Makeover Monday 2024 Week 3— New Year’s Resolutions Walkthrough
Makeover Monday is a weekly project where participants improve how they visualize and analyze data — one chart at a time.
With that goal in mind, I took a stab at this week’s challenge. I remember how helpful it was to hear other’s talk through their design choices when I first started participating in 2019, so I thought for this week I’d put my thought process down on paper. Let’s dig in!
The original chart
The original chart is fine. But — how could it be improved? Here are my thoughts:
- I want to make the categories more readable (angled text is hard)
- The gridlines don’t really add to the viz
- I don’t love the blue/pink colors for men and women.
- I’d like to understand how men and women rank their resolutions, and I can’t easily tell that from this chart.
Making over the Chart in Tableau
First, we have to load the data into Tableau. You can download it from data.world, I’ll be using Full_New_Years_Resolutions_Gender_Age_Breakdown-2.xlsx .
The first thing I am going to do is hide the fields I won’t be using — in this case, age groups.
Next, I am going to pivot the data — this will make it easier for me to color by category later. While I’m here, I’ll rename my fields as well (otherwise it automatically calls them ‘pivot names’ and ‘pivot values’ which isn’t very useful)
1. Make the text more readable
In my sheet, I’ll double click on category, resolution, and percentage to add them to my view. Then, I’ll click on “show me” in the top right, and change the chart type to a bar chart. That gets me here:
I want to keep the bars horizontal, because that solves my problem of the angled text for categories being hard to read. But I want to see the categories side by side, so I’ll drag the category pill from rows to columns:
2. Get rid of the gridlines
I said earlier I didn’t think the gridlines added much — so let’s take care of that. I like to set formatting at the workbook level, so that it applies to anything I’m doing in the workbook. Format > Workbook > Grid Lines > Off.
I’m using percentages, but my numbers don’t look like that. I’ll change the default formatting by right clicking on percentage > default properties > number format. In this case, I’ll use custom, zero decimal points, and add a % sign as the suffix.
3. Choosing my own colors
I mentioned I didn’t like was the blue/pink colors — time to change that. Dragging category onto color in the marks card gives us the good old tableau defaults: blue, orange, and red.
Clicking on Color, you can then hit edit colors and choose any you’d like. I’m going to go with a gray for all, green for men, and purple for women. You can read more about alternative colors for gender data here: https://blog.datawrapper.de/gendercolor/
Now the graph looks like this
4. Calculate and display rank
The last thing I wanted to improve the chart is be able to tell the rank of each category. I’ll solve this with some labeling.
First we’ll create a calculated field:
I’m using rank_dense because if two categories have the same percentage, I want the rank to tie, and then continue to the next number.
Dragging our new rank field and percentage onto label, it looks like this:
That’s…not readable. Click on label, and then you’ll get more options for editing.
After some formatting, and setting the viz to fit the entire screen, I now have a chart that looks like this:
And there we have it — a chart that I believe is better than the original. With some other formatting cleanup, adding titles and such, here is my final dashboard:
If you too would like to get better at improving charts one viz at a time, you can access all current and past Makeover Monday datasets here: https://makeovermonday.co.uk/.
