Bogus forecast

The Economist describes the boom in digital mapping with a story including the most preposterous graph of 2017.

The graph appears to be a solid, clean, better than average business bar chart.

But it only has one actual data point! And we don’t learn what amount that is. There is a phrase for an untethered forecast reaching more than 30 years into the future. It is made up.

History and forecast

A recent Bloomberg article describes the rapid growth of solar power generation. This graph compares the trend for all renewables to the main fossil fuels used for electricity generation.

The graph is nice and clean. It includes the main message in the title and subtitle. There are only three curves and the have distinct colors. (But shouldn’t the renewables be green?) The axes tick labels are simple, which is adequate since we are not invited to pay attention yearly details.

The graph would be improved if the forecast parts of the curves were indicated. For example using dashes.

China is installing solar capacity much faster than any other country.

The message is well expressed in the subtitle. And it’s a relief that the graph is not a pie chart. The heights of the columns are much easier to compare than the areas of slices would be. The emphasis is on the first and tallest bar, so it does not need to be a different color for emphasis.

Interestingly the article states that the primary motivation in China is concern in the population about air pollution and environmental degradation, while the main driver in the United States is economic.

Clutter

A recent article in The Economist is titled Women alone are driving a recovery in workforce participation. An accompanying graph is suggestive but far too cluttered.

Long point labels usually don’t work. A horizontal bar or a dot graph would work well for the long category names. Unfortunately neither is available in Excel. This graph provides the idea.

The evidence is a bit more nuanced. The one category dominated by women had the largest growth. The three categories dominated by men had the smallest growth. The more mixed categories had mixed growth.