1. In a previous article, we showed how to create waffle chart from a single percentage, without extra data table and without table join. Also the labeling is automated instead of manual, comparing to other approaches.

    Now we are going to apply the same method to creating waffle chart for multiple percentages as follows. Note that the percentages are calculated horizontally, showing the percentages in different regions per each category.
    The steps are: (The 2nd step uses a different approach from the previous one for the same result.)

    1.Union the data source with itself.
    This will generate a new column Table Name with two different table names.

    2.Use the two table names to define the first and 100th points
    For each waffle chart, we need to have a 10x10 grid of 100 data points. In the video we created a set for one of the tables to define the first point. If outside the set, it is defined as the 100th point. This is a different approach from the one in the previous post. This way, the formula seems simpler.

    3.Create Point Bins from 1 to 100
    Thus we define the 100 points. Bin size = 1
    The bins will allow us to create a 10x10 grid that is the essence of the Waffle chart. The 98 points are the result of data densification through the bins.

    4.Create X and Y coordinates for the 10x10 Grid
    • X = -First()%10
    • Y = Int(-First()/10)
    Both compute along Point Bin. BTW -First()=Index()-1, an equivalent way to index the points from 0 to 99.

    5.Populate Percent to every point in the grid.
    Percent Omni = Window_Max(Max(Percent))
    We have only two real data records as data points. The other 98 points are the results of data densification. We need table calculation to assign data to them.

    6.Color the grid
    Use this formula to color: Index()/100<= Percent Omni 
    Index()/100 will assign a percentage to each point from 1% to 100%.

    7.Label the grid
    We only need one label in the middle. Here it is:
    If Index()=55 Then Percent Omni End

    That is it. You can watch the video for further details.

    The workbook can be downloaded here.

    Posted by
    0

    Add a comment

  2. Waffle chart is popular for visualizing a single percentage. How can we create such a chart?

    Here are the steps. The example data source has a single data record.

    1.Union the data source with itself.
    Thus the data source has two records now. And it generates a new column: Table Name

    2.Define the first and 100th points
    Based on the two table names, we can define the two points. We need to have 100 points to create a waffle chart, as a 10x10 grid.

    3.Create Point Bins from 1 to 100
    Thus we define the 100 points. Bin size = 1
    The bins will allow us to create a 10x10 grid that is the essence of the Waffle chart.

    4.Create X and Y coordinates for the 10x10 Grid
    • X = -First()%10
    • Y = Int(-First()/10)
    Both compute along Point Bin. BTW -First()=Index()-1.

    5.Populate Percent to every point in the grid.
    Percent Omni = Window_Max(Max(Percent))
    We have only two real data records as data points. The other 98 points are results of data densification. We need table calculation to assign data to them.

    6.Color the grid
    Use this formula Index()/100<= Percent Omni 
    Index()/100 will assign a percentage to each point from 1% to 100%.

    7.Label the grid
    We only need one label in the middle. Then
    If Index()=55 Then Percent Omni End

    That is it. You can watch the video for further details.

    The workbook can be downloaded here.




    Posted by
    0

    Add a comment

  3. Usually when we create a bar chart with 2 hierarchical dimensions such as Category and Sub-Category, we will have bars at the Sub-Category level and roll-ups at the Category level. Here is a chart we can create easily.
    This is simple but it doesn't show the hierarchical relationship very well.

    Today, we have an alternative way to chart the the bars and their hierarchical relationship as follows:

    The grey bars for Category is created using reference lines. 
    The fill-below grey color will create the the Category bars which encompass the Sub-Category bars. This we believe is a better visual representation of hierarchical relationship.

    After we posted it to twitter, Tableau Zen Master Jeff Shaffer said he created similar chart long time ago. So Jeff is the original creator of this kind of chart.

    The workbook can be downloaded here.
    Posted by
    0

    Add a comment

Blog Archive
Blog Archive
Loading
Dynamic Views theme. Powered by Blogger. Report Abuse.