What is a standard deviation calculator?
A standard deviation calculator measures how spread out a set of numbers is around their average. When you report only a mean, you hide whether every value sits close together or whether a few outliers stretch the range. Standard deviation and variance answer that dispersion question with two related statistics: variance squares the distances from the mean, and standard deviation takes the square root so the result stays in the same units as your original data.
The standard deviation calculator on Yaya Tools accepts a pasted list of numbers—separated by commas, spaces, line breaks, or Chinese commas—and returns mean, count, sum, population variance, population standard deviation, sample variance, sample standard deviation, and the coefficient of variation (CV). Results update as you type. There is no sign-up, no file upload, and no server round trip: every calculation runs locally in your browser.
Whether you are checking homework, summarizing lab readings, reviewing sales by region, or comparing two datasets before a meeting, a dedicated variance calculator saves time because you do not have to rebuild spreadsheet formulas each time the list changes. The page shows both population and sample formulas side by side, which makes sample vs population standard deviation choices visible instead of buried in a cell comment.
How to use this standard deviation calculator
The workflow is built for quick paste-and-read analysis.
- Open the standard deviation calculator in your browser.
- Paste or type your numbers into the input box. Separate values with commas, spaces, line breaks, or Chinese commas (，).
- Read the summary cards: mean, count, sum, population SD and variance, sample SD and variance, and CV.
- Use Reset example to restore the built-in sample dataset, or Clear to wipe the field and start over.
The tool validates input continuously. Empty fields, non-numeric tokens, or values that produce invalid results trigger a status message so you can fix the list before trusting the output. When only one number is present, population statistics still compute, but sample variance and sample standard deviation show "—" because they require at least two values. When the mean is exactly zero, CV also shows "—" because dividing by zero is undefined.
For a quick sanity check on the mean before you interpret spread, compare your result with the average calculator. If you need the raw total of the list, the number sum tool adds the same values without dispersion metrics.
Standard deviation and variance formulas
The calculator implements textbook definitions. Below, n is the count of numbers, xᵢ is each value, μ is the population mean (here, the mean of your entered list), and x̄ is the sample mean (numerically the same as μ for one dataset, but the formulas differ in the denominator).
Mean, count, and sum
Mean (μ or x̄) = sum of all values ÷ n
Sum = x₁ + x₂ + … + xₙ
Count = n
The mean is the balance point of the dataset. Sum and count appear on the results panel so you can verify inputs or reuse totals elsewhere—for example, confirming a column total in the number sum tool before analyzing spread here.
Population variance and population standard deviation
Population formulas treat your entered list as the entire group of interest:
Population variance σ² = Σ(xᵢ − μ)² ÷ n
Population SD σ = √σ²
Each term (xᵢ − μ)² is the squared deviation from the mean. Squaring removes negative signs so values far from the mean contribute more than values close to it. Dividing by n yields the average squared deviation. Taking the square root returns to the original measurement unit, which is why standard deviation is easier to interpret than variance when values represent dollars, seconds, or test points.
Sample variance and sample standard deviation
Sample formulas estimate dispersion for a subset drawn from a larger population. They use n − 1 in the denominator (Bessel's correction):
Sample variance s² = Σ(xᵢ − x̄)² ÷ (n − 1)
Sample SD s = √s² (requires n ≥ 2)
When n is 1, the sample standard deviation is undefined; the calculator displays "—" for sample variance and sample SD in that case. Population standard deviation still computes because its divisor remains 1.
Coefficient of variation (CV)
CV = (population SD ÷ |mean|) × 100%
CV expresses relative variability as a percentage of the mean's magnitude. It helps compare spread across datasets measured in different units or with very different averages. If the mean is zero, CV is not defined and the tool shows "—". For percentage-style comparisons of two proportions or rates, the percentage calculator handles ratio questions; CV here is specifically a dispersion-to-mean ratio.
Sample vs population standard deviation
Choosing the wrong divisor is one of the most common errors in introductory statistics. The calculator outputs both so you can pick the row that matches your question.
| Question you are answering | Use | Divisor |
|---|---|---|
| You have every member of the group (full roster, complete census, all parts measured) | Population SD / σ | n |
| You have a subset meant to represent a larger group (survey sample, experimental trial, quality spot-check) | Sample SD / s | n − 1 |
Population standard deviation describes the list you actually entered as the whole population. Example: every test score in a class of 28 students when you have all 28 scores and care only about that class.
Sample standard deviation adjusts for the fact that a sample mean x̄ usually sits closer to the sample values than the true population mean would. Dividing by n − 1 instead of n makes the sample variance an unbiased estimator of population variance under standard assumptions. Example: 50 customer ratings chosen from thousands of transactions, where you want to infer variability in the full customer base.
