A couple things came to mind on this -
Option 1a (bad) : Separate workbooks. Make a few copies of your workbook so that each one contains only the views and selected dimensions your client should be playing with.
Option 1b (not-as-bad) : Use folders for your dimensions. Place X,Y, and Z in a folder called "View 1" and then dimensions A,B, and C in a folder called "View 2".
Option 2 (ideal): Review this link http://onlinehelp.tableau.com/current/pro/desktop/en-us/help.htm#changing-views-using-parameters.html - and use parameters to allow your client to self select from a curated list of dimensions that you want them to play with on a particular view. I wouldn't suggest allowing a client have access to the whole list of dimensions that they could misinterpret.
Hope that helps - let me know if you run into any issues using the suggested solution!
I was just on the tableau page that you suggested and will try to use that to make it a little more interactive for our customer so they can create comparisons themselves.
Some of my measures are percentages while others are raw numbers so it is difficult to want to put them together as it wouldn't make sense if they clicked on a couple of measures at the same time.
I want to compare compliance rates when adding in different sources
Compliance Rates (source A)
Compliance Rates (source A + B)
Compliance Rates (Source A+B+C)
and be able to compare 1, 2 or all of them together in one workbook without seeing their other options.
They could also compare the numerators instead of compliance rates
Numerator (source A)
Numerator (source A+B)
Numerator (source A+B+C)
Therefore, if they were to click on a measure filter and click compliance rate ( Source A ) and Numerator (Source A+B) it would not work. This is what I am trying to avoid
Hopefully this helps and I will try to figure out how to use the parameters better.
Cool. I think you've got some interesting direction to take this in.
I did put together a lil mock-up of what it sounds like you're looking for - hopefully this helps you get going (if you haven't solved it yet!).
The one area I wasn't clear on from your last message was whether or not you wanted to be able to compare aggregations of sources - for example comparing source A to an aggregation of sources B and C (or comparing A to an aggregation of A an B, etc.). That would make things a bit tricky - but isn't impossible. I assume your data structure will dictate how well you're able to replicate based on this workbook - ie, I assume you have individual respondent level data, responses may be more binary in nature, are the sources separate files, etc?
In any event, the attached workbook allows you to:
a) filter by source. (A, B, C) Each source is a color coded line on the chart - I used time series data for this, which might not be what you're looking for at all, but should highlight the general point. Your horizontal axis could be any dimension (counties, age/age groups, etc.) - or you could use a parameter to select between dimensions (as used on the other parameter).
b) select a dimension. (1, 2, 3) These are randomly populated data for three dimensions. All three dimensions have numerators and denominators associated with each time period (the fourth dimension in this case)
c) select between counts and rates. Select counts and you see the numerator only for the selected dimension and source(s) across time. Select rates and you're seeing the particular rate on a given dimension from the selected source(s) across time (numerator / denominator).
Anyway, feel free to touch base with me offline if you want to chat further on this. Parameters are a game changer - if you thought data was fun before, just wait until you start seeing the world in parameters