Convert Tableau calculations into PowerBi dax

[AMorulita]

Hi friends! Please help!

I'm new to PowerBI and I need to convert a couple of calculation from Tableau into PowerBI:

1. I need to calculate the total number of fields that were marked with an error for a specific user.
2. I need to calculate the total number of fields reviewed for a specific user.
3. I need to calculate the above two but cross all users.
4. I need to visualize top 4 fields with errors across all users.

Context: users work/update fields. The reviewers check those worked fields and mark with Error or No Error. We need to calculate the defect ratio between the fields marked with errors versus total fields reviewed for each user. Also, need to calculate the defect ratio overall across all users.

Thanks so much for all your help!!!!

In Tableau the calculations look like:

Defect ratio: 1-(SUM([No Error Count])/SUM([All Field Count]))

No Error Count: {FIXED [User],[ReviewDate], [ErrorType], [ReviewItemID]:SUM(IF [ErrorType]="No Error" THEN 0 Else 1 END)}

All Field Count: {FIXED [User],[ReviewDate], [ReviewItemID]:COUNT([ErrorType])}

I tried putting dax together but didn't get any meaningful results.

Sample data looks like the 3rd table below. So, for example, then I would calculate the Correctness indicator for June, which is =(100-(7*100/28))/100 (Incorrect Info errors=7, all reviewed fields=28). I would also calculate the Completeness indicator for June, which is =(100-(5*100/28))/100 (Missing errors=5, all reviewed fields=28). 

Top 4 fields with errors across all users: 

Document Content	6
Document Remarks	7
Document Title	8
Document Translation	7

User	Assessment Score	Calculation explanation
User 1	55.56%	=(100-(4*100/9))/100
User 2	50.00%	=(100-(5*100/10))/100
User 3	66.67%	=(100-(3*100/9))/100

User	ItemID	Field	ErrorType	Status	Review Date
User 1	12345	Document Title	Incorrect Information	Correction Needed	Jun-24
	12345	Document Content	No Error	No Correction Needed	Jun-24
	12345	Document Translation	No Error	No Correction Needed	Jun-24
	12345	Document Remarks	No Error	No Correction Needed	Jun-24
User 1	123456	Document Title	Missing	Correction Needed	Jun-24
	123456	Document Title	Incorrect Information	Correction Needed
	123456	Document Content	Missing	Correction Needed	Jun-24
	123456	Document Translation	No Error	No Correction Needed	Jun-24
	123456	Document Remarks	No Error	No Correction Needed	Jun-24
User 2	67890	Document Title	No Error	No Correction Needed	Jun-24
	67890	Document Content	No Error	No Correction Needed	Jun-24
	67890	Document Translation	Incorrect Information	Correction Needed	Jun-24
	67890	Document Remarks	Missing	Correction Needed	Jun-24
	67890	Document Remarks	Incorrect Information	Correction Needed	Jun-24
User 2	9876	Document Title	Incorrect Information	Correction Needed	Jun-24
	9876	Document Title	Missing	Correction Needed	Jun-24
	9876	Document Content	No Error	No Correction Needed	Jun-24
	9876	Document Translation	No Error	No Correction Needed	Jun-24
	9876	Document Remarks	No Error	No Correction Needed	Jun-24
User 3	1234567	Document Title	Incorrect Information	Correction Needed	Jun-24
	1234567	Document Content	No Error	No Correction Needed	Jun-24
	1234567	Document Translation	Missing	Correction Needed	Jun-24
	1234567	Document Translation	Incorrect Information	Correction Needed	Jun-24
	1234567	Document Remarks	No Error	No Correction Needed	Jun-24
User 3	654321	Document Title	No Error	No Correction Needed	Jun-24
	654321	Document Content	No Error	No Correction Needed	Jun-24
	654321	Document Translation	No Error	No Correction Needed	Jun-24
	654321	Document Remarks	No Error	No Correction Needed	Jun-24

[Greg_Deckler · Accepted solution]

@AMorulita First, I would use the Fill down feature in Power Query to fill out your User column. Then you can do things like this:

Measure 1 = 
  VAR __Table = SUMMARIZE( FILTER( 'Table', [ErrorType] <> "No Error" && [User] = "User 1" ), [ItemID], [Field] )
  VAR __Result = COUNTROWS( __Table )
RETURN
  __Result

Measure 2 = COUNTROWS( FILTER( 'Table', [User] = "User 1" ) )

Measure 3a = 
  VAR __Table = SUMMARIZE( FILTER( 'Table', [ErrorType] <> "No Error" ), [ItemID], [Field] )
  VAR __Result = COUNTROWS( __Table )
RETURN
  __Result

Measure 3a = COUNTROWS( 'Table' )

For the last one you can use a TopN filter or TOPN DAX function or perhaps RANKX