Looks vs Personality

I’m going back in time for this #WorkoutWednesday challenge to Andy’s 2017 week 53 which is being included in week 1 for 2018. This is all part of my goal of completing all 52 weekly challenges in 2018 along with all my other 2018 dataviz goals. So here we go, a few weeks late. The challenge is to recreate this

So we’ve got a simple dot plot showing % of gender who ranked personality over looks as their number 1 trait.

The data is pretty straightforward. One row for each country and gender. So to build the dot plot, we’ll put [Value] on columns and [Nationality] on rows. Next, I’ll change the mark type to shape and drop gender onto the shape card. In order to get the shapes correct, the men get a closed circle and the women get the open circle. I used a color picker program on my mac to find the hex code to match the color of the original, which is #3a9e9f.

And here’s where we’re at thus far

Screen Shot 2018-02-23 at 9.00.27 PM

Now the original is sorted by women value and we’re gonna need to create a field in order to do that.

IF [Gender]=”Women” THEN [Value] END

Now we’ll sort [Nationality] by this new field.

Screen Shot 2018-02-23 at 9.14.22 PM

Now I need to connect the dots for three of the nationalities. In a recent #WorkoutWednesday challenge, I wrote about a couple different approaches to connected dot plots and as was the case in that challenge, the line is there only for context and not as a measure so I will go for the line technique again.

But first, I need to create a calculated field for only 3 of the countries. This can be done with the following IF statement [Connect the Dots]

IF [Nationality]=”Egyptian” THEN [Value]
ELSEIF [Nationality]=”Saudi Arabian” then [Value]
ELSEIF [Nationality]=”Vietnamese” then [Value]
END

I’m going to put the measure on columns and create a dual axis. Next, I’ll change me chart type to a line and put [Gender] onto the path card and we end up with this

Screen Shot 2018-02-27 at 9.10.30 PM

So we’ll make that a dual axis (don’t forget to sync) and we’re in the home stretch.

Well crap, these shapes are not right because I can’t hide the line behind the open circle because it has a transparent center. So to fix that, I’ll open powerpoint and create a couple circles, one closed and on that looks open but is really just a copy of the closed circle with a white smaller circle on top which will now not be transparent. Nice trickery, Andy!

Now, to get the axis just right. This could be tricky because I can’t hide the bottom axis without the top one disappearing as well. But I can do some formatting magic. For the bottom axis, I’ll format all the text and tick marks to be bright white which will make it look like it’s been hidden. For the top axis, I’ll set the axis marks to only show every 50 units to give me the 0, 50, & 100

My dotted line at 50 is a constant reference line and the gray background is a reference band from 0 to 100. Then I’ll set the gridlines to white and the correct width this is what I’ve got

Screen Shot 2018-02-27 at 9.50.54 PM

Now then, let’s get this on a dashboard for our final touch – the annotations. Clearly, we’re not going to be able to do the curved arrows but we should be able to do everything else.

Looks vs PersonalityWW

Great challenge this week. I like these that have a good mix of Tableau technicality as well as design strategies.

Until next time!

 

Leave A Reply

Your email address will not be published.