data visualisation – Coffee Spoons

Covid Update: 28 July

Another week, more bad news when it comes to Covid-19 in the states of Pennsylvania, New Jersey, Delaware, Virginia, and Illinois. Last week I wrote about how the slight upticks in new cases in the states were not so slight. This week it’s more of the same, though it would be fair to say that the spread of Covid is beginning to accelerate, although in some states more quickly than others.

To start, the most basic of refreshers. Pennsylvania, New Jersey, and Delaware are all in the Northeast. Virginia is in the South. Illinois is in the Midwest. At a broad level, Covid is beginning to strike places with low vaccination rates and in particular has been hitting the South and Midwest fairly severely. You might say that Virginia has the northern suburbs of Washington and Illinois has Chicago, but in both states simply head south and west and you’ll very quickly find yourself in an entirely different culture, almost a different state in both. And it is in those parts of the states where the increase is most noticeable.

Conversely, the Northeast has had some of the highest rates of vaccination and the new Delta variant is taking longer to take root. Because, that’s the entire point of the vaccination process. We want to starve the virus of potential host bodies. But we needed to reach much higher than the 50-odd percent of the states to be fully vaccinated. So where are we? Let’s take a look.

New cases curves for PA, NJ, DE, VA, & IL.

The recent tails are curving up now and it’s plain to see. Two weeks ago the tails were barely noticeable in most states. You had to really look at the numbers to see the differences. Even last week in Pennsylvania and Delaware the tails were still fairly flat. This week is unmistakable. You may be over the pandemic and done with Covid, but the pandemic isn’t over and it’s not done with us.

In both Virginia and Illinois the seven-day averages for new cases rose by nearly 50% in the last week. Yesterday Illinois reported more than 2,000 new cases, the first single-day report that high since the beginning of May. And in Virginia we had more than 1,000 cases yesterday, again the most in a single day since the end of April.

In the tri-state area we also saw increases over the last week, albeit not as great as in Virginia and Illinois. Though they weren’t off the pace by much. In fact, at this rate, I would not be surprised if next week I write how both Pennsylvania and New Jersey see more than 1,000 cases in a single day. Neither state is there yet, as New Jersey reported 830 yesterday and Pennsylvania only 645—though it reported 986 on Tuesday. Both states’ seven-day averages are also a bit behind Virginia and Illinois, but again, I would not be surprised to see them nearing 1,000 though maybe not surpassing it this time next week.

I cannot overstate this next part, however. The virus is primarily infecting, sickening, and killing the unvaccinated amongst us. If you have not received your vaccine shots yet, please, please do. The vaccines have been proven safe. They have been proven effective. And they are free.

The good news, if we want to find some, is that the death rates largely continued to fall overall.

Largely. Keep in mind as I noted above, the deaths are almost all occurring in those who have not been vaccinated. Now in Virginia we are seeing the death rate increase, though not yet dramatically. Last week the seven-day average was three per day. This week we are at four.

But compare that to three of the other four states. (Delaware continued to have almost deaths and its average is zero.) Last week both Pennsylvania and New Jersey had averages of 5 deaths per day and this week they area at 4. And in Illinois the average fell from 7 to 4.

What I will be watching over the next week or two is whether these death rates begin to increase beyond Virginia. I mentioned before how deaths are a lagging indicator and we are beginning to reach the point at which, during earlier waves, where we could begin to see increasing numbers of deaths.

But again, the key is for any and all my readers who are unvaccinated, please make an appointment to get your shots as soon as possible.

Credit for the piece is mine.

Olympic Recap/Retro

Every four years (or so) I have to confess that I think fondly back upon my former job, because I worked with a few wonderful colleagues of mine on some data about the Olympics. And the highlight was that we had a model to try and predict the number of medals won by the host country as we were curious about the idea of a host nation bump. In other words, do host countries witness an increase in their medal count relative to their performance in other Olympiads?

We concluded that host nations do see a slight bump in their total medal count and we then forecast that we expected Team GB (the team for Great Britain and Northern Ireland) to win a total of 65 medals. We reached 64 by the final day and it wasn’t until the women’s pentathlon when, in maybe the last event, Team GB won a silver medal bringing its total to 65, exactly in line with our forecast.

Probably the most Olympics I’ve ever watched.

Of course we also looked at the data for a number of other things, including if GDP per capita correlated to Olympic performance. We also looked at BMI and that did yield some interesting tidbits. But at the end of the day it was the medal forecast that thrilled me in the summer of 2012.

