How do I calculate standard deviation?

To calculate standard deviation: (1) Find the mean of your data set. (2) Subtract the mean from each value and square the result. (3) Sum all squared differences to get the sum of squares. (4) Divide by N for population or N-1 for sample to get variance. (5) Take the square root of the variance.

What does a high standard deviation mean?

A high standard deviation means data points are spread far from the mean, indicating high variability. A low standard deviation means data clusters tightly around the mean, indicating consistency. In investing, high SD means higher risk. In manufacturing, high SD means inconsistent product quality.

What is the formula for standard deviation?

Population SD: sigma = sqrt(sum of (xi - mu)^2 divided by N). Sample SD: s = sqrt(sum of (xi - x-bar)^2 divided by (N-1)). The only difference is the denominator: N for population, N-1 for sample.

What is variance vs standard deviation?

Variance is the average of squared differences from the mean, expressed in squared units. Standard deviation is the square root of variance, expressed in the original units. SD is more interpretable for reporting; variance is preferred in inferential statistical formulas.

How do I calculate a z-score?

z = (x minus mean) divided by standard deviation. It tells you how many standard deviations a value is from the mean. A z-score of +2 means 2 SD above average. Values with absolute z-score greater than 3 are considered potential outliers.

Why does sample standard deviation divide by N-1?

Dividing by N on a sample consistently underestimates the true population standard deviation. Using N-1 (Bessel's correction) corrects this bias. For large samples the difference between N and N-1 becomes negligible.

How is standard deviation used in finance?

In finance, standard deviation measures investment volatility. Higher SD means greater risk and potential reward. The Sharpe ratio divides excess return by SD for risk-adjusted performance comparison. Options pricing models like Black-Scholes use SD as implied volatility.

What is a good standard deviation?

There is no universally good SD. Use CV = (SD/mean times 100%) to assess relative variability. A CV below 15% is generally low variability. In academic testing, SD of 10 to 15 on a 100-point exam is typical. In manufacturing, smaller SD means better precision.

Standard Deviation Calculator

📋 Enter your data

Data Set Minimum 2 values required. Please enter at least 2 valid numbers.

Standard Deviation

—

⚠️ Disclaimer: This calculator is for educational and general reference purposes. Results are based on standard statistical formulas. For clinical research, financial decisions, or academic submissions, verify with qualified statistical software or a professional statistician.

Sources & Methodology

✓Formulas verified against NIST/SEMATECH e-Handbook of Statistical Methods. Bessel's correction sourced from Kenney & Keeping (1951). Z-score methodology per Khan Academy Statistics curriculum.

📘

NIST/SEMATECH e-Handbook of Statistical Methods — Measures of Scale

Authoritative US government reference for standard deviation and variance formulas, including population and sample variants, used as primary formula source for this calculator.

🏫

Khan Academy — Calculating Standard Deviation Step by Step

Widely used educational reference for the step-by-step standard deviation procedure and the conceptual basis for Bessel's correction in sample SD.

📊

Scribbr — Standard Deviation & Variance Guide

Peer-reviewed statistics reference covering sample vs population distinction, coefficient of variation, standard error, and interpretation guidelines used for content accuracy.

Methodology: Population SD: σ = √[ ∑(xᵢ − μ)² / N ] Sample SD: s = √[ ∑(xᵢ − x̅)² / (N−1) ] SEM = s / √n | CV = (s / x̅) × 100% | z = (x − μ) / σ Sample SD uses Bessel's correction (N−1 denominator). All results rounded to 4 decimal places. Welford one-pass algorithm used for numerical stability on large data sets.

Last reviewed: April 2026

How to Calculate Standard Deviation — Complete Guide

Standard deviation (SD) is the most widely used measure of statistical dispersion. It tells you, on average, how far each data point lies from the mean. A low standard deviation means your data clusters tightly around the average — values are consistent and predictable. A high standard deviation means your data is spread across a wide range — values are highly variable.

Understanding standard deviation is fundamental to data analysis in finance, research, quality control, medicine, education, and virtually any field involving numbers. Once you understand what SD tells you, you can interpret datasets, compare distributions, identify outliers, and make evidence-based decisions.

Standard Deviation Formula

There are two formulas depending on whether your data represents a complete population or a sample drawn from a larger one.

Population SD: σ = √[ ∑(xᵢ − μ)² / N ] Sample SD: s = √[ ∑(xᵢ − x̅)² / (N−1) ]

Where xᵢ = each data value • μ or x̅ = mean • N = count of values
The only difference: population divides by N, sample divides by N−1 (Bessel's correction).

Step-by-Step Example

Using the data set: 4, 8, 15, 16, 23, 42 (sample)

Step 1: Count: n = 6
Step 2: Sum = 108 → Mean = 108 ÷ 6 = 18.0000
Step 3: Squared deviations: (4−18)²=196 | (8−18)²=100 | (15−18)²=9 | (16−18)²=4 | (23−18)²=25 | (42−18)²=576
Step 4: Sum of squares = 910
Step 5: Sample variance = 910 ÷ (6−1) = 182.0000
Step 6: Sample SD = √182 = 13.4907

Population vs Sample Standard Deviation

💡 Rule of thumb: Use Population SD (σ) when your data includes every single member of the group (e.g., all 30 students in one class). Use Sample SD (s) when your data is a subset selected from a larger group (e.g., 500 survey respondents out of a city of 500,000). When in doubt — use sample SD.

