The dot plot calculator above draws one dot for every observation in your dataset, stacked above the exact value it takes. Nothing is grouped into ranges, nothing is averaged, and no width has to be chosen. It also reports the count at each distinct value, the observations, the number of distinct values, the median and the range, and it names the gaps, clusters and isolated points visible in the display.
The distinction from the other one-variable displays is the whole point of the page, so Arb Digital will state it plainly. A histogram collapses observations into bins of a chosen width and draws a bar per bin; a frequency distribution presents those same bins as a table; a stem-and-leaf plot splits each number into a leading stem and a trailing digit; a box plot discards the individual values entirely and draws the five-number summary. A dot plot alone keeps every observation as a separate, countable mark at its own value.
What This Dot Plot Calculator Does
It parses your data, groups identical values, and draws a stack of dots at each one. The height of a stack is the frequency of that value, and the number of dots across the whole display equals the number of observations. There is no binning parameter because there are no bins.
The one rounding it applies is a stacking precision, which decides how many decimal places two numbers must agree to before they count as the same value. This exists because measured data almost never repeats exactly: eleven readings of 3.14159, 3.14161 and so on would each get their own stack of one and the display would tell you nothing. Setting the precision to two decimals stacks them together at 3.14. That is a deliberate, visible, reversible choice, and it is not the same as binning — the label still names an exact value rather than an interval.
For whole-number data the tool can keep the empty positions on the axis, drawing a labelled row with no dots where a value simply does not occur. This is what makes a gap legible. Turning it off compresses the display to only the values present, which is more compact and hides exactly the structure a dot plot exists to reveal.
Alongside the drawing, the shape panel reports what the display shows: whether the counts are roughly symmetric, which tail is longer, whether there is more than one local peak, and whether any value sits separated from the rest by a run of empty positions. Those are the four things a reader is supposed to take from a dot plot, and stating them in words makes the display checkable.
How to Use It
- Paste your data using any mix of spaces, commas, semicolons or line breaks. Non-numeric entries are ignored rather than treated as zero.
- Set the stacking precision. Zero for counts and scores, one or two for measurements, higher only if repeated values really do agree that closely.
- Keep the empty positions on for whole-number data, so gaps stay visible.
- Read the stacks. Each row is one value; the dots are the observations at it, and the number after them is the count.
- Check the shape panel for the symmetry, the peaks, the gaps and any isolated point, then confirm the total matches your sample size.
How the Plot Is Built
The construction is deliberately trivial, and that is its strength. Sort the values, tally identical ones, and draw a column of that many dots above each distinct value on a common horizontal scale. Wolfram MathWorld's entry on the dot plot gives the same construction and the code that produces it.
Work the default dataset through by hand: 4, 5, 5, 6, 6, 6, 7, 7, 8, 9, 9, 12. That is twelve observations. Tallying gives one at 4, two at 5, three at 6, two at 7, one at 8, two at 9 and one at 12 — seven distinct values, and the counts sum to 1 + 2 + 3 + 2 + 1 + 2 + 1 = 12, which matches. The tallest stack is at 6, so 6 is the mode.
With twelve values the median sits between the sixth and seventh in order, which are 6 and 7, giving 6.5. The range is 12 − 4 = 8. The mean is 84 ÷ 12 = 7.0, above the median, which is the arithmetic signature of the right tail you can see in the drawing.
The interesting part is the axis. Positions 10 and 11 have no observations at all, so the value at 12 stands two clear positions away from the rest of the data. That two-position gap is the entire reason to draw a dot plot for this dataset: a histogram with a bin width of 5 would put 12 in the same bar as 9, 10 and 11 and the separation would vanish without trace.
Why No Bin Width Is a Real Advantage
Every histogram requires two arbitrary choices: how wide the bins are and where the first one starts. Both change the picture, sometimes dramatically. The same forty numbers can look symmetric with one bin width and bimodal with another, and there is no rule that settles which is right. The NIST/SEMATECH e-Handbook's section on the histogram sets out what the display shows and why the bin choice governs it.
