4 Replies Latest reply on Apr 13, 2011 12:19 PM by Joe Mako

    Combining/blending members from same dimension (different data sources, different aggregation levels)

    Tim Harris

      Hi -

      I have two data sources – one, a small file with historical data aggregated at a dimension level; and two, a large file with current data stored at the individual observation level.  (I no longer have the individual records for the historical data). 

       

      I want to combine the data so all data points are at the level of aggregation in the first data set, and then display the results in a graph -- as one continuous set of data poins within a dimension.. 

       

      The data sets look like the following:

       

      Data at five points aggregated and stored on the source file as a percentage

      Snapshot date    Group1    Respresentation   

      12/31/2005    Cats    55%   

      12/31/2005    dogs    45%   

      12/31/2006    Cats    55%   

      12/31/2006    dogs    45%   

      12/31/2007    Cats    55%   

      12/31/2007    dogs    45%   

                 

                 

                 

                 

       

      Sixth data point generated from individual observations and percentage generated by tableau.

      Snapshot Date    Group1    Animal Name       

      12/31/2008    Dogs    Timber       

      12/31/2008    Dogs    Ralph       

      12/31/2008    Cats    Mr. Big       

      12/31/2008    Dogs    Don Enrique       

      12/31/2008    dogs El Pedon       

      12/31/2008    dogs    Ol Yeller       

      12/31/2008    Cats    Annabella       

      12/31/2008    dogs    Rex       

      12/31/2008    Cats    Yogi       

      12/31/2008    Cats    Boots       

      12/31/2008    Cats    Luther       

      12/31/2008    Cats    TS Elliot       

                     

      On the second data set, the percent representation will be calculated by tableau.

       

      I am not sure how to approach this. Not clear if this is possible, and the video on data blending doesn’t quite address it.  Appreciate any guidance here.

       

      Thanks.