1.TBL_Personal
PersonID -- PK
Gender
Country

2.TBL_PersonalTravelProfile
PersonID -- PK
TravelID -- PK

3.TBL_TravelProfile
TravelID -- PK
QuestionID
ChoiceID

Data for TBL_Personal
1 M USA
2 F USA
3 M Can
4 M USA
5 M USA

TBL_PersonalTravelProfile
1 1
1 6
1 7
2 1
2 2
3 10
3 9
3 1

Data for TBL_TravelProfile
1 100 111
2 100 112
3 100 113
4 100 114
5 100 115
6 200 211
7 200 212
8 200 213
9 200 214
10 200 215

TravelProfile will be a dimension which should look as follows:

TBL_PersonalTravelProfile stores all the Travel Choices the person has
selected.
Dimensions are Gender, Country and TravelProfile
I want to see count by each dimension. However if I link these tables
directly, Analysis Services creates a inner-join, whereas I want is an outer
join from TBL_Personal to TBL_PersonalTravelProfile.
Basically the count for Person by Gender and/or country should be 5 and by
TravelProfile dimension should be 8 based on the above data.

Any design suggestions for the tables and/or cube. Please remember I will
have over 200 choices for Travel, so I cannot flatten it out.

-Subhash