How to read a chart

In the 2023 bar of a chart of Korean births, 230.0 means 230,000 babies, because the axis is labelled in thousands.

A chart turns a table of numbers into shapes, and our eyes are quick with shapes. That speed is the reason to draw a chart, and it is also the risk: you can take in the shape before you know what it shows. Good chart readers do the slow part first. They read the words around the picture, and only then look at the bars or the line. This unit uses one real data set, the number of babies born in Korea each year from 2015 to 2025, to practise that habit.

Four things to read before the shape

First, the title: what was counted, where and when. Second, the vertical axis: what the numbers measure and in what unit. Third, the horizontal axis: here it is the year, but on other charts it may be a category, an age group or a measurement. Fourth, the source and the date. The births chart below comes from the Ministry of Data and Statistics, the Korean agency formerly called Statistics Korea, in its report "Birth Statistics 2025". A chart with no source is a chart you cannot check, and a chart with no date may describe a world that has already changed.

Units change the meaning of every bar

The vertical axis of the births chart is labelled "thousand". The 2015 bar reaches 438.4, which means 438,400 births, and the 2023 bar reaches 230.0, which means 230,000. Leave out the unit and a reader could think 438 babies were born. Units also hide in words such as "per 1,000 people", "percent" or "children per woman". The total fertility rate, which you will meet in the unit on line graphs, is measured in children per woman, so 0.72 does not mean 72 percent of anything.

Difference, ratio and percent change

Once the labels are clear, read values by sliding across from the top of a bar to the axis. Then compare them in one of three ways. The difference between 2015 and 2023 is 438.4 − 230.0 = 208.4 thousand births. The ratio is 230.0 ÷ 438.4 ≈ 0.52, so the 2023 bar is a little more than half as tall as the 2015 bar. The percent change is (230.0 − 438.4) ÷ 438.4 ≈ −47.5%. After 2023 the numbers rose again: from 230.0 to 254.3 thousand in 2025, a rise of 24.3 thousand or about 10.6%. Notice how the same chart supports both "births nearly halved" and "births rose two years in a row". Both are true; they describe different stretches of time.

What a chart cannot tell you

A chart shows what happened, not why. The births chart cannot tell you whether the 2024 and 2025 rise came from more marriages, from the age structure of the population or from policy. Questions like that need other data and careful reasoning. Keep the description of the picture and the explanation of the picture apart, in your head and in what you write.