Hiding Matrix values by adding a flag to the source data.

I’m trying to create a column that will flag a row as being included, or not, for display.

I need basically to hide any rows where the number of unique respondents in a subgroup who gave that particular response to that particular question is one or less.

The data is unpivoted, so each row of the source table is one question response from one respondent.

So – the code I’m trying to write is attempting to do a filtered calculate where it counts the distinct number of respondents (each has a unique ID, “UserNo”), where these people meet these critera:

1:They have given the same answer (“1-5 num” – just a number between 1 and 5)

2: They have answered the same question (“QCode” – Text string)

3: They are all in the same organisation (“Org Name (Display)” - Text string)

4: They are all Training in the same subject (“Training Subject” - Text string)

then, if this count comes to more than one, return “true”, otherwise, “false”

This is the code I’ve got:

ShowResponse =

IF(CALCULATE(DISTINCTCOUNT

('Response Data'[UserNo]),

    'Response Data'[1-5 num]='Response Data'[1-5 num],

        (IF('Response Data'[Org Name (Display)] = 'Response Data'[Org Name (Display)],"TRUE","FALSE")),

                (IF('Response Data'[QCode] = 'Response Data'[QCode],"TRUE","FALSE")),

                    (IF('Response Data'[Training Subject]='Response Data'[Training Subject],"TRUE","FALSE"))

                ) > 1,"TRUE","FALSE")

It runs, but it comes back with “FALSE”, for every row. I think I know why (the DISTINCTCOUNT one comes up with 1 because it's comparing to itself, then you can’t subfilter lower than 1…) so I need a different approach. Basically, how to calculate, for each row, the number of other rows in the same table that are similar to it based on the above criteria. Is that even possible, or is it considered circular and therefore impossible?

Basically, the desired end result is a matrix that needs to act as if these rows aren’t there at all when it calculates each cell. I’ve tried applying filters to the visual (based on a creating a measure doing distinct count of UserNo) but that doesn’t work, so I’m trying to flag the data upstream, if that makes sense.