Filtering across multiple dimensions

Hi Brill

Thank you for reaching out to the Microsoft Fabric Forum Community.

After reviewing the details you provided, I was able to reproduce the scenario, I have used it as sample data on my end and successfully implemented it.   

Dax Measure:  

ValidCustomerSatisfaction = 
VAR SelectedSurveys =
    VALUES('SurveyResults'[SurveyName])
VAR NumSelected =
    COUNTROWS(SelectedSurveys)
RETURN
CALCULATE (
    AVERAGE('SurveyResults'[SatisfactionScore]),
    FILTER (
        'SurveyResults',
        VAR CurrentCustomer = 'SurveyResults'[CustomerID]
        VAR CustomerSurveys =
            CALCULATETABLE (
                VALUES('SurveyResults'[SurveyName]),
                'SurveyResults'[CustomerID] = CurrentCustomer
            )
        RETURN
            COUNTROWS (
                INTERSECT (SelectedSurveys, CustomerSurveys)
            ) = NumSelected
    )
)

Attached the PBIX file for reference.

If this post helps, then please give us ‘Kudos’ and consider Accept it as a solution to help the other members find it more quickly. 
 
Thank you

You could create a calculation group so that users could choose between showing customers who have completed any of the selected surveys or only those who have completed all the selected surveys.

The calculation item for any of the selected surveys would just use SELECTEDMEASURE(), the item for showing just those customers who have completed all the selected surveys would be

All Selected Surveys = 
VAR NumChosen = COUNTROWS( ALLSELECTED( 'Survey Type'[Survey Type] ) )
VAR CustomersAndNumSurveys = ADDCOLUMNS(
    VALUES( Customer[Customer] ),
    "@num rows", CALCULATE( COUNTROWS( 'Table' ) )
)
VAR ValidCustomers = FILTER( CustomersAndNumSurveys, [@num rows] = NumChosen )
VAR Result = CALCULATE( SELECTEDMEASURE(), ValidCustomers )
RETURN Result

By implementing this in a calculation group it will work for any measure, e.g. counting the number of customers, showing average scores etc.

You can see an example in the attached file.