A dot plot removes both choices. There is exactly one dot plot of a dataset, so two analysts looking at the same numbers see the same picture and cannot disagree about the display before they start disagreeing about the data. For a class exercise, a quality-control sample or any dataset under about a hundred points, that reproducibility is worth more than the smoother shape a histogram gives.
The cost is that dot plots do not scale. Past a few hundred observations the stacks grow too tall to count, repeated values become the norm rather than the exception, and the display degrades into a ragged histogram with the worst possible bin width. That is the point to switch, and it is a genuine limit rather than a matter of taste.
What to Read Off a Dot Plot
Four features, in roughly this order. Centre: where the mass of dots sits, which the median in the grid pins down numerically. Spread: how far the dots reach, summarised by the range but better judged by eye, since the range is fixed by the two most extreme points alone.
Shape: whether the stacks fall away evenly on both sides, or trail off in one direction. A longer right tail pulls the mean above the median, and the gap between the two is the quickest numerical check on what your eye is telling you. Our mean, median and mode calculator gives all three at once for exactly this comparison.
Gaps and isolated points: runs of empty positions, and single dots stranded beyond them. This is where the dot plot beats every alternative. A box plot will flag a point as an outlier only if it falls beyond 1.5 times the interquartile range from the box, which is a rule, not an observation; a dot plot simply shows you that the point is alone. The NIST/SEMATECH section on the box plot explains what that summary keeps and what it necessarily throws away. Our outlier calculator applies the formal rules if you need a decision rather than a look.
Arb Digital builds free tools that draw the data rather than a summary of it.
Browse All Free Tools Talk To Our TeamCommon Mistakes to Avoid
- Dropping the empty positions — a compressed axis puts adjacent stacks side by side and makes a two-unit gap look identical to no gap at all.
- Setting the stacking precision too high on measured data — every observation gets its own stack of one and the display becomes a flat, useless row.
- Using a dot plot for hundreds of points — the stacks become uncountable and you have built a badly binned histogram by accident.
- Reading a lone dot as an error — a separated point is a fact about the data, not a verdict on it, and it may be the most informative observation you have.
- Comparing two dot plots on different scales — the axes must match, or the apparent difference in spread is an artefact of the drawing.
Related Free Tools From Arb Digital
Bin the same data into bars with the histogram calculator, tabulate those bins with the frequency distribution calculator, split the digits instead with the stem and leaf plot calculator, reduce it to a five-number summary with the box plot calculator, get the three averages with the mean, median and mode calculator, or apply the formal rule with the outlier calculator. The full free online tools hub lists every statistics tool we publish.
Frequently Asked Questions
A histogram groups observations into bins of a chosen width and draws one bar per bin. A dot plot draws one dot per observation at its exact value, so nothing is grouped and no bin width has to be chosen.
It sets how many decimal places two numbers must agree to before they stack together. It is the only rounding applied, and the label still names an exact value rather than an interval.
Because a run of positions with no dots is what a gap in the data looks like. Compressing the axis to only the values present makes a two-unit gap indistinguishable from none.
Past roughly a hundred the stacks become too tall to count and the display degrades. At that size a histogram or a box plot conveys the shape far more efficiently.
It shows isolated points directly, as a dot separated from the rest by empty positions. It applies no rule, so whether that point counts as an outlier remains your judgement.
A stem-and-leaf plot splits each number into a leading stem and a trailing digit and prints the digits themselves. A dot plot places an anonymous mark at the value, which is easier to count and harder to misread.
Counted. The total number of dots equals the sample size exactly, which makes a dot plot the easiest display to verify against the underlying data.
Yes. Negative values are placed correctly on the axis, and decimals stack according to the precision you set. Empty positions are only filled in for whole-number data.
This page explains a standard exploratory data display for educational purposes. It processes only the numbers you paste into it, entirely in your own browser, and sends nothing anywhere.
