Data Science Capstone Project Example Guide & Walkthrough

1. From the Data Pane, drag Category into both the Rows and Columns shelf.
1. The Columns shelf should be aggregated as a COUNT while the Rows should remain a dimension, if not, you can right click the cell and hover over Measure, and from there change the type to COUNT.

2. You can add more depth by dragging Category to the Color Shelf.

3. If need be, you can also place Category under filter and change it so that Null is an unchecked value, that way Null isn’t reflected in the bar chart.

4. Now for the important part, drag Category into the Text Shelf under in the Marks tab. Then, change the Measure to COUNT and down below by the bottom of the pop up window there will be text that says “Add Quick Table Calculation.” Make sure this is set to Percent of Total. Now, the percentages of each category will be presented alongside each bar! 🎉

5. Video below!

1. The pie chart is a bit different than you would expect, and has some more involved steps than the bar chart. The first thing you want to do is drag Category from the Data Pane onto the Rows Shelf twice, one is a dimension and the other is a COUNT value.
1. For the COUNT Category, we want to change this by adding in a Quick Table Calculation. Add in Percent of Total, just like we did for the bar chart percentages. You might have to click into “Edit Quick Table Calculation” and adjust the settings to that it performs the calculation using Table (down) instead of Table (across) based on the way this is formatted by row.

2. If need be, you can also place Category under filter and change it so that Null is an unchecked value, that way Null isn’t reflected in the bar chart.

3. Next, we need two new data values (if not already created by default). These will be Measure Names and Measure Values. To do this, in the Data Pane you can right click on any existing data value (for ex. Names) and click Duplicate. From there, you can change the name and right click again to convert to a Measure.
1. Add Measure Names to the Filters section. You can leave the default selections as this will update later on.

4. Under the Marks tab where it says “Automatic” click on the drop down and select “Pie.” This will change our setup and allow us to now drag over Measure Names to the Color Shelf and Measure Values to the Angle Shelf. ❗At this point, you should see your pie has been divided into slices!❗

5. We need another new data value in order to show the percent of each category. This time, we have to create a Calculated Field which can be done by clicking on the small arrow next to the Search Bar in the Data Pane and selecting “Create Calculated Field.”

6. Rename this as “1 - % (cutting slice into pie)” for more clarity and then input the following code snippet which will take the percentage out of 100:

//the remaining percentage out of 100
1 - COUNT([Category]) / TOTAL(COUNT([Category]))

7. Almost done! In the Measure Values subsection, which can be seen under the Marks tab, you want to make sure that the only green cells in that area are the calculated field we just made and the COUNT Category (again, you can make a duplicate of the Category dimension and use this as the input). This step can be seen in the picture above as well!
1. Once COUNT Category has been moved, right click and add in a Quick Table Calculation with Percent of Total and Table (down) as the selections, just like we did for the one in the Rows Shelf.
2. As for the percentage calculation, right click and in the Add/Edit Table Calculation area, choose “Specific Dimensions” and select Category in the checkbox area. Your pie chart should now reflect only the percent of one category!

8. Video Below!

1. From the Data Pane, drag Age to the Rows Shelf and Rank to the Columns Shelf. You should see the beginnings of the chart start to take shape! ❗(Make sure these are both Dimensions!)❗

2. Next, in the Marks tab, I decided to add more depth to mine by putting Category in the Colors Shelf and Names in the Detail Shelf. This just adds more description to each data point, which is always helpful to add more clarity to a visualization.
1. Change “Automatic” to “Circle” if you prefer, it’s more of a stylistic preference but up to you to decide!

💡

Use your Tools(tip)!

If you click on the Tooltip Shelf in the Marks tab, you can edit how information is displayed/what is displayed when you hover over each data point. You can also change the font or sizing to your liking, which is something I found useful when organizing my formatting later on!

3. If need be, you can place Category under filter and change it so that Null is an unchecked value, that way Null isn’t reflected in the scatterplot.

4. Also under filters, I grouped my ages by SUM, so you can drag the Age cell to this area and switch to SUM Age under Measure.
1. I wanted to allow the user to be able to see different ranges of ages, so I right clicked “Show Filter” on this cell, selected “Range of Values”, and just chose a range of ~10 years for the slider bar. If all went according to plan, you should see a fancy new scatterplot on your screen! 🎉

5. Video Below!

1. In order to make a histogram, we have to edit our data values first. Right click on Age in the Data Pane and hover over “Create”, then select “Bins.” This will create a new discrete value titled “Ages (Bins)” where we can now choose the number of bins or groupings for this plot.
1. This is up to personal preference, but I chose 5 as my bin size. Tableau also has a suggest button that will provide a reasonable number for bin sizes based on your data. Feel free to play around with this!

2. Now, we can drag and drop Ages (Bins) to the Columns Shelf and Names to the Rows Shelf. Note that we want a COUNT Name (by selecting Measure and clicking on COUNT) and also need to add a Quick Table Calculation of the Percent of Total (Table Across, not Down) as well.

