Age groups based on difference - Date slicer

  • Hello,

    I have a cube I'm working with that has a dimension called Employees.

    In the dimension table contains a Birthdate.

    What I need to do in a query is count the amount of Employees However, I also have to put them into age groups using the Birthdate. Below is an example of the dimension table:

    Employee Birthdate

    1 1951-04-04

    2 1987-05-05

    3 1986-11-04

    4 1977-10-10

    So, for simplicity, lets say three age groups;

    Under 22;

    22 to 34;

    Over 35

    I have no idea how I will achieve this as I need to use a Date slicer.

    The way this would have to work is whatever dates in the Date slicer (MDX WHERE), would have to evaluate the date difference between the birthdate and the selected date. and then add them to the predefined groups so the example below could be shown.

    <22 22-34 35>

    EmployeeCount 15 40 70

    Is there a way/function/MDX what I can create these groups and then group each of the employee? How would I store the birthdate for this to work; member property? attribute?

    Any help would go a long way.

    Thanks in advance

  • Couldn't you do this in a calculated member and use the DateDiff() function to get the number of years. Then use IIF() to put in groups?

    Just thinking out loud here.

    HTH,

    Rob

  • robert.gerald.taylor (11/5/2013)


    Couldn't you do this in a calculated member and use the DateDiff() function to get the number of years. Then use IIF() to put in groups?

    Just thinking out loud here.

    HTH,

    Rob

    Thanks for responding Rob. I can certainly look into it. The first thing I thought was DateDiff. Because I'm using the selected date, which is the slice, I'm not really sure how I'd use Datediff. If you have a simple example, I'd really appreciate it. then I'd need to put this logic into a function to reuse. My MDX skills aren't the strongest.

    I'll go away and look at IIF() for groups.

Viewing 3 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply