3 Replies Latest reply on Jan 27, 2019 10:07 AM by Shinichiro Murakami

    Correlation Matrix Between Two Different Dimensions

    Stephen Groff

      I've been working on figuring out the correlation matrix for a few hours now and something seems pretty clear to me... most tutorials will only show you how to compare 1 dimension with any measure.

       

      In my scenario I would like to compare 2 dimensions with one measure.

       

      Dimensions used:

      [Model Number]

      [Work Center]

       

      Measure used:

      [Number of Defects]

       

      I'm trying to find a correlation between the model number and where a defect occurs on the assembly line.

       

      i.e. When I'm building Model X which work centers are more likely to cause a defect?

       

      Defects are characterized (in my data) as event numbers.  There can be 100's of Events (defects) written on the assembly line each day.

       

      As shown within the tutorials I've watched, I add the same table twice in the data source and did an inner join on the [Event No] and ran through the calculation that each tutorial shows you... but this gets me no where.

       

      The tutorials all have the same dimensions (i.e. Model & Model1) on the X and Y axis.

       

      I need my Model numbers on the Y axis and my Work Centers on the X axis.  Then I need my Pearson Corr Calc to be added to the color card to show the correlations.

       

      Here's the calc per the tutorial:

       

       

      Does anyone have any ideas as to how to go about doing this?