A choropleth map is one of the most familiar map types: districts, counties, or countries shaded from light to dark according to a number, such as population density, rainfall, or income. It is also one of the easiest maps to make badly, because the choice of class breaks and colours can make the same data tell very different stories. Done honestly, however, a choropleth is a powerful way to show geographic patterns at a glance. This guide walks through a first choropleth in QGIS, with emphasis on the decisions that keep the map truthful and readable.

A choropleth needs two things joined together: polygon boundaries and a numeric attribute for each polygon, with a shared key such as a district code or name. Check the join before styling anything: count how many polygons matched, look for spelling differences in names, and watch for polygons with missing values, which will draw blank or with a default colour. Normalise counts when areas differ in size, because raw totals favour large areas even when rates are low. Population density, cases per thousand people, or rainfall per area are usually more honest than raw counts. If your table has text where numbers should be, clean it in a spreadsheet first, because QGIS classification tools need real numeric fields.

Classification means grouping continuous values into a small number of classes, often five, each with one colour. Equal intervals divide the range into equal-sized bins and work well when values spread evenly. Quantiles put the same number of polygons in each class and work well for ranking, but they can hide large gaps between values. Natural breaks look for clusters in the data and suit uneven distributions. There is no universally correct method, so try two methods on the same data, compare the maps, and pick the one that reflects real gaps in the numbers rather than the one that looks most dramatic. State the method in the map legend or notes, so readers know how the shades were chosen.

Use a single-hue ramp, light to dark, for data that runs low to high, and a diverging ramp, with a neutral middle, only when the data has a meaningful midpoint such as above and below average. Avoid rainbow ramps, which imply categories where none exist and are hard for colour-blind readers to follow. Keep the number of classes small, usually four to six, so neighbouring shades stay distinguishable in print and on screen. Test the map in greyscale: if the classes still read in order from light to dark, the ramp is sound. Add a clear legend with the value range for each class, a title that names the variable and year, and a source credit.

Before exporting, zoom out and ask what the map claims at a glance. Does one huge, sparsely populated area dominate the eye even though its rate is ordinary? That is the classic choropleth trap, and a note about area versus rate, or a companion chart, can correct it. Check that no polygon is missing, that labels do not hide the shading, and that the title, legend, and source are all present. Export at a resolution that keeps text readable, and keep the QGIS project file with the data so you can revise the breaks later. A first choropleth made slowly, with documented choices, teaches more than a polished-looking map made from default settings.

A good choropleth is a data decision before it is a design decision. Clean and normalise the values, choose breaks that reflect real gaps, use a restrained colour ramp, and document the method. Follow that order and your first shaded map will be one you can defend, not just one that looks impressive.