Distinct count based on 2 or more values of another column

[Moe_M]

Good day all, 
I have a dataset that looks like below. I was trying to get a distinct count of 'sentence_id' column when 'level_3' is both 'actions' and 'contacts'. 

sentence_id	level_3
1280590690	Actions
1280590690	Contact
1280590690	Actions
1280590690	Actions
1280590690	Actions
1280590690	Contact
1280590690	Contact

[v-tangjie-msft · Accepted solution]

Hi @Moe_M ,

According to your description, here are my steps you can follow as a solution.

(1) This is my test data.  

(2) We can create a measures. 

Measure = CONCATENATEX(FILTER(ALLSELECTED('Table'),[sentence_id] in VALUES('Table'[sentence_id])),[level_3],",")

Measure 2 = var a={"Actions","Contact"}
var b=COUNTROWS(FILTER(a,CONTAINSSTRING([Measure],[Value])))
return IF(b=COUNTROWS(a),1,0)

Measure 3 = CALCULATE(DISTINCTCOUNT('Table'[sentence_id]),FILTER('Table',[Measure 2]=1))

(3) Then the result is as follows.

Best Regards,

Neeko Tang

If this post  helps, then please consider Accept it as the solution  to help the other members find it more quickly. 

[amitchandak]

@Moe_M , not very clear but try like

countrows(summarize(Table, Table[sentence_id], Table[level_3]) )