I Have an App for That Too

Well, everyone, we made it to Friday. So let’s all reflect on how many things we did on our mobile phones this week. xkcd did. And it’s fairly accurate. Though personally, I would only add that I did not quite use my mobile for a TV remote. Unless you count Chromecasting. In that case I did that too.

What about boarding passes?
What about boarding passes?

If I have to offer a critique, it’s that it makes smart use of a stacked bar chart. I normally do not care for them, but it works well if you are only stacking two different series in opposition to each other.

Credit for the piece goes to Randall Munroe.

How Worldly Is the World Series?

The World Series began Tuesday night. But, as many people reading this blog will know, baseball is not exactly a global sport. But is it really? CityLab looked at the origin of Major League Baseball players and it turns out that almost 30% of the players today are from outside the United States. They have a number of charts and graphics that explore the places of birth of ball players. But one of the things I learned is just how many players hail from the Dominican Republic—since 2000, 12% of all players.

There are quite a few players from countries around the Caribbean.
There are quite a few players from countries around the Caribbean.

The choropleth here is an interesting choice. It’s a default choice, so I understand it. But when so many countries that are being highlighted are small islands in the Caribbean, a geographically accurate map might not be the ideal choice. Really, this map highlights from just how few countries MLB ball players originate.

Fortunately the other graphics work really well. We get bar charts about which cities provide MLB rosters. But the one I really enjoy is where they account for the overall size of cities and see which cities, for every 100,000 people, provide the most ballplayers.

One of the other things I want to pick on, however, is the inclusion of Puerto Rico. In the dataset, the designers included it as a foreign country. When, you know, it’s part of the United States.

Credit for the piece goes to David H. Montgomery.

Canadian Election Results

Yesterday Canada went to the polls for the 43rd time. Their prime minister, Justin Trudeau, has had a bad run of it the last year or so. He’s had some frivolous scandals with wearing questionable fashion choices to some more serious scandals about how he chose to colour his face in his youth to arguably the most serious scandal where an investigation concluded improperly attempted to influence a criminal investigation for political gain. (Sound familiar, American readers?) Consequently, there was some chatter about whether he would lose to the Conservatives.

But nope, Trudeau held on.

So this morning I charted some of the results. It was a bad night for Trudeau, but not nearly as bad as it could have been. He remains in power, albeit head of a minority government.

That's a steep drop in seats, but it could have been worse
That’s a steep drop in seats, but it could have been worse

Credit for the piece goes to me.

Prorogation of Parliament

If you’re among my British/European audience, you are probably well aware Boris Johnson has prorogued, or suspended, Parliament. He and cabinet ministers stated it was a normal, average-length prorogation to prepare for a Queen’s Speech. (The Queen’s Speech is the formal opening of a new session of Parliament that sets out a new legislative agenda and formally closes/kills any unpassed legislation from the old session.) Except that in documents revealed in a Scottish court case, we now know that the real reason was to shut down Parliament to prevent it from interfering in Boris’ plans for a No Deal Brexit. And just this morning the Scottish High Court did indeed rule that the prorogation is illegal. The case now moves to the UK Supreme Court.

But I want to focus on the other claim, that this is a prorogation of average length. Thankfully instead of having to do a week’s hard slog of data, the House of Lords Library posted the data for me. At least since 1900, and that works well enough for me. And so here we go.

Back to the 1930s?
Back to the 1930s?

So yeah, this is not an average prorogument. If you look at only proroguments that do not precede a general election—you need time for the campaigning and then hosting the actual election in those cases—this is the longest prorogument since 1930. (Also, a Parliament does not necessarily need to be prorogued before it is dissolved before an election. And that happened quite often in the 1960s, 70s, and 80s.)

And as I point out in the graphic, Parliament was prorogued during the depths of World War II to start new legislative sessions. But in those cases, Parliament opened the very next day, during a time of national crisis. One could certainly make the argument that Brexit is a national crisis. So wherefore the extraordinarily long prorogument? Well, quite simply, Brexit.

Credit for the piece goes to me.

Greenland, the 51st State?

If you haven’t heard, President Trump wants to buy Greenland from Denmark. So is Greenland going to beat Puerto Rico to joining the Union as the 51st state?


Not even close.