So yeah, today’s a shameless plug for some old work of mine. But I’m still proud of it two olympiads later.

If you’d like to see some of the pieces, I have them in my portfolio.

Credit for the piece is mine.

But Where Are the Spiders?

Yesterday I mentioned more about revolutions, well today we’re talking about Mars, a planet that revolves around the Sun. Late last week scientists working with the InSight lander on the Red Planet published their findings. Turns out we need to rethink what we know about Mars.

First, the planet is probably much older than Earth. Mars’ composition also differs from Earth in some significant ways. InSight mapped the interior of Mars by studying the seismic waves (think like sound waves but through the inside of planets) of marsquakes.

The Wall Street Journal published a great article spelling out the findings in detail that is well worth the read. It also included some nice graphics helping to support the piece. The one I wanted to highlight, however, was a brilliant comparison of Mars to Earth.

But how many licks to get to the centre?

Conceptually this is important, because many diagrams and graphics I’ve seen about these findings only detail the interior of Mars. But what makes Mars important is how it differs from Earth, and let’s be honest, how many of us remember our Earth science classes at school and can diagram out the interior of Earth?

And right here the designer compares the smaller—and now older—brother of Earth. Again, read the article for the details, but in short, some of the key findings are that the core is larger, but also lighter, than we thought. Our planet’s core differs because Earth has two parts: a solid and heavy ball of iron and nickel surrounded by a liquid core that spins. That spinning core creates the magnetic fields that protect our planet from the Sun and have kept our atmosphere intact. Mars doesn’t have that. And that’s in part because, given the core is larger than we thought, the mantle is smaller.

A smaller mantle means that certain materials couldn’t form that insulate the Earth’s core. So while Earth’s core has been prevented from cooling and slowing down, Mars was not. And so while it did have a magnetic field at one point, that slowing, cooling core slowly dissipated the magnetic field. That may be why the planet, once rich in water, now is a barren rock exposed to the Sun.

Again, this is a big deal in terms of planetary science. Consider that Earth and Mars are broadly made of the same materials that orbited the Sun billions of years ago. But Mars took those same ingredients and made itself into a very different planet. And now we know quite a good deal more about the Red Planet.

One last point to make about the graphic above. Again, many illustrations will increase the size of the crust to make it more visible. Here the designer chose to keep it more in proportion to the scale of the planets’ interiors. (Even though Mars’ crust is quite a bit thicker than Earth’s.) I think that’s important because it puts us into perspective. We can build monuments like the Pyramids that last thousands of years and dig bore holes miles deep and tunnel out connections through mountain ranges, but that also just scratches the surface of the crust. But that crust is the thinnest of shells over the mantle and cores of these planets.

That life began and took hold on Earth, on that thinnest of shells protected by a magnetic field because of a spinning molten core buried at the centre of the planet…something to think about sometimes.

Credit for the piece goes to Merrill Sherman.

Covid Update: 22 July

I know that I typically post more light-hearted content on Fridays. But after taking a holiday Monday and my internet being down all day yesterday—fun fact, trying to type up a blog post on my mobile is…well suffice it to say I am glad that it is only an emergency backup plan—we are going to enter the weekend with some updates on the spread of Covid-19 in Pennsylvania, New Jersey, Delaware, Virginia, and Illinois.

A little over a week ago I wrote ab o ut the emergence of slight upticks in our region of coverage. Nine days hence, those slight upticks are no longer so slight.

New case curves for PA, NJ, DE, VA, & IL.

In all five states we are seeing significant increases in the numbers of new cases. In New Jersey we hit the milestone of 900,000 total cases. Three weeks ago, the Garden State had bottomed out with an average of about 160 new cases per day. Yesterday the moving seven-day average surpassed 500 for the first time since mid-May.

But New Jersey is not alone. In Illinois the seven-day average has doubled since my last post. The state reported just under 2000 new cases yesterday and that brought the seven-day average above 1000 for the first time since late-May. Before the spring surge earlier this year, Illinois had bottomed out with a seven-day average of about 1500 new cases per day. If these numbers continue, I would not be surprised to see the state back at those levels within another week or two.

Nor has Virginia been spared and yesterday Old Dominion’s seven-day average rose to just short of 500 new cases per day. That too is a level we haven’t seen since…wait for it…mid-May. The pattern emerging here is fairly clear. Things are beginning to get worse and get worse in a hurry. The average has nearly doubled since my last post about Covid.

