCOUNT( 'Sessions'[User ID] ),
        KEEPFILTERS( 'Sessions'[Event Category] = "Training" )
    )
    
    [Users in Both] =
    var FollowUpUsers =
        FILTER(
            VALUES( Assessment[User ID] ),
            CALCULATE(
                DISTINCTCOUNT( Assessment[assessment_id] ) >= 2
            )
        )
    var TotalUsersTrainingUsers =
        CALCULATETABLE(
            DISTINCT( 'Sessions'[User ID] ),
            KEEPFILTERS( 'Sessions'[Event Category] = "Training" )
        )
    var Result =
        COUNTROWS(
            INTERSECT(
                FollowUpUsers,
                TotalUsersTrainingUsers
            )
        )
    RETURN
        Result

     

8 Replies

    • 91asma2's avatar
      91asma2
      Helper I

      Thank you, the article is helpful. But my model is correct and I know it easy but I am not sure how to approach it. What is your recommendation? 

      • Anonymous's avatar
        Anonymous
        Not applicable

        91asma2 

         

        Your model might be correct but I really suspect it's sub-optimal. Is your model a star-schema with conformed dimensions?

         

        By the way, your problem with that much information is not solvable right now. You have to show us the measures, (ideally) the model and what it means "to count the useres that exist in both measures." One can't write DAX in a vacuum...

  • Anonymous's avatar
    Anonymous
    Not applicable

    Can you please mark the answer as the solution? This helps me and others who will be looking for something smilar... Thanks.