It would be the smallest state in terms of population, but also one of the smallest US territories. But in terms of area, Greenland dwarfs every state but Alaska. Though it still beats Alaska by almost 50% of its land area.

It's like a super-charged Seward's Folly
It’s like a super-charged Seward’s Folly

I had hoped to include some more economic data, but that will have to wait for a different post. Acquiring the population data was actually the most difficult—the US Census Bureau does not actually have easy to access data on the populations of US territories not called Puerto Rico.

This piece is mine.

From Frying Pan to the Fires of a War Zone

Moving away from climate change now, we turn to the lovely land of Afghanistan. While the Trump administration continues to negotiate with the Taliban in hopes of ending the war, the war continues to go worse for Afghanistan, its government, and its allies, including the United States.

It is true that US and NATO ally deaths are down since the withdraw of combat troops in 2014. But, violence and sheer deaths are significantly up. And as this article from the Economist points out, the deaths in Afghanistan are now worse than they are in Syria.

The beginning of the article uses a timeline to chart the history of Afghan conflicts as well as the GDP and number of deaths. And it is a fascinating chart in its own right. But I wanted to share this, a small multiples featuring graphic looking at the geographic spread of deaths throughout the country.

Getting hotter (because red obviously means heat)
Getting hotter (because red obviously means heat)

It does a nice job by chunking Afghanistan into discrete areas shaped as hexagons and bins deaths into those areas. All the while, the shape remains roughly that of Afghanistan with the Hindu Kush mountain range in particular overlaid. (Though, I am not sure why it is made darker in the 2003–04 map.)

To highlight particular cities or areas, hexagons are outlined to draw attention to the population centres of interest. But overall, the rise in violence and deaths is clear and unmistakable. And it has spread from what was once pockets in the south to the whole of the country that isn’t mountains or deserts.

Tamerlane would be proud.

Credit for the piece goes to the Economist graphics department.

How Mass Shootings Have Changed

A few weeks ago here in the United States, we had the mass shootings in El Paso, Texas and Dayton, Ohio. The Washington Post put together a piece looking at how mass shootings have changed since 1966. And unfortunately one of the key takeaways is that since 1999 they are far too common.

The biggest graphic from the article is its timeline.

Getting worse over time
Getting worse over time

It captures the total number of people killed per event. But, it also breaks down the shootings by admittedly arbitrary time periods. Here it looks at three distinct ones. The first begins at the beginning of the dataset: 1966. The second begins with Columbine High School in 1999, when two high school teenagers killed 13 fellow students. Then the third begins with the killing of 9 worshippers in a African Episcopal Methodist church in Charlestown, South Carolina.

Within each time period, the peaks become more extreme, and they occur more frequently. The beige boxes do a good job of calling out just how frequently they occur. And then the annotations call out the unfortunate historic events where record numbers of people were killed.

The above is a screenshot of a digital presentation. However, I hope the print piece did a full-page printing of the timeline and showed the entire timeline in sequence. Here, the timeline is chopped up into two separate lines. I like how the thin grey rule breaks the second from the third segment. But the reader loses the vertical comparison of the bars in the first segment to those in the second and third.

Later on in the graphic, the article uses a dot plot to examine the age of the mass shooters. There it could have perhaps used smaller dots that did not feature as much overlap. Or a histogram could have been useful as infrequently used type of chart.

Lastly it uses small multiples of line charts to show the change in frequency of particular types of locations.

Overall it’s a solid piece. But the timeline is its jewel. Unfortunately, I will end up talking about similar graphics about mass shootings far too soon in the future.

Credit for the piece goes to Bonnie Berkowitz, Adrian Blanco, Brittany Renee Mayes, Klara Auerbach, and Danielle Rindler.

Quantifying Part of the Opioid Crisis

Two weeks ago the Washington Post published a fascinating article detailing the prescription painkiller market in the United States. The Drug Enforcement Administration made the database available to the public and the Post created graphics to explore the top-line data. But the Post then went further and provided a tool allowing users to explore the data for their own home counties.

The top line data visualisation is what you would expect: choropleth maps showing the prescription and death rates. This article is a great example of when maps tell stories. Here you can clearly see that the heaviest hit areas of the crisis were Appalachia. Though that is not to say other states were not ravaged by the crisis.

