Generate stem-and-leaf plots with frequency counts, back-to-back comparison mode, key interpretation, and summary statistics. Auto-detects stem unit.
The stem-and-leaf plot calculator builds stemplots from raw data, preserving the original values while still showing the overall distribution shape.
Unlike a histogram, which groups values into bins, a stem-and-leaf plot keeps each observation visible. That makes it useful for teaching, small-to-medium datasets, and quick exploratory analysis where you want both the shape and the exact values in the same display.
This page supports both single-dataset stemplots and back-to-back comparisons, with automatic stem-unit detection and a frequency summary for each stem.
Stem-and-leaf plots are still one of the clearest ways to examine small and medium datasets without throwing away the original values. They make medians, gaps, clusters, and repeated values easy to spot at a glance.
The back-to-back layout is especially helpful when you want to compare two groups on a shared scale without reducing them to summary statistics alone.
For each value x: stem = floor(x / unit), leaf = round((x/unit − stem) × unit). Stem unit is auto-detected from the data range: range ≤ 10 → unit=1, ≤ 100 → unit=10, ≤ 1000 → unit=100.
Result: Stem-and-leaf plot with stems 6–9, each showing sorted leaves
6|2 8 9 → values 62, 68, 69. 7|1 2 4 6 7 9 → values 71, 72, 74, 76, 77, 79. 8|1 3 4 5 6 8 → values 81, 83, 84, 85, 86, 88. 9|0 1 1 3 5 → values 90, 91, 91, 93, 95. Leaves are always read in order. The 7 stem has the most values (6), showing the concentration in the 70s.
John Tukey introduced the stem-and-leaf display in his 1977 book "Exploratory Data Analysis." It was designed as a quick, hand-drawable alternative to histograms that preserves the original data values. Despite the rise of computer graphics, stemplots remain a standard topic in introductory statistics courses because they teach students to think about distribution shape while staying close to the data.
Both stemplots and dot plots show individual values. Dot plots place a dot for each value on a number line; stemplots group values by leading digits. For small datasets (< 20 values), dot plots are often clearer. For moderate datasets (20–100), stemplots are more compact. Both are "EDA tools" in Tukey's framework, designed for quick understanding rather than formal presentation.
Beyond basic and split stemplots, there are: (1) rounded stemplots where values are rounded to smooth noise, (2) truncated stemplots where leaves are truncated (not rounded) for simplicity, (3) five-stem plots where each stem is split into five parts (leaves 0–1, 2–3, 4–5, 6–7, 8–9) for fine-grained shape analysis. These variations offer different trade-offs between resolution and readability.
Last updated:
A stem-and-leaf plot (stemplot) is a way to display quantitative data that shows individual values while revealing the distribution shape. Each number is split into a "stem" (leading digits) and a "leaf" (trailing digit). For example, 85 becomes stem=8, leaf=5. Values with the same stem are grouped on the same row, with leaves sorted in order.
Read the key first (e.g., "7|2 = 72"). Then for each row, combine the stem with each leaf to get the original values. Row "7|1 2 4 6" means 71, 72, 74, 76. The shape of the plot — which rows have the most leaves — shows the distribution. Count the leaves for frequency; read them for exact values.
A back-to-back stemplot displays two datasets sharing the same stems. Dataset 1 leaves extend to the left (reversed order), Dataset 2 leaves extend to the right, and stems are in the center column. This makes it easy to compare distributions, medians, and spreads of two groups side by side.
Use a stemplot when: (1) you have a small to moderate dataset (15–150 values), (2) you want to see exact values, not just bin counts, (3) you want to quickly find the median (count to the middle leaf), (4) you're doing exploratory analysis. Use a histogram when you have hundreds+ of values or need to present to a general audience.
A split stemplot divides each stem into two parts to create more rows and spread out the leaves. Typically, leaves 0–4 go on the first half-stem (marked with *) and leaves 5–9 on the second (marked with .). This is useful when most data falls in just a few stems, creating overcrowded rows.
Outliers appear as isolated stems with very few leaves, separated by empty stems (gaps) from the rest of the data. For example, if most data has stems 5–8 and there's a lone "12|3", the value 123 is visually obvious as an outlier. The gap between stems 8 and 12 immediately signals something unusual.