VizWiz

Data Viz Done Right

March 30, 2019

Groundwater Contamination and Cow Poo: A Major Contributor to Global Warming

No comments
This is a project I've been working on for a while now, mostly because time has not permitted me to finish it and I've had other "issues" to deal with. I've been doing lots of research about global warming, water contamination and whether or not the two are related.

While watching a documentary, they mentioned how methane from cows (i.e., cow farts) are a major contributor to the greenhouse gasses and how cow manure is a major source of nitrate released into groundwater used for drinking. Fortunately, there is tons of data available, the primary source being the Environmental Protection Agency (EPA).

I wanted to understand the geographical distribution of three factors:

  1. The percentage of each State with high groundwater nitrate concentrations.
  2. The total area (square miles) of each State with high groundwater nitrate concentrations.
  3. Where the cow crap comes from that pollutes groundwater used for drinking.

I decided to create a map for each of these topics, as a scrolling story, with three actions you can take to help reduce the impact of cow manure pollution. We all want safe drinking water after all.

March 26, 2019

#TableauTipTuesday: Create a Region to State Drill Down Map with Set Actions

No comments
In this tip, I show you how to use set actions to create a map that allows the user to click on a region and show the states for the region, but keeping all other areas at the region level.

March 25, 2019

#MakeoverMonday: Consumer Spending by Generation

No comments
For week 13, we're making over this viz from Business Insider:


What works well?

  • The generations are sorted from youngest to oldest.
  • The title is clear.
  • The gridlines help guide the eye across the viz.
  • It's easy to compare the general/misc category and the restaurants across generations.
  • A stacked bar chart is easy to understand.

What could be improved?

  • The story in the data, from the article, is about how millennials are spending more on restaurants. It would be good to make that a more obvious focus of the viz. 
  • There are too many colors.
  • While the title is clear, if you don't read the article, you could miss the purpose for the chart.

What I did

I really enjoyed using Google Data Studio last week, so I thought I'd give it another try to continue my learning. Since this was a simple stacked bar chart, I wanted to create a "set" for restaurants vs. all others. I needed to create a calculated field using a case statement that checks the category field. That's it!

From there, it was formatting, which is pretty intuitive as well. I'd highly recommend you give Data Studio a try, especially if you know exactly what you want to build; it's not a data exploration tool.

 

March 20, 2019

#MakeoverMonday Data Studio Edition: Reykjavik Index for Leadership in G7 Countries

No comments
Yesterday I posted a Power BI version of Data Schooler Hanna Nykowska's Makeover Monday viz. Today, I recreated her viz in Google Data Studio.

DATA PREP REQUIRED

  1. Add values for the remainder (100 - index)
  2. Add a sort column
  3. Pivot the data so that the index and the remainder were in the same column

WHAT WORKED WELL

  • To create the stacked bar chart, all you need to do is select the chart type and drop the fields on the appropriate shelves.
  • Customizing the split of the colors for the index and the remainder was easy.
  • I was able to customize the size of the viz.
  • You can choose any font that Google supports!!
  • The tooltips are super responsive.
  • Everything looks very crisp.
  • Hiding the gridlines leaves a nice thin black line on the y-axis without me needing to fiddle around with a few different settings.
  • The overall UX is quite intuitive. I see they have a data explorer now too.

WHAT I COULDN'T OVERCOME

  • I couldn't find a way to show only the mark labels for the purple bars.
  • I couldn't add a reference line for the G7 average so I had to leave it in the view.
  • I couldn't hide the x-axis only. When you do, the y-axis gets hidden as well.

With that, here's my third Makeover Monday for week 12 2019.

March 19, 2019

#MakeoverMonday Power BI Edition: Reykjavik Index for Leadership in G7 Countries

No comments
Data Schooler Hanna Nykowska create a viz this week for Makeover Monday that was quite similar to my first idea. I didn't publish mine, so I thought instead of creating my version again, I would try to recreate her viz in Power BI.

DATA PREP REQUIRED

  1. Add values for the remainder (100 - index)
  2. Add a sort column
  3. Pivot the data so that the index and the remainder were in the same column

WHAT WORKED WELL

  • Creating a stacked bar chart in Power BI was quite simple.
  • Customizing the split of the colors for the index and the remainder was easy.
  • The viz layout is super intuitive and automatically adjust to the size of the screen while maintaining the original chart ratio.
  • The fonts look super crisp.
  • Simple to add a constant reference line for the G7 average.

