Create frequency distributions from raw data. Visualize data distribution with customizable bin widths.
Last updated: March 2026
Separate values with commas or spaces
1-50 bins recommended
| Range | Freq | Rel Freq |
|---|---|---|
| 23.0 – 34.0 | 2 | 10.0% |
| 34.0 – 45.0 | 4 | 20.0% |
| 45.0 – 56.0 | 4 | 20.0% |
| 56.0 – 67.0 | 3 | 15.0% |
| 67.0 – 78.0 | 4 | 20.0% |
| 78.0 – 89.0 | 3 | 15.0% |
A histogram is a graphical representation of the distribution of numerical data. It organizes data into "bins" (intervals) and displays how many observations fall into each bin using bars. The height of each bar represents the frequency (count) of observations in that interval.
Unlike bar charts which compare separate categories, histograms show continuous data distributions. They help identify patterns such as symmetry, skewness, multimodality, and outliers. Histograms are essential in exploratory data analysis and quality control.
The key parameter in histogram creation is the number of bins—too few bins hide detail, while too many bins create noise. A good rule of thumb is the square root rule: bins ≈ √n, where n is the number of observations.
Test Scores Distribution
There's no perfect answer. Common rules: Square Root Rule (√n), Sturges' Rule (1 + 3.322·log n), Scott's Rule, or Freedman-Diaconis Rule. Experiment to find what best reveals your data's structure.
Bar charts display categorical data with gaps between bars. Histograms show continuous numerical data with no gaps, representing frequency distributions. Histograms emphasize data shape; bar charts compare categories.
Skewness indicates asymmetry. Right-skewed (tail right) has high values pulled out; left-skewed (tail left) has low values pulled out. Symmetric histograms suggest normally-distributed data.
You can create unequal-bin histograms, but they require frequency density (frequency/width) on the y-axis for proper interpretation. Most standard histograms use equal bin widths.
Outliers can distort histogram appearance by creating very wide ranges. Consider: removing outliers, using log scale, or creating separate analysis. Sometimes outliers reveal important data characteristics.
Yes! The mode (most common value) corresponds to the tallest bar. Histograms can show unimodal (one peak), bimodal (two peaks), or multimodal (many peaks) distributions.
Count (frequency) shows raw number of observations. Relative frequency is count divided by total (shown as proportion or percentage). Use relative frequency to compare datasets of different sizes.
The first bin's lower edge equals the minimum value, ensuring all data fits within the bins. Bins are right-inclusive on the last bin (≤ upper) to capture the maximum value.
Related Tools