Something like the following:
Count(ID) / count (TOTAL <[Job Level]> ID)
,assuming that each person is identified by the field ID.
Upgrade your Qlik skills at the Masters Summit for Qlik - coming soon to Milan, Italy and Austin, TX!
Does your top table represent your raw data, i.e. a table from your model? Or do you have a table with fields Job Level, Sex, EmpID instead?
In latter case, you can do what Oleg suggested, create a chart with dimensions Job Level and Sex and the above expression (adapted to your field names).
In first case, you can consider transforming the table using CROSSTABLE LOAD prefix ( I think this would be best), then follow second case.
Or create a chart with dimension Job Level and two expressions:
=Num(Sum(Female) / Sum(Total),'#0.00%')
=Num(Sum(Male) / Sum(Total),'#0.00%')
Oleg & Swuehl -
Thank you for your prompt responses. Appreciate it!
Good question, thanks for checking Swuehl. I have the latter. So my data does look like this:-
Name ID Job Level Region Gender asf 123 Executive US Male asdfasdf 234 Professional APAC Female asdfasdf 345 Professional US Male asda 456 Manager Europe Male sdasd 567 Professional US
I tried the formulae you suggested as well as what Oleg suggested but I am not able to get it right. Based on my data above, do you have other thoughts?
What issues do you see? Incorrect numbers? No result? Error messages?
I think Oleg's solution should work:
Create a chart with two dimensions, [Job Level] and Gender
As expression, use
=Count(ID) / Count(TOTAL<[Job Level]> ID)
I assume you have distinct IDs in the ID column, so you don't need to use explicitely a count distinct.
If you still have issues, could you describe what exactely you have done and what you see (and maybe what you expect to see)?