We see the same situation in Delaware though it’s increase is on the smaller side. The seven-day average in the First State hit 50 new cases per day earlier this week, though it’s ebbed back slightly and sat at 48 yesterday. You have to go back to early June to find levels that high.

And finally it’s not as if Covid has spared Pennsylvania. Three weeks ago we were nearing 100 new cases per day. At the time of my last post, the seven-day average was just a little over 200 new cases per day. The average is double that number as of yesterday at 438 new cases per day. The Commonwealth has now reported three consecutive days of over 500 new daily cases for the first time since…early June. Surprised at that timing?

The common thread in these states is that the virus, almost certainly the Delta variant, is racing through the unvaccinated populations. And we are beginning to see rises in the numbers of hospitalisations, something that I don’t chart but on which I keep tabs. But what I do track visually are the deaths from Covid.

We need to keep in mind again that deaths are a lagging indicator and that it takes a few weeks from someone’s infection to their death, if that person’s case is severe enough.

The good news is that we are not yet seeing a corresponding increase in the numbers of deaths. Though it might still be too early for that, should it occur. Of course lack of death does not preclude one from suffering from long Covid, a debilitating and persistent illness that lasts for months. As for deaths, instead we see the numbers entering something of a holding pattern.

Starting with Pennsylvania, deaths from Covid-19 have halved since this time last month. But we were largely at that point when I last wrote nine days ago. Deaths haven’t fallen significantly since, dropping from 7.3 per day to 5.3 per day.

And that’s a pattern we see elsewhere. In Illinois deaths are also down nearly half from an average of 12 per day last month to just 7 per day. But they were at 9 when I last wrote. In New Jersey deaths were averaging 6 per day last month and now they sit at an average of 4.7. What was the average nine days ago? 4.7. Virginia is also similar, going from an average of 6 deaths per day to 4 and now just 3.

Delaware is the one state where we have some genuinely good news. Deaths had been averaging about 2 per day last month and we are at 0 today. Since the beginning of July we’ve seen only three additional deaths. That’s great news, but Delaware is also one of the smallest states.

The takeaway from this should be please get vaccinated if you’re still not. They are safe. They are effective. And they are free. Especially as this new Delta variant begins ripping through unvaccinated populations, I really hope my unvaccinated readers will take an opportunity this weekend to get their shots. Because remember, except for the Johnson & Johnson vaccine, you need two shots and you need to wait between those two doses. And then the vaccine takes about two weeks to kick in after that second dose.

Credit for the piece is mine.

The Western Heat Dome(s)

For the last two days Philadelphia and much of the East Coast suffered from a heavy haze of smoke that blanketed the region. This wasn’t just any smoke, however, but smoke from the wildfires on the West Coast. This post isn’t about the wildfires, but rather something that exacerbated them. We are talking about the heat domes that formed earlier this summer. The ones that melted trolley cables in Portland.

This was a nice print graphic in the Guardian Weekly, a magazine to such I subscribe that had several articles about the domes.

It does a nice job of showing the main components of the story and sufficiently simplifying them to make them digestible. One quibble, however, is how in the second map how oddly specific the heat dome is depicted.

The first graphic in particular is more of an abstraction and simplified illustration. But here we have contours and shapes that seem to speak with precision about the location of this heat dome. It also contains shades of red that presumably indicate the severity of the heat.

There’s nothing wrong with that, but it stuck me as odd juxtaposed against the top illustration.

Credit for the piece goes to the Guardian Weekly graphics department.

Substandard Housing in Philadelphia

I took a holiday yesterday and headed down the street to the Philadelphia City Archives, which houses some of the oldest documents dating back to the founding of the colony. But I was there primarily to try and find deeds and property information for my ancestors as part of my genealogy work.

When I walked into the building—the archives moved a few years ago from an older building in University City into this new facility—an interactive exhibit confronted me immediately. Now I did not take the time to really investigate the exhibit, because I anticipated spending the entire day there and wanted to maximise my time.

But there was this one graphic that felt appropriate to share here on Coffeespoons.

Philadelphia’s population crested in the 1950 census, it would decline continually until the 2010 census.

Like a lot of statistical graphics from the mid-20th century we have a single-colour piece because colour printing costs money. It makes use of a stacked bar chart to highlight the share of housing in the city that can be classified as substandard, i.e. dilapidated or without access to a private bath.

