Population pyramids don’t have to be tedious, inefficient, and ineffective. The traditional ones (bar charts, males to the left, females to the right) certainly are. Since demographers look at population pyramids to detect specific shapes, I don’t think we should change the basic structure (population along the x-axis, age groups along the y axis). Other than that, we can discuss everything:
- Using lines instead of bars can multiply the number of populations in a single chart or a population over multiple points in time.
- If you are displaying a single population (or very few), consider “folding” the axis and represent male and female on the same side, making comparisons much more straightforward.
- In most cases, the split by sex is not relevant since the population is almost symmetrical. You’ll probably find a more exciting comparison (change over time, other regions) when that happens. The header above, for Qatar, shows that the split by sex is relevant there, so you should always check.
The population pyramids below show how the population is changing in each country. I used estimates (1950-2019) and projections (2020-2100) from the UN Population Division (2019 revision). Each series displays population by single year age groups (both sexes) in a given year, from 1950 (darker color) to 2100 (lighter color). The charts share the same scales (ages from 0 to 100+ along the y axis, and 0% - 50% along the x-axis).