Type a list of numbers — 2, 4, 4, 4, 5, 5, 7, 9 — and this statistics calculator returns the five-number summary, the range and IQR, both the population and the sample variance and standard deviation, and any outliers past the 1.5×IQR fence — every answer recomputed from the raw data and proved.
✓100% free — no paywall, ever
✓No sign-up, no app
✓Every answer verified before you see it
In short
A dataset has a centre and a spread, and the centre alone misleads: two classes can share a mean of 70 while one clusters tight and the other sprawls. This statistics calculator measures the spread — the five-number summary, the range and IQR, the population and sample variance and standard deviation, and outliers — with every answer recomputed from the raw numbers and proved.
Four ideas carry this page
Quartiles, the IQR fence, and the two variances with their square roots. The solver applies all four in the steps, so the method — not just the answer — is what you take away.
Five-number summary
min,Q1,median,Q3,max
Range and IQR
R=max−minIQR=Q3−Q1fences: Q1−1.5IQR,Q3+1.5IQR
Variance — both kinds
σ2=n∑(xi−xˉ)2s2=n−1∑(xi−xˉ)2
Standard deviation
σ=σ2s=s2
Worked examples — a classic set, an outlier, five values, decimals
A textbook dataset with an exact standard deviation, a list where one value breaks away, five values with Tukey’s halves on show, and decimals carried exactly. Each check below is the same recomputation this statistics calculator prints on every problem.
2, 4, 4, 4, 5, 5, 7, 9
1.Sorted: 2, 4, 4, 4, 5, 5, 7, 9 — count 8, sum 40, mean 5
2.Five-number summary: min 2, Q1 4, median 4.5, Q3 6, max 9
3.Range 7, IQR 2; fences 1 and 9 — 9 sits on the fence, so no outliers
exact fractions throughout — no rounding until the final decimal
Proof: the exact lane carries fractions end to end; decimals appear only alongside, labelled ≈.
Common mistakes — and how to dodge them
✕
Dividing by n when you should divide by n − 1
Population variance divides by n; sample variance divides by n − 1 — and using the wrong one is the single most common statistics error. The data 2, 4, 4, 4, 5, 5, 7, 9 has population SD 2 but sample SD ≈ 2.1381: same data, different question. If the data IS the whole group, divide by n; if it is a sample standing in for a larger group, divide by n − 1. This statistics calculator shows both, labelled, on every problem, so a wrong divisor shows up as a visible choice instead of a silent error.
✕
Quoting quartiles without saying which method
Quartiles are not one calculation: Tukey’s hinges, inclusive halves, and percentile interpolation can all give different Q1 and Q3 for the same data. A quartile without a stated method cannot be reproduced or checked. This statistics calculator states Tukey’s hinges — the median of each half, the median itself excluded when the count is odd — and shows both halves in the steps. Whenever you quote a quartile anywhere, name the method with it.
✕
Judging spread by the range alone
The range uses exactly two values — the smallest and the largest — so one outlier rewrites it: 10 to 50 looks like a spread of 40, while seven of the eight values sit between 10 and 20. The IQR (6 here) and the standard deviations tell the truer story because they use the whole dataset. This statistics calculator reports the IQR and both SDs alongside the range, so the range is never the only voice.
Statistics Calculator — questions
What does a statistics calculator do?
A statistics calculator summarizes a dataset. This one measures spread and shape: the five-number summary (minimum, first quartile, median, third quartile, maximum), the range and the interquartile range, the population and sample variance and standard deviation, and any outliers past the 1.5×IQR fence. It does not answer centre-only questions (those belong to the Average Calculator), and it does not do probability or combinations — the boundary is stated plainly on the page, and refused inputs get the reason named.
What is the difference between population and sample standard deviation?
The population standard deviation divides by n — it describes the whole group in front of you. The sample standard deviation divides by n − 1 — it estimates the spread of a larger group from a sample of it, and the smaller divisor corrects the bias from using the sample mean. Use ÷ n when the data IS the whole group (every student in the class); use ÷ n−1 when the data is a sample (20 voters polled). This statistics calculator shows both, labelled, on every problem, so the choice is visible instead of silent.
How are quartiles calculated here?
By Tukey’s hinges: sort the data, find the median, then take the median of the lower half for Q1 and the median of the upper half for Q3 — with the median itself excluded from both halves when the count is odd. Other methods exist (inclusive halves, percentile interpolation) and can give different quartiles, which is why this statistics calculator states its method and shows both halves in the steps: every quartile here can be checked by hand.
What is the IQR, and how does it find outliers?
The IQR — the interquartile range, Q3 minus Q1 — is the width of the middle 50% of the data, the spread that ignores both tails. Outliers are values strictly beyond the fences: below Q1 − 1.5 × IQR or above Q3 + 1.5 × IQR. A value exactly on a fence is not an outlier. In 10, 12, 14, 15, 16, 18, 20, 50, the fences are 4 and 28, so 50 is flagged and nothing else is.
Does this statistics calculator do probability or combinations?
No — and it says so rather than faking it. This page summarizes the spread of one dataset only: it never computes a probability, never counts combinations or permutations, and never compares two datasets. Ask it for a probability and it refuses with the reason named. The boundary keeps every answer on this page honest: what it computes, it proves.
Is the statistics calculator free?
Yes — free, no sign-up, no app, full steps and verification included on every problem. The scope is one dataset of 2–50 numbers: five-number summary, range, IQR, both variances, both standard deviations, and outliers. Anything outside that scope — probabilities, combinations, centre-only questions — is refused with the reason named instead of answered approximately.
A note on honest use: VerifySolve is built for learning. Use the steps to understand the method, then close the tab and try one on your own — that is where the grade actually comes from. If your teacher or school has a policy on calculators, follow it.