Histograms

Learn how to use histograms to understand the distribution of a numerical variable.

Skill Level: Beginner
Prerequisites: Scatter Plots
Estimated Time: 25 minutes

Story Time

You have already learned how to compare categories with bar charts, show trends with line charts, and explore relationships with scatter plots.

Now we move to a different kind of analytical question:

How are the values of one numerical variable distributed?

For example:

• Are most mathematics scores clustered in the middle?
• Are reading scores spread widely or narrowly?
• Are there unusually low or high scores?
• Does the data appear balanced, skewed, or uneven?

Histograms help answer these questions.

In this lesson, you will learn how to create histograms with Matplotlib and how to interpret what they reveal about a variable’s distribution.

Learning Objectives

By the end of this lesson, you will be able to:

• explain when a histogram should be used
• create a histogram with Matplotlib
• understand the role of bins
• interpret shape, center, and spread
• identify possible outliers or unusual patterns
• distinguish a histogram from a bar chart

Why This Topic Matters

Many important data-analysis questions involve a single numerical variable.

Examples include:

• How are exam scores distributed?
• Are most values concentrated in one range?
• Is the data spread out or tightly grouped?
• Are there unusually high or low observations?

A histogram helps answer these questions by grouping numerical values into ranges called bins and showing how many observations fall into each range. This makes it easier to study the overall distribution of the data.

What Is a Histogram?

A histogram is a chart used to display the distribution of a numerical variable.

It works by:

• dividing the range of values into intervals called bins
• counting how many observations fall into each bin
• drawing bars to represent those counts

The x-axis shows ranges of values, and the y-axis shows frequency, or how many observations fall in each range.

When Should You Use a Histogram?

Use a histogram when:

• you are working with one numerical variable
• you want to understand how the values are distributed
• you want to inspect shape, concentration, spread, or possible outliers

For example:

• mathematics scores → histogram
• reading scores → histogram
• age → histogram
• salary → histogram

A histogram is used for distribution questions, while a bar chart is used for comparison questions across categories.

Your First Histogram

You can create a histogram in Matplotlib with plt.hist().

import matplotlib.pyplot as plt

math_scores =[5][6][7][8][9][3]

plt.hist(math_scores)
plt.show()

This groups the score values into bins and displays how many values fall into each range.

Adding a Title and Axis Labels

As with other charts, labels help the viewer interpret the plot more clearly.

import matplotlib.pyplot as plt

math_scores =[6][7][8][9][5][3]

plt.hist(math_scores)
plt.title("Distribution of Mathematics Scores")
plt.xlabel("Math Score")
plt.ylabel("Frequency")
plt.show()

Loading interactive histogram…

Now the chart clearly communicates what variable is being shown and what the bar heights represent.

Understanding Bins

Bins are the score ranges used to group the data.

For example, one bin might include values from 50 to 60, another from 60 to 70, and another from 70 to 80.

Matplotlib chooses default bins automatically, but you can also set the number of bins yourself:

import matplotlib.pyplot as plt

math_scores =[7][8][9][5][6][3]

plt.hist(math_scores, bins=5)
plt.title("Distribution of Mathematics Scores")
plt.xlabel("Math Score")
plt.ylabel("Frequency")
plt.show()

The choice of bins affects how the distribution looks, so it is important to remember that different bin settings can change the visual appearance of the histogram.

What a Histogram Helps You See

A histogram helps you understand several important features of a distribution:

• shape — Is the distribution balanced, skewed, or uneven?
• center — Where do most values seem to be concentrated?
• spread — Are the values tightly grouped or widely spread out?
• unusual values — Are there very low or very high observations?

These ideas are central to descriptive data analysis.

Reading the Shape of a Histogram

When reading a histogram, the first thing to notice is the overall shape.

You might ask:

• Is there one main peak?
• Are there two peaks?
• Is the data concentrated in the center?
• Is the distribution skewed to one side?

A histogram with most values clustered in the middle and fewer values at the ends may look roughly balanced. A histogram with a long tail on one side may appear skewed.

Understanding Center

The center of a distribution is the area where many values seem to cluster.

For example, if most bars are tallest around scores from 65 to 75, that suggests the center of the distribution is in that range.

A histogram does not directly calculate the mean or median, but it gives a useful visual impression of where most values lie.

Understanding Spread

Spread refers to how widely the data values are distributed.

If most bars are packed into a narrow range, the spread is smaller.

If the values extend across many bins from low to high, the spread is larger.

Understanding spread helps analysts judge whether the data is tightly grouped or highly variable.

Spotting Unusual Values

A histogram can also suggest possible outliers or unusual observations.

