GIS is all about mapping data. Whether you’re visualizing trends, highlighting patterns, or presenting raw facts, the way you map your data is critical for how your audience understands it. But you’re not just throwing data on a map; you’re telling a story. In this lesson, you’ll explore the essentials of mapping qualitative and quantitative data and discover how to make complex information clear and compelling using classification techniques.
Think about the last map you saw that changed your view on something. What made it clear or confusing?
Descriptive or nominal information that categorizes features, such as land-use types, political boundaries, or road classifications.
Mapping qualitative data
Many times, you’ll find yourself mapping qualitative data, such as land-use types, political boundaries, or road classifications. Creating these types of maps often involves assigning unique symbols, colors, or patterns to represent each category. For example, you might assign colors such as green to forests, yellow to residential areas, and gray to industrial zones.
Most GIS software displays polygons in a default solid color when you add a polygon layer to your map. The color choice may seem arbitrary at first, but think of it as a blank canvas ready to be customized with colors that best represent your data.
Driving your visual color palette choices are two common attribute types:
- Descriptive attributes, like land-use type or road type
- Identifiers, such as country names or state codes
Using textures or patterns instead of colors is a classic cartography trick—essential for maps that will be printed in black-and-white or viewed by people with color vision differences.
Not all maps will be in color. For black-and-white printing or accessibility purposes, consider using textures or patterns to distinguish categories. (Refer to the earlier section, “Designing for accessibility,” for some tips).
What challenges might arise if you use only colors to distinguish categories in your map?
Numerical information that measures or counts features, such as population, temperature, or income.
Mapping quantitative data
When you work with numbers, statistics, or other measurements, maps are powerful tools for visualizing trends. How you display that data is just as important as the data itself. Here are three strategies for effective mapping:
- Show raw counts: If your goal is to show raw data (like population), map the numbers as is. Be careful, though: Large numbers don’t always tell the full story. Imagine looking at the number of newspaper subscribers in New York City — a population of millions — versus Santa Fe, New Mexico, a city with fewer than 100,000 people. This type of map would simply reflect the area’s population. Normalizing the data, explained in the next bullet, would make it more meaningful.
- Normalize your data: Raw numbers can be misleading when areas vary in size or population. To make comparisons fairer, normalize your data by converting it into rates, percentages, or densities. For example, instead of total population, map population density (people per square mile) or crime rates per 1,000 residents. This way, you can compare large and small areas more accurately.
- Use ranking: Replace raw values with rankings to show relative differences. For example, ranking cities by their average income can make comparisons easier to grasp.
City planners often use normalized data, like crime rates per 1,000 residents, to allocate resources more equitably across neighborhoods—ensuring small communities aren’t overlooked just because their raw numbers are lower.
Consider your audience. Do they need raw numbers or are they better served by percentages, rates, or rankings? Tailor your approach to their needs, and be sure to express the data appropriately.
When might raw numbers be more useful than normalized data for a map’s audience?
Creating classes
Imagine trying to map 3,000 U.S. counties, each with its own unique shade or pattern. It would look as though a rainbow had exploded. Users would find it visually overwhelming and impossible to interpret. By grouping similar values into classes, you reduce the visual clutter, simplifying the story your map tells. For example, you can group the 3,000 counties into five classes of population ranges.
When creating classes, you need to make two big decisions:
-
How many classes do you need?
Fewer classes mean simpler maps but may hide important details. Too many classes can make maps hard to read. A good rule of thumb is to aim for 4 to 7 classes.
-
How should you divide the data into groups?
The method you choose depends on your data and what you’re trying to show. Here’s a guide to common methods:
- Equal interval: This method divides data into equal-sized ranges; for example, 1–100, 101–200, 201–300, and so on. The equal-interval method is simple but can hide patterns if your data isn’t evenly distributed. This method works best when you have data ranges that are fairly well-known, such as percentages or temperatures.
- Defined or fixed interval: You set the range size, and the software calculates the number of classes. This method works great if you have meaningful ranges in mind; for example, test scores grouped into intervals of 20 points become 1–20, 21–40, 41–60, and so on.
- Quantiles: With this approach, each class has the same number of features. This method is great for evenly distributing data visually, but it can distort values if you have outliers in your data. To address this issue, try increasing the number of classes.
- Natural breaks: This method is driven by data and finds clusters and gaps in your values. It’s a great way to highlight natural patterns in your data. This method is also my personal favorite because it’s more likely to represent how the data should logically be grouped. For example, you can use natural breaks to group housing prices to show clear divisions between low, medium, and high costs.
- Geometric intervals: This method groups data by creating class ranges that increase in size at a consistent rate, such as by doubling or tripling. It’s ideal for datasets that grow quickly or unevenly, such as population growth rates, for which small values are common but large values stand out. (In QGIS, this method is called logarithmic scale.)
- Standard deviations: This method is rooted in statistical analysis, dividing data based on their distance from the mean (average). It works best if your data are distributed in a bell shape (that is, normally distributed), like showing temperature anomalies above or below average.
Want to go deeper? The science behind classification choices
Statistical classification methods, like natural breaks (Jenks optimization), use algorithms to minimize variance within classes and maximize variance between classes. This makes patterns clearer but can introduce subjectivity, so always consider your map’s purpose and audience when choosing a method.
Practitioners often try different classification methods in their GIS software and preview the results—choosing the one that best balances clarity and accuracy for their intended audience.
Understanding the different classification types is one thing, but how do you know which one to pick? It comes down to two simple questions: What story do you want to tell, and are your data evenly distributed, clustered, or skewed? For example:
- If you want to highlight natural clusters, use natural breaks.
- If you’re comparing values relative to an average, go with standard deviations.
- If you’re mapping percentages and temperatures, equal interval might work best.
The right classification method isn’t always the most scientific approach; instead, it’s the one that makes your data easy to understand and tells the story you want to share. Just be sure your choices are clear and defendable.
More classes always mean a more detailed and better map.
Too many classes can overwhelm users and hide patterns—clarity often comes from simplifying to just a few meaningful groups.
- Qualitative and quantitative data require different mapping approaches
- Classification helps reduce visual clutter and clarify the story your map tells
Create a legend for a map showing three types of land use: forest, residential, and industrial. Decide on colors or patterns for each, and explain your reasoning.
- List the three land use types.
- Choose a color or pattern for each (write or sketch).
- In one or two sentences, explain why you made these choices and how they help the map reader.
Reflect on a map you’ve seen or created: How did the classification method or symbol choices affect your interpretation of the data? What might you do differently next time?
What does it mean to normalize data in mapping?
Tap to revealIt means converting raw data into rates, percentages, or densities to allow fair comparisons between areas of different sizes or populations.
What are “natural breaks” in data classification?
Tap to revealNatural breaks are class breaks determined by clusters and gaps in the dataset, highlighting patterns that naturally occur in the data.
Why might you use textures or patterns instead of colors on a map?
Tap to revealTo ensure that maps remain readable in black-and-white prints or for users with color vision differences.
The way you classify and symbolize your data can make the difference between a confusing map and a powerful, story-driven communication tool.
Choosing the right classification method depends on both your data’s distribution and the story you want your map to tell.
By grouping similar values into classes, you reduce the visual clutter, simplifying the story your map tells.
Imagine you are mapping income data for a set of cities. Which classification method would you choose and why?
By converting raw numbers into or densities, you can make fairer comparisons between areas of different sizes.
Every map with more classes gives viewers more useful information.
How confident are you that you can choose an appropriate classification method for mapping quantitative data?