A histogram displays quantitative (numerical) data, not categorical data. If you need to visualize categorical data, the chart you’re looking for is a bar chart. The two look similar at first glance, but they represent fundamentally different types of information.
Why Histograms Are Quantitative
A histogram’s x-axis shows a continuous number line, broken into ranges called bins. Each bin covers a span of values (say, 0 to 10, 10 to 20, 20 to 30), and the height of each bar tells you how many data points fall within that range. The y-axis shows either a raw count (frequency) or a percentage (relative frequency). Common examples include test scores, heights, incomes, or wait times.
The key detail: the x-axis is always drawn to scale. The numbers have a meaningful order and consistent spacing, which is what makes the chart quantitative. You can’t rearrange the bars in a histogram the way you might sort a bar chart from tallest to shortest, because the position of each bar corresponds to a real numerical range.
How Bins Shape What You See
Because quantitative data can contain a huge range of values (or even infinite values for continuous measurements like weight or temperature), a histogram groups those values into bins. Choosing bin width matters more than most people realize. Too few bins and you lose meaningful detail. Too many bins and the chart becomes noisy and hard to read.
The same dataset can look quite different depending on how you define the bins. With one bin width, a distribution of test scores might appear perfectly symmetric with a clear peak in the middle. Change the bin width, and that same data can look slightly skewed to one side. A set of bins aligned to letter grade cutoffs (60–70, 70–80, 80–90) might tell a more intuitive story than bins offset by five points (65–75, 75–85). There’s no single correct bin width, but the goal is to reveal the overall shape of the data without distorting it.
Histogram vs. Bar Chart
The confusion between histograms and bar charts is understandable because both use rectangular bars. Here’s how they differ:
- Data type: Histograms show quantitative data (numbers). Bar charts show categorical data (labels like “New York,” “London,” “Tokyo”).
- X-axis: A histogram’s x-axis is a scaled number line with continuous ranges. A bar chart’s x-axis lists a finite set of categories.
- Bar spacing: Histogram bars touch each other because the bins represent adjacent, continuous ranges with no gaps in between. Bar chart bars typically have spaces between them because the categories are distinct and unrelated.
- Bar order: You can reorder bars in a bar chart (alphabetically, by size, or any other way) without changing the meaning. Rearranging histogram bars would destroy the information, since position on the number line is part of the data.
If your x-axis labels are words or names, you’re building a bar chart. If your x-axis labels are numbers representing ranges, you’re building a histogram.
Can Histograms Show Discrete Data?
Histograms are most commonly associated with continuous data (measurements that can take any value, like height or time), but they also work for discrete quantitative data. Discrete data involves countable whole numbers, like the number of pets in a household or the number of emails received per day. These are still numerical values with a meaningful order, so a histogram is appropriate.
For discrete data with only a few possible values, a bar chart sometimes works just as well, since each value acts almost like a category. But once the range of possible values gets large (say, the number of steps someone walks in a day), grouping those values into bins and using a histogram becomes far more practical.
What a Histogram Tells You
The whole point of a histogram is to reveal the shape of a distribution. When you look at one, you’re trying to answer a few questions: Where do most values cluster? Is the data symmetric or skewed to one side? Are there any unusual gaps or outliers? Is there one peak or multiple peaks?
A histogram of exam scores might show a bell-shaped curve centered around 75, telling you most students scored near that mark with fewer students at the extremes. A histogram of home prices in a city might be heavily skewed to the right, with most homes in a moderate range but a long tail stretching toward expensive outliers. These patterns are exactly the kind of insight histograms are designed to surface, and they only make sense when the underlying data is numerical.