Possible odd DAX Behaviour of a DISTINCTCOUNT of a dimension key

[DebbieE]

I have created the following DAX to get the count of all records in a dimension (Getting rid of NAs)

Possible odd DAX Behaviour of a DISTINCTCOUNT of a dimension key

# Entities across Company = CALCULATE(DISTINCTCOUNT('Dim'[Key]), 'Dim'[Sub Term] <> "NA")

If I add it into a table with the key its fine

Its counting 16 of them without the NA row. However. When I remove the table from the key it goes wrong

Now each row shows as 16. I cant work out what I have done wrong in the DAX apart from use the key but normally this works fine. 

Is there anything I can do with the DAX so it counts each record once even without the key in the table?

Thanks in advance

[MFelix · Accepted solution]

Hi @DebbieE ,

This has to do with the context and the way the calculate works. The use of calculate with a filtering when applied make the same calculation has a filter all.

In your case when you take out the ID from the measure it will pick up all the values from the ID and then getting the 16 instead of 1 for each of the values.

Check the link about the CALCULATE and how it works.

https://dax.guide/calculate/

Context is everything especcially when working with calculate, in this case you should substitute the ID by the SUBTERM and believe you will get the correct result.