Course contentsShow
← Data Science
Lesson 1220 of 2,14531. Data Visualization PrinciplesFree lesson

Histograms for Continuous Distributions

Understand how histograms display the distribution of continuous variables through binning and frequency counts.

Histograms for Continuous Distributions

What you'll learn: How histograms reveal the shape and spread of continuous data through smart binning and counting.

The Core Idea

A histogram is like sorting a pile of coins by their minting year into labeled cups—each cup represents a range (bin), and the height shows how many coins fall into that range. For continuous numerical data (like height, temperature, or response time), a histogram groups values into intervals and displays the frequency of observations in each interval as bars.

Unlike bar charts that show discrete categories (which you just learned), histograms show continuous distributions. The bars touch each other because the data flows continuously from one bin to the next—there are no natural gaps.

How Histograms Work

The x-axis represents your continuous variable divided into equal-width bins. The y-axis shows the count (or sometimes proportion) of data points falling within each bin. The resulting bar heights reveal the distribution's shape: is it symmetrical? Skewed left or right? Does it have multiple peaks?

Binning matters: Too few bins and you miss important details; too many bins and you see noise instead of pattern. Think of it like adjusting camera focus—you want just enough resolution to see the true structure.

What Histograms Reveal

  • Central tendency: Where most values cluster
  • Spread: How wide or narrow the distribution is
  • Shape: Normal (bell-shaped), skewed, uniform, or multimodal
  • Outliers: Isolated bars far from the main distribution
  • Gaps: Ranges where no data exists

This is your first deep look at continuous distribution visualization—building on the distribution checks you performed during EDA profiling.

Key Takeaway: Histograms transform continuous data into a visual frequency distribution through binning, revealing shape, spread, and concentration patterns that summary statistics alone cannot show.