WHAT I COULDN'T OVERCOME

  • I couldn't find a way to show only the mark labels for the purple bars.
  • I had to change the mark labels so that the values of the grey bars wouldn't be visible by making the text the same color as the grey bars.
  • I couldn't copy/paste into a text box.
  • I couldn't customize the font size for the reference line.
  • I'm sure there's a way, but I couldn't figure out how to color code the bars based on whether they were above or below the G7 average. For example, I wanted to make those countries below the G7 average a lighter shade of purple.

With that, here's my second Makeover Monday for week 12 2019.

March 17, 2019

#MakeoverMonday: To what extent are women and men viewed equally in leadership positions?

No comments
The chart Eva chose this week shows that, not surprisingly, people are less likely to view women and men equally in leadership positions. Scores of 100 mean that people view women and men are equally suited to leadership roles. Sadly, women still have a long way to go in their seemingly never ending fight for equality.

Let's have a look at the chart:

Source: World Economic Forum

WHAT WORKS WELL?


  • Ordering the countries from highest to lowest in terms of people that view women and men equally in leadership positions
  • Including the G7 average for context
  • Assigning a different color to the G7 average
  • Labeling the end of the lines

WHAT COULD BE IMPROVED?

  • Circular bar charts are horrible for comparisons.
  • The title is meaningless.
  • The lines start thin, get thicker, then get thin again. Why?
  • The title and the center of the chart are the same. That's certainly unnecessary redundancy.

WHAT I DID

I started by creating a simple bar chart and that was fine. I also added a grey bar to have it as a stacked bar for each country that goes up to 100%. I then thought about doing a waffle chart (with circles) and then I remembered this viz from Andy Cotgreave back in Makeover Monday week 4 2016. I decided to replicate Andy's work since it looks great and gives lots of context. I created a mobile version like Andy did too.

With that in mind, here's my makeover for week 12.

March 12, 2019

#TableauTipTuesday: How to add a one pixel line to a dashboard

No comments
In this week's tip, I show you how to use text boxes combined with containers to add divider lines in your dashboards. You can download the workbook here.

March 11, 2019

#MakeoverMonday: Has Philadelphia recovered from the Great Recession?

No comments
For this week's Makeover Monday, we're looking at this dashboard from OpenDataPhilly.


What works well?

  • Consistency of colors
  • Simple design
  • Using an area chart with a bold line at the top
  • Bar chart is sorted
  • Interactive actions
  • Automatic proportional brushing

What could be improved?

  • Reduce the outline of the zip codes on the map
  • Remove the background from the map
  • Add a dashboard title
  • Change the chart titles to be more meaningful

And here's my makeover. Click to interact.


March 5, 2019

Makeover Monday Power BI Edition: Births Attended by Skilled Health Staff vs. Female Life Expectancy as a Motion Chart

No comments
One of the things I couldn't do with my Makeover Monday this week was animate the visualization. The animations is what makes the story unfold and neither Tableau Public nor Tableau Server support animation.

What's one to do? Try a tool that does support animation. In this case, Power BI. Scatter plots in Power BI support animation natively and it took less than five minutes to create this.

  1. Upload the data.
  2. Choose the measure for the x-axis and place it on the X Axis shelf.
  3. Choose the measure for the y-axis and place it on the Y Axis shelf.
  4. Add a dimension to the Details shelf to draw more dots.
  5. Place the dimension to animation across, i.e., years, on the Play Axis shelf.
  6. Add a title.
  7. Add a text box as a footer.

BOOM! Done! Easy peasy. Check it out below.

March 4, 2019

Makeover Monday: Are skilled health staff an indicator of female life expectancy in fistula countries?

No comments
For Makeover Monday week 10, Eva presented us with this map of adolescent fertility rates on Data Wrapper:


What work well?

  • Using a continuous color palette
  • There are no exceptionally large countries compared to the others, so a filled map is a good choice.
  • Normalizing the data to make comparisons across countries more relevant.
  • Using grey for countries with no data.
  • Good title and subtitle

What could be improved?

  • If there is data across years, it would provide additional context to the data. In other words, is the situation improving?
  • Make the title bigger; it's too small compared to the large map.

My Goals

  • Compared the metrics between fistula and non-fistula countries
  • Look at change over time
  • Figure out how to deal with all of the nulls
  • Be done