For large n, population and sample standard deviation converge. For small samples, the gap matters. With the example dataset 2, 4, 4, 4, 5, 5, 7, 9 (n = 8), the mean is 5, population SD is 2, and sample SD is approximately 2.138. Reporting 2 when your instructor expects the sample formula—or the reverse—will cost points even when the dataset is identical.
If you are unsure which label your assignment or report requires, check whether the prompt treats the data as a complete census or a sample from something larger. When in doubt, sample SD is the safer default for real-world measurements drawn from a broader process.
Worked examples
These examples use numbers you can paste directly into the tool.
Example 1: Built-in sample dataset
The default list is:
2, 4, 4, 4, 5, 5, 7, 9
- Count: 8
- Sum: 40
- Mean: 5
- Population variance: 4 → population SD: 2
- Sample variance: 4.571… → sample SD: ≈ 2.138
- CV: (2 ÷ 5) × 100% = 40%
The mean matches what you get from the average calculator on the same eight values. Most values cluster near 4–5; 2 and 9 pull the standard deviation up to 2 (population).
Example 2: Identical values (zero spread)
Input:
10, 10, 10, 10
Every value equals the mean, so all squared deviations are zero. Population variance, sample variance, population SD, and sample SD are all 0. CV is 0% because variability is nil relative to the mean. This is a useful edge-case check: if your real-world sensor readings look like this but you expected noise, inspect the data source before continuing analysis.
Example 3: Single value
Input:
42
Mean, sum, and count display normally. Population variance and population SD are 0 because one point has no spread around itself. Sample variance and sample SD show "—" with a note that at least two numbers are required. This matches the behavior described in the tool FAQ.
Example 4: Negative and decimal values
Input:
-1.5, 0, 2.25, 3.75
The calculator accepts negatives and decimals. Mean = (-1.5 + 0 + 2.25 + 3.75) / 4 = 1.125. Squared deviations are computed from that mean; population and sample SD return in the same units as the inputs. CV uses |mean| in the denominator, so a negative mean still produces a positive CV percentage.
Example 5: Comparing two lists with CV
Suppose team A daily sales average $1,000 with population SD $50, and team B averages $200 with population SD $40. Raw SD suggests A is more volatile, but CV tells a different story:
- Team A CV ≈ 5%
- Team B CV ≈ 20%
Relative to typical sales level, B fluctuates more. CV is built from population SD on this page; use it when means are non-zero and you want a unitless comparison.
When to use variance vs standard deviation
Variance (σ² or s²) is the average squared deviation. It appears in analysis of variance (ANOVA), many statistical tests, and theoretical derivations because squared terms combine cleanly in algebra.
Standard deviation (σ or s) shares the unit of the original measurements. If heights are in centimeters, SD is in centimeters; if reaction times are in milliseconds, SD is in milliseconds. That makes SD the better choice for reports, charts, and verbal summaries ("scores varied by about 12 points").
This calculator returns both so you can copy the statistic your audience expects. In spreadsheets, SD is often the column you plot as error bars; variance is the intermediate step you rarely show to non-specialists.
For related arithmetic on the same list—mean without spread, or sum without averaging—use the average calculator and number sum respectively. For expressing one dispersion value as a share of another quantity, the percentage calculator complements CV when your question is about ratios rather than spread around a mean.
Common mistakes when calculating standard deviation
Using the sample formula on a full census (or the reverse). If the data are the entire population of interest, population SD with divisor n is appropriate. If the data are a sample, sample SD with divisor n − 1 is standard. The tool prints both; pick intentionally.
Forgetting that sample SD needs at least two points. With one observation, spread around the mean is always zero for population formulas, but sample SD is undefined. Do not treat "—" as a bug—it reflects the mathematics.
Confusing standard deviation with standard error. Standard deviation describes variability in the data. Standard error describes variability in an estimate of the mean (typically SD / √n). This page does not compute standard error.
Mixing up variance and standard deviation when reporting. Saying "the variance is 4 minutes" when you mean SD = 4 minutes misstates spread by a factor related to the square. Always label σ² vs σ (or s² vs s) explicitly.
Entering text or mixed separators incorrectly. The parser splits on whitespace, commas, and Chinese commas. Tokens must be valid numbers. Labels like "N/A" or currency symbols attached without spaces can invalidate the list.
Interpreting CV when the mean is near zero. CV divides by |mean|. Very small means inflate CV; a mean of exactly zero makes CV undefined. For data centered around zero, prefer raw SD or another scale-free method suited to your field.
Rounding intermediate steps manually. The tool keeps full precision internally and formats for display. If you hand-compute with rounded means, your variance can drift. Re-enter the same list to audit spreadsheet work.
References
- Standard deviation (NIST/SEMATECH e-Handbook of Statistical Methods) — Definitions of variance and standard deviation for process and measurement data.
- Sample variance and Bessel's correction (OpenStax Statistics) — Why sample variance uses n − 1 when estimating population spread.