The opposite filter of ALLEXCEPT - exclude one filter in measure

[carinat]

I'm trying to create a measure that gives me the percentage within groups. The following formula is working like a charm until I start slicing or cross-filtering:

Measure = DIVIDE(COUNT(data[Respondent]),CALCULATE(DISTINCTCOUNT(data[Respondent]),ALL(data)))

What I would like is a function that only ignores one column in my data when filtering, i.e. only the column which contains the categories I'm trying to create a percentage of. I want kind of a %GT effect but one that amounts to more than 100% (therefore, I don't use the build-in quick calc at the graph level) since the respondents can have selected several categories.

Any suggestions?

[Anonymous]

calculate( COUNT(Main_data_set[column_x]) , ALLSELECTED(Main_data_set), REMOVEFILTERS(Main_data_set[column_y]) )

[pdelia]

Replying to this old thread because it came up in a search for me. DAX now has a REMOVEFILTERS function which I believe accomplishes what this is asking about (and certainly works in the case I was dealing with).

[credit_check]

Thank you, was looking for exactly this.

[BetterCallFrank]

Hi @carinat

have you tried:

Measure = 
DIVIDE(
  COUNT(data[Respondent]),
  CALCULATE(
    DISTINCTCOUNT(data[Respondent]),
    ALL(data[category])
  )
)

HTH,

Frank

[carinat]

Hi Frank,

Thanks for your suggestion but that actually just makes it worse.

What I have look like the following (anonymized):

And adding the column within ALL gives me:

[v-sihou-msft]

@carinat

In DAX, there's no function which is the opposite of ALLEXCEPT(). You have to use All(Column1, Column2,...) to include all other column as filter context expect that column you want to ingore. For more details, please see: ALL Function (DAX).

Regards,

[BetterCallFrank]

can you please post a screenshot of your data table?

Frank