Sample and population statistics.
Population: σ = √[ Σ(xᵢ − μ)² / N ] Sample: s = √[ Σ(xᵢ − x̄)² / (n − 1) ]
There are two versions of the standard deviation formula depending on whether your numbers represent a complete population or just a sample drawn from one.
Population Standard Deviation (σ) — use this when you have data for every member of the group.
Sample Standard Deviation (s) — use this when your data is a subset (sample) of a larger group. It applies Bessel's correction (dividing by n − 1 instead of n) to produce an unbiased estimate of the true population spread.
Step 1 — Mean (x̄): (4 + 7 + 13 + 2 + 1) / 5 = 27 / 5 = 5.4 Step 2 — Squared deviations: (4 − 5.4)² = (−1.4)² = 1.96 (7 − 5.4)² = ( 1.6)² = 2.56 (13 − 5.4)² = ( 7.6)² = 57.76 (2 − 5.4)² = (−3.4)² = 11.56 (1 − 5.4)² = (−4.4)² = 19.36 Step 3 — Sum of squared deviations: 1.96 + 2.56 + 57.76 + 11.56 + 19.36 = 93.20 Step 4 — Sample variance (s²): 93.20 / (5 − 1) = 93.20 / 4 = 23.30 Step 5 — Sample SD (s): √23.30 ≈ 4.83
Result: Sample Standard Deviation (s) ≈ **4.83** Mean (x̄) = 5.4 Sample Variance (s²) = 23.30
A sample standard deviation of 4.83 means the typical data value in this set deviates from the mean of 5.4 by about 4.83 units. The value 13 pulls the spread upward significantly — if removed, the SD would drop considerably. In practice, a larger SD signals more variability in your data, while a smaller SD indicates the values are tightly grouped around the mean.
Standard deviation is one of the most widely used statistics in science, finance, quality control, and research. It tells you how consistent or variable a set of numbers is — a crucial question whenever you compare groups, evaluate risk, or check whether a process is under control.
| Situation | Use | |---|---| | You have data on every member of the group (e.g., all 30 students in one class) | Population SD (σ) | | Your data is a random sample from a larger group (e.g., 30 students from all schools) | Sample SD (s) |
Dividing by n − 1 (Bessel's correction) compensates for the fact that a sample's deviations tend to underestimate the true population spread.
Variance (s² or σ²) is the average of the squared deviations. Standard deviation is simply its square root, which restores the original units of measurement (e.g., dollars, centimetres) and is therefore easier to interpret.
For data that follows a normal (bell-curve) distribution:
This rule is used constantly in quality control (Six Sigma), grading on a curve, and risk assessment in finance.
In investing, SD measures volatility. A stock with a high SD has prices that swing widely; a low-SD bond fund is more stable. Portfolio managers use SD as a core risk metric alongside expected return.
SD is sensitive to outliers. A single extreme value can inflate the SD dramatically (as seen in the worked example above). For skewed distributions, also consider the median, interquartile range (IQR), or mean absolute deviation (MAD) for a fuller picture.
It means the typical data value is 1 unit away from the mean. Whether that is large or small depends entirely on the context — a SD of 1 is very tight for IQ scores (mean 100) but significant for a chemical purity measurement that should be 99.9%.
Teachers sometimes 'curve' grades by setting the mean as a benchmark and awarding letter grades based on how many SDs a student's score falls above or below it. For example, scores within 1 SD of the mean might earn a C, those 1–2 SDs above earn a B, and so on.
The coefficient of variation (CV) = (SD / Mean) × 100%. It expresses SD as a percentage of the mean, allowing you to compare variability across data sets with different units or scales — for example, comparing price volatility of a $10 stock vs. a $500 stock.
Adding more data points does not necessarily increase or decrease SD — it depends on whether the new values are close to or far from the mean. However, a larger sample size does reduce the *standard error* of the mean, making your estimate of the population mean more precise.
In a normal distribution, SD defines the exact shape of the bell curve. The mean sets the centre, and SD controls the width — a larger SD creates a flatter, wider curve while a smaller SD creates a taller, narrower peak. This relationship underlies z-scores, confidence intervals, and hypothesis testing.
Population SD (σ) is used when your data includes every member of the group; it divides by N. Sample SD (s) is used when your data is a subset of a larger group; it divides by n−1 (Bessel's correction) to produce an unbiased estimate of the population spread.
Variance (SD²) is useful in mathematical derivations and statistical tests because it has additive properties. Standard deviation is preferred for reporting and interpretation because it is expressed in the same units as the original data, making it immediately meaningful (e.g., '$15' not '$225²').
No. Because it is the square root of a sum of squared values, standard deviation is always zero or positive. An SD of zero means every value in the data set is identical.
You need at least 2 values to calculate a sample SD (since dividing by n−1 = 0 is undefined for n = 1). For population SD you technically need only 1 value, though a SD of zero is trivial. More data points generally produce a more reliable estimate.
No. Standard deviation depends only on the values themselves, not the order in which they appear. The calculator will produce the same result regardless of how you sequence the input.
Excel's STDEV() and STDEV.S() functions compute sample SD, while STDEVP() and STDEV.P() compute population SD. Make sure you are selecting the same type (sample vs. population) in both tools to get matching results.
Only sources that have been reviewed are shown. Unverified citations are never published.
Spotted an inaccuracy or have an idea to improve this calculator? Let us know.
We use cookies to operate the site, and — with your consent — to show advertising and measure usage. You can change your choice anytime in Cookie Settings.