For example, if most scores are between 60 and 85 but one or two scores appear far away from the rest, those values may deserve closer attention.

Histograms do not label outliers as precisely as some other charts, but they can still reveal unusual gaps or extremes in the data.

Histogram vs Bar Chart

Histograms and bar charts may look similar, but they are used for different purposes.

A bar chart is used for:

• categories
• grouped comparisons
• separate labels such as gender or lunch type

A histogram is used for:

• one numerical variable
• grouped ranges of values
• studying distribution

For example:

• gender counts → bar chart
• mathematics score distribution → histogram

Also, histogram bars usually touch because the bins represent continuous ranges of values, while bar chart bars are separated to show distinct categories.

Example: Mathematics Scores

Suppose you want to study the distribution of mathematics scores.

import matplotlib.pyplot as plt

math_scores =[10][2][8][9][5][6][7]

plt.hist(math_scores, bins=6, color="skyblue", edgecolor="black")
plt.title("Distribution of Mathematics Scores")
plt.xlabel("Math Score")
plt.ylabel("Frequency")
plt.show()

A useful interpretation might be:

• most scores fall in the middle to upper-middle range
• there are fewer very low and very high scores
• the distribution appears concentrated around the center rather than spread evenly across all values

Example: A Wider Spread

Now imagine a second set of scores:

import matplotlib.pyplot as plt

scores =[11][8][9][5][7]

plt.hist(scores, bins=6, color="orange", edgecolor="black")
plt.title("Distribution of Scores")
plt.xlabel("Score")
plt.ylabel("Frequency")
plt.show()

A strong interpretation might be:

• the scores are spread across a wider range
• there is more variation in the data
• the distribution does not appear as tightly clustered

Reading a Histogram as an Analyst

When interpreting a histogram, ask:

1. What variable is being shown?
2. What does the x-axis represent?
3. What does the y-axis represent?
4. Where are most values concentrated?
5. Is the distribution narrow or wide?
6. Does the shape appear balanced or skewed?
7. Are there unusually low or high values?

For example:

Most mathematics scores are concentrated in the middle range, with fewer values at the extremes.

That is a stronger interpretation than simply saying, “The histogram has bars.”

Google Colab Activity

Using the Students Performance dataset, create a histogram for:

• mathematics score

Then write a short interpretation that answers:

• Where are most scores concentrated?
• Does the distribution seem narrow or wide?
• Are there any unusually low or high values?

Practice in Colab

Notebook

Open in Google Colab

Practice histograms with short runnable examples and exercises.

After completing the activity in Google Colab, mark it as complete below.

Practice Exercise

Create a histogram for reading score.

Then answer:

• What score range appears most common?
• Does the distribution seem balanced or skewed?
• Is the spread smaller or larger than you expected?

Check Your Understanding

1. When should you use a histogram?

When comparing categories such as gender or lunch typeWhen studying the distribution of one numerical variableWhen showing the relationship between two numerical variablesWhen displaying change over time

2. Which Matplotlib function creates a histogram?

plt.bar()plt.scatter()plt.hist()plt.plot()

3. What does the y-axis of a histogram usually represent?

CategoriesFrequencyCorrelationTime

4. What are bins in a histogram?

Category labelsRanges of numerical valuesColors used in the chartTitles placed above the bars

5. What is one common mistake when interpreting a histogram?

Reading the overall shape of the distributionThinking it is the same as a bar chart for categoriesLooking at concentration and spreadUsing it for numerical data

Challenge Exercise

Choose one numerical variable from your dataset and create a histogram that clearly shows its distribution.

Your chart should include:

• a meaningful title
• clear axis labels
• an appropriate number of bins
• a short written interpretation

Your interpretation should mention shape, concentration, and spread.

Common Mistakes

Confusing a Histogram with a Bar Chart

A histogram shows the distribution of one numerical variable using value ranges. A bar chart compares categories. They may look similar, but they answer different questions.

Using Too Few or Too Many Bins

The number of bins can change how the distribution appears. Too few bins can hide detail, while too many bins can make the chart look noisy.

Looking Only at the Tallest Bar

The histogram should be read as a whole. Focus on the overall shape, center, and spread, not just one bar.

Forgetting to Interpret the Distribution

Do not stop after creating the plot. Ask what the histogram suggests about clustering, spread, and unusual values.

Key Takeaways

In this lesson, you learned how to:

• create histograms with Matplotlib
• use histograms to study the distribution of one numerical variable
• understand the role of bins
• interpret shape, center, spread, and possible unusual values
• distinguish histograms from bar charts

Continue Your Journey

Next, you will learn Boxplots, which provide another powerful way to summarize distributions and identify outliers.