Thursday, December 22, 2016

Math Thursday:
The math of life and death, part 1
Population pyramids

Over the next few weeks, I'll be covering population statistics. This week's graphics come from the website, which has a set of population pyramids for nearly every country on earth as well as the entire world population (shown to the left) and regions like Western Europe and Western Asia.

Note: I've shrunk down the screen shots so they fit in the blog format. You can click on any picture to get a larger version or visit the btlas website to get more data about different countries or different years.

The population is split into five year cohorts, starting at 0-4. The pink and blue bars represent the number of females and males respectively in that cohort rounded to the nearest tenth of a percent. Let's look at a few features of the data.

Mode: In data sets where the size is a knowable number, the mode would be the largest cohort and any ties for largest, which is this case would be just the 0-4 cohort with 4.4% female and 4.7% male. But in sets like this where the exact number is changing, a cohort counts as a mode if it is greater than or equal to both the cohort immediately below and immediately above. In terms of the graph, if we have pinches and bulges, any bulge can be a mode, even if it is not the widest overall. In this case, the 25-29 cohort is larger than both the 20-24 group and the 30-34 group, so it counts as a mode, though it isn't nearly as large as the 0-4 cohort.

The age at which the female population surpasses the male: It is typical that there are more males under five years old than females and also typical that males die at higher rates than females. At ages less than 50, men are dying faster than women in large part because of accidents. Eventually, the proportion of females catches up and surpasses the males. In this case, the 45-49 cohort is tied at 3.1% for both males and females and the 50-54 cohort is tied at 2.8%. All of the cohorts older than 50-54 have more females than males.

Youth, Adults and Seniors: One way to group these cohorts is to look at the Youth for 0 to 19, the Adults from 20 to 64 and the Seniors, 65 and older.  In this case, the percentages add up to 34.0% Youth, 57.4% Adults and 8.4% Seniors. (if often happens that the total isn't 100% due to rounding error.) It is assumed that the Adult cohort are the wage earners for the most part, caring for the Youth and Seniors. It's a bad sign when the Adult cohort dips near to 50%.

Population growth: The blue curve with the cross hairs shows overall population growth. According to these numbers, the world population in 1950 was about 2.5 billion, it is now around 7.4 billion and projected to be 11.2 billion in 2100.


Here are the numbers for the good old U.S. of A. Birth rates have been decreasing, in large part due to declining teen pregnancy rates, so the 0-4 cohort does not count as a mode at all. The three modes here are at 20-24, the very tail end of the Baby Boom at 50-54 and a much smaller bulge at 40-44. The Youth/Adult/Senior split is 25.3%, 59.8% and 15.0%, more adults and seniors than the worldwide numbers and less kids.

The age at which women outnumber men is the 55-59 cohort. You'll notice the 60-64 cohort is represented by exact numbers instead of percents. That is a feature of the site when the cursor scrolls over a cohort bar. In my age group, there are about 600,000 more women than men. Lucky me.

The website speculates that the population will keep increasing throughout the century. My best guess is the numbers we have worldwide and nationally now are not truly sustainable. The best guess fifty years ago was we would have a famine large enough the reduce the population, but food production has kept up with population so far. We have paid for that with increased carbon emissions and terrifying air and water pollution in developing nations throughout Asia.

That shit is gonna catch up with us, as mathematicians are fond of saying.

And then we have the Russians, our new best buddies and a completely failed state. We have a boatload of pinches and bulges, so much so that 75-79 is a mode due to World War II. We have a crazy pinch when comparing 20-24 vs. 15-19, the kind of fall off that is usually indicates a major war, but this time was instead the result of the fall of the Soviet Union. Women outnumber men in the 35-39 cohort, due in no small part to young men killing themselves with alcohol. The very pronounced squeeze from 0-4 to 15-19 is typical in a country with a horrible medical system that can't keep babies alive. The good news in all of this is their Adult cohort vastly outnumbers their Youth and Seniors, so they should be able to support their children and elderly.

And then we have the population curve. The population of Russia has decreased over the past few decades, though the decline is leveling off. The author of the website, a Belgian named Martin De Wulf, thinks there will be even more decline in the decades to come.

I won't be around to collect, but if I was betting on the world population in 2100, I'd say there's a good chance it will be less than it is today. While this would good news for the planet and our chances for long-term survival, it's more than a little depressing that it will take the whole world turning into the kind of shithole Russia has been for decades.

More in the first comment.

1 comment:

Matthew Hubbard said...

I was very insulting to Russia in this post partly as an experiment. Russia dominates the visits to this webpage, but I assume it's almost all bots. If Russian commenters come here to defend their country's honor, that will be evidence that I'm wrong about the bots. If they don't, my hypothesis looks good.