The designer chose to separate the nonwhite from the white population on different sides of the date labels, though the scale remains the same. I wonder what would have happened if the nonwhite bars sat immediately below the white bars within each year. That would allow for a more direct comparison of the absolute numbers of housing units.

That would then free up space for a smaller chart dedicated to a comparison of the percentages that are otherwise written as small labels. Because both the absolutes and percentages are important parts of the story here.

The white housing stock increased and the number of substandard units decreased in an absolute sense, leading to a strong decline in percentages.

But with nonwhite housing, the number of substandard units slightly increased, but with larger growth in the sheer number of nonwhite housing units overall, that shrank the overall percentage.

Put it all together and you have significant improvements in white housing, though in an absolute sense there still remain more substandard units for whites than nonwhites. Conversely, we don’t see the same improvements in housing for nonwhites. Rather the improvement from 45% to 35% is due more from the increase in housing units overall. You could therefore argue that nonwhite housing did not improve nearly as much as white housing between 1940 and 1950. Though we need to underline that and say there was indeed improvement.

Anyway, I then went inside and spent several hours looking through deed abstracts. Not sure if those will make it into a post here, but I did have an idea for one over a pint at lunch afterwards.

Credit for the datagraphic goes to some graphics person for some government department.

Credit for the exhibit goes to Talia Greene.

Covid Update: 13 July

So Mondays no longer work for these regular updates, because as we know Illinois no longer reports weekend data. Starting next week, neither will Virginia. Furthermore, keeping track of the vaccinations is tough, because the same. But also, then we have Pennsylvania, which includes Philadelphia, but Philadelphia only updates vaccination data twice per week.

Consequently, I’m not sure what I’m going to continue doing. But at the least, these updates of cases and deaths could continue, because after one cycle the zero numbers for Saturday and Sunday will average out.

The question is where are we today?

Well when it comes to new cases, we are seeing slight upticks across the board.

But whereas the upticks were slight last week in Pennsylvania and Delaware, they were a little bit greater this week. Still, we are far from drastic upswings, but they are noticeable. Look at the very tail end of Delaware and you will see a slight change in the curve’s slope. Pennsylvania’s is, for now, less noticeable.

In New Jersey, where cases had still been declining last week, the numbers are now heading back up. You can see this as well in the chart with a sudden little jump in the last day or so.

Meanwhile in Virginia and Illinois, the upward swings have clearly begun and they are plainly visible. But the numbers are increasing, because in Virginia today’s seven-day average is now the highest it’s been since the end of May. In Illinois, yesterday’s average was higher than today’s, but both are about the same as the average was in the beginning of June.

State departments of health indicate these increases are mostly all in unvaccinated people. And that’s not terribly surprising given that the new Delta variant beginning to take root in the United States is far more effective at viral spread than its predecessors. The worry is that the variant may be more lethal.

And to that point we are also seeing the seven-day average for deaths rebound in most states.

First, the exception. Delaware has now gone over a week without any Covid-19 death and its average now sits at 0.0.

In all other states, the trend is pretty clear if not visible in the charts. Illinois is the most obvious where the recent rise in deaths from Covid-19 can be seen in the sharp jump of the orange line at the tail end of the chart. The state is now flirting with double-digit death rates after hitting 11 and 13 deaths per day Sunday and Monday.

Elsewhere, we have numbers creeping up, but still below the levels we saw two weeks ago. On 28 June, Pennsylvania averaged 12 deaths per day. That had fallen to 5 for last week’s write-up, but today it sits at 7. New Jersey went from 8 to 4 but is back up to 5. And Virginia went from 6 to 3 and is now at 4.

Again, these are not catastrophic increases, to be clear. However, after several weeks of declining numbers of deaths the death rates are climbing once more. As with new cases, state departments of health point to the deaths being in the unvaccinated populations.

If you haven’t been vaccinated yet, I encourage you to do so. They have been proven both safe and effective. And if cost is your concern, they’re free. This new Delta variant can make you sick even if you’re only partially vaccinated—only one of the two required shots for Pfizer and Moderna. Factor in the month+ you need from your first of the Pfizer and Moderna doses through the second and the two-week waiting period, it’s critical you schedule your shots as soon as possible.

I’ll try to look at writing up the numbers again next Tuesday or Wednesday, and over the course of the week I’ll be following these numbers. Though I’m not entirely certain I’ll continue posting them daily to my social media feeds. (In fairness, I’ve been busy enough to preclude me from doing that the last two weeks.)

