Employee Count - Returns strange numbers

[Anonymous]

Hello all,

I have an emloyee table and created a measure called employee count.

Employee Count =

VAR selectedDate = MAX('Date'[Date])

RETURN

CALCULATE(COUNTROWS('All Employees'), FILTER('All Employees', 'All Employees'[DTHired]<= selectedDate && OR('All Employees'[DTTerminated] >= selectedDate, ISBLANK('All Employees'[DTTerminated]))))

However, the results it returns are incorrect. I did some research and found out that it is counting theincorrect number of terminstions. Also, if I change the function to say all employees who where terminated before they hire it finds employees.... this is impossible. Here is the impossible function that still seems to find people.

Employee Count =

VAR selectedDate = MAX('Date'[Date])

RETURN

CALCULATE(COUNTROWS('All Employees'), FILTER('All Employees', 'All Employees'[DTHired]>= selectedDate && 'All Employees'[DTTerminated] <= selectedDate))

and here is the graph it returns:

This should be zero every month.

This should be zero every month as employees are not terminated before hire.

Any help or ideas for a measure that would actually count my emplyees would be lovely.

[Greg_Deckler]

@Anonymous Maybe try this:

Employee Count =
  VAR __selectedDate = MAX('Date'[Date])
  VAR __Table = 
    FILTER(
      ALL('All Employees'),
      [DTHired] <= __selectedDate && ( [DTTerminated] = BLANK() || [DTTerminated] >= __selectedDate )
    )
  VAR __Result = COUNTROWS(DISTINCT(__Table))
RETURN
  __Result

[Anonymous · Accepted solution]

Thank you for the idea! It gives diferent numbers but they stil lare not correct and it finds employees who were terminated before they were hired. Using a different measure I can calculate it properly: 

Current Employees = CALCULATE(COUNTx(FILTER('All Employees','All Employees'[DTHired]<=max('Date'[Date]) && (ISBLANK('All Employees'[DTTerminated]) || 'All Employees'[DTTerminated]>max('Date'[Date]))),('All Employees'[Employee Id])),CROSSFILTER('All Employees'[DTHired],'Date'[Date],None))