Pivot Table calculations

[#1]

I have a pivot table and it has a category of "Functional System".  For each Functional System (a type of highway) it gives me the total number fatalities for a 30 year period.  When I expand the Functional System, it then breaks out the the number of fatalities by year.  I am actually looking to have it show the percentage of fatalities of the total as opposed to the total number.  I have tried Set Analysis, but this does not seem to work.  I want it to show the percentage when the fields are collapsed or expanded.  Currently I have the expression as: Sum([Number Of Fatalities]) .

[Anonymous]

I can describe what is there.

Script for the test data, three years, three types:
load * inline [
Year, Type,  Number
2000, A,  10
2000, B,  8
2000, C,  7
2001, A,  5
2001, B,  3
2001, C,  1
2002, A,  13
2002, B,  9
2002, C,  11];

Front end - five pivot tables, all with dimensions Type and Year.  Expressions are different in each table

1. sum(Number)
2. sum(Number)/sum(total Number)
3. sum(Number)/sum(total<Type> Number)
4. sum(Number)/sum(total<Year> Number)
5. sum(Number)/ if(Dimensionality()<2, sum(total Number), sum(total<Year> Number))

See what makes more sense for you.

[#3]

Michael,

#5 works...sort of.  When I expand to the second dimension it gives 100% for all of the years.  So, I changed the Dimension for when I expand it out from Year to Functional System and it worked flawlessly!  Thanks for your help!