Histogram vs. Bar Chart: Key Differences and When to Use Each
Last updated: October 16, 2026
Histograms and bar charts are both made of rectangles, so they are easy to mix up. But they answer different questions. A histogram shows how numerical data is spread out. A bar chart compares values across separate categories. Pick the wrong one and your chart can mislead.
The one-question test
Is the x-axis a number line? If the things along the bottom are ranges of a measured quantity (0–10 kg, 10–20 kg…), you need a histogram. If they are names or labels (Paris, London, Tokyo), you need a bar chart.
On the left, the bins form a continuous scale from 0 to 60 minutes, so the bars touch and can’t be rearranged. On the right, the subjects are independent categories. They are sorted by size here, but any order would be correct.
Key differences at a glance
| Histogram | Bar chart | |
|---|---|---|
| What’s on the x-axis | Numbers on a continuous scale, grouped into ranges (bins) | Separate categories or labels |
| Question it answers | How are the values distributed? | How do these groups compare? |
| Bars | Touch each other; a gap means an empty bin | Separated by gaps |
| Order of bars | Fixed by the number line | Any order, often sorted by size |
| Bar width | Means something: the bin width | Means nothing; purely visual |
| Input data | Raw values, which you group into bins | One value per category |
| Typical examples | Heights, test scores, wait times, incomes | Sales by region, votes by party, favorite fruit |
Why the bars touch in a histogram
A histogram’s bins sit right next to each other on the number line: 10–20 ends exactly where 20–30 begins. Leaving a gap would suggest that values between 20 and 21 can’t exist. So in a histogram, a gap carries meaning: that bin is empty. In a bar chart, gaps simply separate categories that have nothing in between.
Tricky cases
Counting whole numbers (dice, number of children, goals per match)
These are numbers, but only whole ones. Both charts are used. A histogram with bins of width 1 centered on each whole number (1.5–2.5, 2.5–3.5…) keeps the number-line meaning, and that is what statistics courses usually expect. A bar chart of counts per value is also common and fine for a quick summary. See 300 rolls of two dice as a histogram.
Data that’s already grouped (age bands, income brackets)
A table like “18–24: 120 people, 25–34: 180 people” is still numerical data, so draw it as a histogram with touching bars. Watch out for unequal groups, such as 18–24 (7 years) next to 25–34 (10 years). A wider bin collects more people just because it is wider. Show density (count ÷ bin width) on the y-axis so the area of each bar, not its height, represents the count.
Rating scales (1–5 stars, “strongly disagree” to “strongly agree”)
These are ordered categories, not measurements: the gap between “agree” and “strongly agree” isn’t a measurable distance. Use a bar chart, keep the categories in their natural order, and don’t sort them by size.
Values over time (monthly sales, yearly population)
Neither. When each bar is a single value at a point in time, it is a bar chart of a time series, or better, a line graph. A histogram of monthly sales would instead show how often sales fell into each range, ignoring which month they happened in.
When to use each
Use a histogram when you want to…
- See the shape of numerical data: symmetric, skewed or with two peaks.
- Find where most values lie and how spread out they are.
- Spot outliers, gaps or data-entry errors.
- Compare the distribution of two groups, such as test scores before and after a course.
Use a bar chart when you want to…
- Compare totals or averages across categories, such as sales by store.
- Rank items from largest to smallest.
- Show survey answers or counts for named groups.
Common mistakes
- Adding gaps to a histogram because “charts have gaps”. It hides the continuity of the data.
- Sorting histogram bins by height. The order is the number line; sorting destroys the shape.
- Making a “histogram” from a list of categories. If the labels are words, it’s a bar chart.
- Using unequal bin widths with counts on the y-axis, which exaggerates the wider bins.
Ready to make one? Paste your numbers into the histogram maker; it chooses sensible bins and keeps the bars touching. New to histograms? Start with What is a histogram?