Free Histogram Maker

How to read skewness and what it means for your data

Skewness measures how lopsided a distribution is. The maker reports it as a signed number and translates it into plain English so you can act on it without a stats textbook open.

Right-skewed (positive skew) means a long tail toward higher values. Household income is the textbook example: most people cluster in a modest range, but a small number of very high earners drag the tail out to the right. The mean sits noticeably above the median because those extreme values pull the average up. Reporting a mean by itself for right-skewed data can be misleading — the median usually tells a truer story about a typical value.

Left-skewed (negative skew) means a long tail toward lower values. Exam scores near a ceiling look like this: if a test was easy for the class, most students bunch up near 100 and a few strugglers pull the tail down. The mean falls below the median, and again the median is often the more honest summary.

Approximately symmetric is the classic bell shape. Heights, measurement errors, and many biological quantities behave this way. Mean and median line up, skewness is near zero, and the mean is a perfectly good summary of the typical value.

The maker prints the numeric skewness alongside the label, so you can distinguish "mildly skewed" from "seriously skewed" — anything with an absolute value under about 0.5 is close to symmetric, 0.5 to 1 is moderate, and above 1 is heavy.

What is a histogram?

A histogram is a chart that shows how a set of numbers is distributed. The range of values is split into consecutive intervals called bins, and the height of each bar shows how many data points fall inside that interval. Because the bars sit directly next to each other, a histogram makes the overall shape of your data obvious at a glance — where values cluster, how spread out they are, and whether the distribution is symmetric or lopsided.

Histogram vs bar chart

It is easy to confuse a histogram with a bar chart, but they answer different questions. A bar chart compares separate categories — sales by country, votes by candidate — and the bars usually have gaps between them. A histogram works with continuous numeric data, the bars touch, and the order of the bars is fixed because the x-axis is a number line. If your x-axis is made of labels, you want a bar chart; if it is made of numbers grouped into ranges, you want a histogram.

How to choose the number of bins

The number of bins changes the story a histogram tells. Too few bins hide detail and the chart looks blocky; too many bins create a noisy, spiky shape. A common starting point is the square-root rule — take the square root of the number of data points and round up. Sturges’ rule (log₂n + 1) and the Rice rule (2 × n^(1/3)) are popular alternatives. This maker shows all three suggestions for your dataset so you can compare and pick whichever reads best.

Reading the shape of a distribution

Once your histogram is drawn, look at its silhouette. A roughly symmetric, bell-shaped curve points to a normal distribution. A long tail to the right is right-skewed, which is common for income or waiting times, while a long tail to the left is left-skewed. Two separate peaks suggest your data may be a mix of two groups. The maker measures skewness automatically and tells you what the shape implies, along with where the mean sits relative to the median.

When not to use a histogram

Histograms are powerful but not perfect. The same data can look very different depending on bin width, they hide the exact values inside each bar, and they struggle with very small samples. For tiny datasets a dot plot or stem-and-leaf plot can be clearer, and for comparing several groups side by side a box plot often works better. A good habit is to try two or three bin counts before settling on the chart you publish.