Credit for the piece is mine.

Covid Update: 6 July

In trying to limit my Covid-19 updates to Monday, bank holidays definitely affect the schedule. For my international audience, that meant Monday was a day of no posting. It also meant that some states—Illinois—were not reporting data. Add to that Illinois had already stopped reporting data on Saturday and Sunday, I wanted to wait until we had Tuesday’s data before putting it all together. And so here we are.

Last week I discussed “divergent patterns at the margins”. We saw some states continue their progress in decreasing the numbers of new cases with falls in their seven-day averages. You could group the tri-state area in this category. Then in the other group we had Virginia and Illinois where the seven-day average had begun to rise.

In the week since, this pattern largely held the same. Both Virginia and Illinois continued to see rising numbers of new cases. Compared to 20 June, before this pattern emerged, Virginia’s seven-day average sat at 129 new cases per day and Illinois was just slightly worse at 156. Yesterday those numbers were 180 and 328, respectively. A far cry from an emergency, yes, but also not ideal.

What about the tri-state area? Well we can now lump Pennsylvania and Delaware in with Virginia and Illinois since both states saw a rise in their seven-day averages. Back on 28 June Pennsylvania sat on an average of 177 new cases per day and Delaware was at 19. Yesterday those numbers were 181 and 27, respectively.

The difference here is that in both Pennsylvania and Delaware this recent rise is still below the numbers from 20 June. On that date Pennsylvania’s seven-day average was 261 new cases per day and Delaware’s was 28. So it’s not great, but it’s still not bad either.

How about New Jersey? The Garden State continues to see declining numbers of new cases. From the 20th to the 28th to yesterday the average has fallen from 179 to 176 to 162. Certainly not dramatic, but it’s progress nonetheless.

With deaths we saw broad and general progress, however, so that’s good.

Last week I mentioned how I had hoped we would see Pennsylvania’s seven-day average of deaths slip into the single digits. Instead it had climbed higher. Well, Pennsylvania finally fell into the single digits and rests at 5 deaths per day.

When we look at the remaining states we see good news across the board. In Illinois the seven-day average fell from 9 to 7 deaths per day. In two states the numbers fell by half. New Jersey fell from 8 to 4 and Virginia went from 6 to 3. Finally, Delaware now averages just 0.1 deaths per day.

I don’t have the data on vaccination, because there are some holes and I want to see if I can fill that data out. But in the three states we track, we are talking about less than a percentage point increase in fully vaccinated people over the course of over a week. That continues to be not ideal.

Credit for the piece is mine.

We’re Gonna Need a Bigger Boat

A little over a week ago the Philadelphia Inquirer posted an article about sharks. It wouldn’t be the American holiday of 4th of July without mentioning Jaws. Think of it, there really are no good Hollywood films about the Constitutional Convention or Declaration of Independence. I mean we have Mel Gibson’s The Patriot. But, that’s terrible. I digress, we’re here to talk about sharks.

A great white shark swam past the Jersey coast in mid-June. She swam just off the shore of Atlantic City’s beaches before moving on towards a shark nursery in more northern waters near New York and New Jersey.

Why are we discussing it? Because sharks have always fascinated me. It’s my blog. Oh, and it had a map.

There isn’t much to say about this map, it makes a good use of contrasting red for the shark’s path against the light blue of the ocean.

Hopefully none of you were eaten by a shark over the holiday.

Credit for the piece goes to John Duchneskie.

Yep, Still Hotter

Like I said yesterday, I wanted to compare cities, surprise, Philadelphia vs. Chicago. And so with some extra time I was able to finish this graphic that took the data from Climate Central to compare the two cities.

What you can see below is that Philadelphia has seen more significant temperature growth in both summer highs and summer lows. And, importantly, the increase in low temperatures, i.e. nighttime, has been greater than that of daytime highs. That means that we have less of an opportunity to cool down after a hot summer day, adding stress to the system.

Chicago on the other hand has seen less overall growth, though it’s still present. And there too we see the same pattern of greater increases in low, i.e. nighttime, temperatures than of daytime highs.

It’s remarkable to think that the flat where I lived seven of my eight years in Chicago had no air conditioning unit in the bedroom, only in the living room. It was, of course, an older concrete building from the 1960s/70s when, as the chart above shows, nighttime temperatures didn’t really require air conditioning.

But like I said yesterday, I’m just glad I’ve been able to crank the air conditioning the last several days.

Credit for the piece is mine.