3. If need be, you can place Ages (Bins) under filter and change it so that Null is an unchecked value, that way Null isn’t reflected in the histogram.

4. I also wanted to add more depth by bringing the Names cell into the Color Shelf, which shows a density distribution of the more populated bins with a deeper color. Congratulations, you’ve now made a histogram!!! 🎉

5. Video Below!

1. From the Data Pane, find the data value that says Country/Territory. Right click this and ensure that in the Geographic Role it’s set to Country/Region, that way we know we’re on the right path to generate a map. ❗There should be a globe icon next to it 🌎❗
1. Now go ahead and drag and drop Country/Territory into the main workspace, which will auto populate the Row and Column Shelf with Latitude and Longitude respectively.
2. ❗This is a very important step for Dual-Axis maps! ❗ You have to include another Latitude cell in the Rows Shelf, which can be found as a Measure value in the Data Pane. Don’t be alarmed! This will create two split maps but we can adjust this later on.

2. Take a look at the Marks tab, there are now two Latitude subsections. For the first subsection, change from “Automatic” to “Circle” and drag Names as a COUNT value to the Size Shelf. You should see that we now have individual density bubbles on the map based on how many people are from that country. For the Detail Shelf we should have Country/Territory already populating this (if not just drag and drop).
1. For the second Latitude subsection we will take COUNT Names again and place this in the Color Shelf, generating a heat map version. For this we also want to make sure that Country/Territory is added to the Details Shelf.

3. You’re probably wondering, “how do we get these two maps to merge together?” THIS is the step you’ve been waiting for! To do this, go to the second Latitude cell in the Rows Shelf, right click, and choose Dual-Axis (as shown below). The two maps are now merged together and in the “All” section of the Marks tab, you can see that “Multiple Fields” now represents some of the shelves.

4. In the filters section, feel free to place Country/Territory and COUNT Names, customizing to your liking. Fantastic work, you’ve just created an amazing looking Dual-Axis map! This one was a bit more complicated to figure out, so give yourself a nice pat on the back!! 🎉

5. Video Below!

IF CONTAINS(LOWER([Designation]), "ceo") OR
   CONTAINS(LOWER([Designation]), "executive") OR
   CONTAINS(LOWER([Designation]), "founder") OR
   CONTAINS(LOWER([Designation]), "coo") OR
   CONTAINS(LOWER([Designation]), "cfo") OR
   (
     CONTAINS(LOWER([Designation]), "president") AND 
     (
       CONTAINS(LOWER([Designation]), "company") OR
       CONTAINS(LOWER([Designation]), "business") OR
       CONTAINS(LOWER([Designation]), "firm") OR
       CONTAINS(LOWER([Designation]), "group") OR
       CONTAINS(LOWER([Designation]), "chairperson") OR
       CONTAINS(LOWER([Designation]), "bank")OR
       CONTAINS(LOWER([Designation]), "finance")OR
       CONTAINS(LOWER([Designation]), "lucasfilm")OR
       CONTAINS(LOWER([Designation]), "citigroup")
       
     )
   ) THEN
   "Business Leader"

ELSEIF CONTAINS(LOWER([Designation]), "minister") OR
       (
         CONTAINS(LOWER([Designation]), "president") AND
         NOT (
           CONTAINS(LOWER([Designation]), "company") OR
           CONTAINS(LOWER([Designation]), "business") OR
           CONTAINS(LOWER([Designation]), "firm") OR
           CONTAINS(LOWER([Designation]), "group") OR
           CONTAINS(LOWER([Designation]), "chairperson") OR
           CONTAINS(LOWER([Designation]), "finance")
         )
       ) OR
       CONTAINS(LOWER([Designation]), "prime minister") OR
       CONTAINS(LOWER([Designation]), "vice president") OR
       CONTAINS(LOWER([Designation]), "speaker") OR 
       CONTAINS(LOWER([Designation]), "chancellor") OR
       CONTAINS(LOWER([Designation]), "governor") OR
       CONTAINS(LOWER([Designation]), "monarch") THEN
   "Politician"

ELSEIF CONTAINS(LOWER([Designation]), "director") OR
       CONTAINS(LOWER([Designation]), "partner") OR
       CONTAINS(LOWER([Designation]), "chair") OR
       CONTAINS(LOWER([Designation]), "chief") OR
       CONTAINS(LOWER([Designation]), "head") OR
       CONTAINS(LOWER([Designation]), "evp") THEN
   "Upper Management & Guidance"

ELSEIF CONTAINS(LOWER([Designation]), "actress") OR
       CONTAINS(LOWER([Designation]), "philanthropist") OR
       CONTAINS(LOWER([Designation]), "author") OR
       CONTAINS(LOWER([Designation]), "musician") THEN
   "Media Figures"

ELSE
   "Other"
END