There are some clear geographic patterns to see here
There are some clear geographic patterns to see here

For me, however, the true gem in this piece is the tool allowing you the user to find information on your county. Because the data is granular down to county-level information on things like pill shipments from manufacturer to distributor, we can see which pharmacies were receiving the most pills. And, crucially, which manufacturers were flooding the markets. For this screenshot I looked at Philadelphia, though I only moved here in 2016, well after the date range for this data set.

It could be worse
It could be worse

You can clearly see, however, the designers chose simple bar charts to show the top-five. I don’t know if the exact numbers are helpful next to the bars. Visually, it becomes a quick mess of greys, blacks, and burgundies. A quieter approach may have allowed the bars to really shine while leaving the numbers, seemingly down to the tens, for tables. I also cannot figure out why, typographically, the pharmacies are listed in all capitals.

But the because I lived in Chicago for most of the crisis, here is the screenshot for Cook County. Of course, for those not from Chicago, it should be pointed out that Chicago is only a portion of Cook County, there are other small towns there. And some of Chicago is within DuPage County. But, still, this is pretty close.

Better numbers than Philly
Better numbers than Philly

In an unrelated note, the bar charts here do a nice job of showing the market concentration or market power of particular companies. Compare the dominance of Walgreens as a distributor in Cook County compared to McKesson in Philadelphia. Though that same chart also shows how corporate structures can obscure information. I was never far from a big Walgreens sign in Chicago, but I have never seen a McKesson Corporation logo flying outside a pharmacy here in Philadelphia.

Lastly, the neat thing about this tool is that the user can opt to download an image of the top-five chart. I am not sure how useful that bit is. But as a designer, I do like having that functionality available. This is for Pennsylvania as a whole.

For Pennsylvania, state-wide
For Pennsylvania, state-wide

Credit for the piece goes to Armand Emamdjomeh, Kevin Schaul, Jake Crump and Chris Alcantara.

It’s Boris Time, Baby

Today Boris Johnson begins his premiership as the next prime minister of the United Kingdom. He might not be popular with the wide body of the British population, but he is quite popular with the Conservative base.

The Economist looked at how Boris polled on several traits, e.g. being more honest than most politicians, compared to his prime minister predecessors before they entered office. And despite being broadly unpopular outside the Tories, he still polls better than most of his predecessors.

Boris rates higher than many previous prime ministers before they came to power
Boris rates higher than many previous prime ministers before they came to power

Design wise, it’s a straight-forward use of small multiples and bar charts. I find the use of the light blue bar a nice device to highlight Boris’ position amongst his peers.

But now we see where Boris goes, most importantly on Brexit.

Credit for the piece goes to the Economist graphics department.

The Rent Is Too Bloody High

This is a graphic from the Guardian that sort of mystified me at first. The article it supports details how the rising rents across England are hurting the rural youth so much so they elect to stay in their small towns instead of moving to the big city.

But all those segments?
But all those segments?

The first thing I noticed is that there really is no description of the data. We have a chart looking at something from 1997 and comparing it to 2018. The title is more of a sentence describing the first pair of bars. And from that title we can infer that these bars are income changes for the specified move, e.g. Sunderland to York, for the specified year. But a casual reader might not pick up on that casual description.

Then we have the issue of the bars themselves. What sort of range are we looking at? What is the min? The max? That too is implied by the data presented in the bars. Well, technically not the bars, but in the numbers at the end of each bar. I will spare you the usual rant about numbers in graphics defeating the purpose of graphics and organisation vs. visual relationship. Instead, the numbers here are essential because we can use them to suss out the scale of the grey bars. After looking at a few bars, we can tell that the white lines separating the grey boxes are most likely 10% increments. And from that we can gather the minimum is about -40% and the maximum 100%. But instead of making the reader work to figure this out, would not some min/max labels at the bottom of the chart be far clearer?

And then there is the issue of the grey boxes/bars themselves. Why are they there in the first place? If the dataset were more about an unmet value, say reservoirs in towns were only at x% of capacity, the grey bars could relate the overall capacity and the coloured bars the actual values. But here, income is not a capacity or similar type of value. It could expand well beyond the 100% or decline beyond the -40%. These bars imply the values are trapped within these ranges. I would instead drop the grey bars entirely and let the coloured bars exist on their own.

Overall this is a confusing graphic for a fascinating article. I wish the graphic had been a little bit clearer.

Credit for the piece goes to the Guardian’s graphics department.