Formula	When to Use	Denominator	Symbol	Example
Population SD	Complete data set	N	σ (sigma)	All 30 students in one class
Sample SD	Subset / sample	N−1	s	500 of 50,000 surveyed customers
Variance (pop.)	Complete data set	N	σ²	ANOVA, regression models
Variance (sample)	Subset / sample	N−1	s²	Inferential statistics tests

Variance vs Standard Deviation

Variance (s² or σ²) is the average of the squared differences from the mean. Because the differences are squared, variance is expressed in squared units which are mathematically useful but hard to interpret. Standard deviation solves this by taking the square root, returning the result to the original unit. This is why SD is reported in descriptive statistics while variance is used in ANOVA, regression, and hypothesis testing.

Standard Error of the Mean (SEM)

SEM measures how accurately a sample mean estimates the true population mean:

SEM = s / √n

A smaller SEM indicates more precise estimation of the population mean. SEM decreases as sample size grows. A 95% confidence interval is approximately x̅ ± 1.96 × SEM.

Coefficient of Variation (CV)

CV = (s / x̅) × 100%

CV allows fair comparison of variability between data sets with different units or scales. A CV below 15% is generally low variability; above 30% is high. Use CV when comparing two data sets measured in different units.

Z-Score (Standard Score)

z = (x − μ) / σ

A z-score of +2 means 2 SD above average. A z-score of −1.5 means 1.5 SD below. Values with |z| > 3 are typically considered statistical outliers (bottom/top 0.3% of a normal distribution).

The 68–95–99.7 Rule (Empirical Rule)

Range	Data Covered	Z-Score Range	Example (mean=100, SD=15)
μ ± 1σ	~68%	−1 to +1	85 to 115
μ ± 2σ	~95%	−2 to +2	70 to 130
μ ± 3σ	~99.7%	−3 to +3	55 to 145
Beyond ± 3σ	~0.3%	\|z\| > 3	Below 55 or above 145

Real-World Applications

Finance & investing: SD measures stock or portfolio volatility. Higher SD = greater risk and potential reward. The Sharpe ratio divides excess return by SD for risk-adjusted comparison.
Quality control (Six Sigma): Manufacturers target processes where defects fall beyond 6σ from the mean — near-zero defect rates.
Medicine & clinical trials: SD determines whether treatment differences are statistically significant vs. natural patient variation.
Education: Standardized tests use SD to check score distributions. Grading on a curve applies the empirical rule to assign grades.
A/B testing & data science: SEM and SD are core inputs for confidence intervals, statistical significance, and required sample size calculations.
Weather & climate: SD of temperature data reveals climate consistency. A coastal city and an inland city can have the same mean temperature but very different SDs.

Common Mistakes When Using Standard Deviation

Using population SD on sample data: Always use N−1 for samples. Using N produces a biased, systematically low estimate.
Confusing SD with SEM: Standard error is much smaller than SD and answers a different question — precision of the mean estimate, not data spread.
Applying the 68–95–99.7 rule to non-normal data: This rule only applies to normally distributed data. Skewed or bimodal distributions require different interpretation methods.
Ignoring units: Standard deviation is in the same units as your data. Variance is in squared units. Always check which statistic is being reported.

Frequently Asked Questions

Find the mean of your data, subtract the mean from each value and square the result, sum all the squared differences, divide by N (population) or N−1 (sample) to get variance, then take the square root. Use the calculator above to get instant results with full step-by-step working.

Population SD divides by N and is used when your data covers every member of the group. Sample SD divides by N−1 (Bessel's correction) and is used when your data is a subset of a larger population. Sample SD gives a slightly larger result to correct for estimation bias. In most real-world analyses, sample SD is the correct choice.

A low SD means values cluster tightly around the mean — consistent and predictable data. A high SD means values are widely spread — variable data. Context matters: high SD in stock returns means higher investment risk; low SD in manufacturing dimensions means high precision. Use CV = (SD/mean × 100%) to compare variability across different scales.

Variance is the average of squared differences from the mean, in squared units. Standard deviation is the square root of variance, in the original units. SD is more intuitive for reporting; variance has better mathematical properties and is used in ANOVA and regression.

Paste your numbers into the data field above — commas, spaces, or new lines all work. Select Sample or Population, then click Calculate. The result shows SD, variance, mean, count, sum, SEM, and CV instantly with full step-by-step working and a 95% confidence interval estimate.

SEM = standard deviation divided by the square root of sample size. It measures how accurately your sample mean estimates the true population mean. A smaller SEM means a more reliable estimate. A 95% confidence interval is approximately mean ± 1.96 × SEM.

CV = (SD / mean) × 100%. It expresses variability as a percentage of the mean, allowing fair comparison between data sets with different units. A CV below 15% is typically low variability; above 30% is high. CV is especially useful when comparing consistency across different measurement systems.

In finance, SD measures investment volatility — how much returns deviate from the average. Higher SD = greater risk and potential reward. The Sharpe ratio divides excess return by SD to measure risk-adjusted performance. Options pricing models like Black-Scholes use historical SD as implied volatility directly.

For normally distributed data, approximately 68% of values fall within 1 SD of the mean, 95% within 2 SD, and 99.7% within 3 SD. This rule is used in quality control (Six Sigma targets ±6σ), grading on a curve, clinical reference ranges, and identifying outliers. Values beyond 3 SD are extremely unusual.

Dividing by N on a sample systematically underestimates the true population SD. Using N−1 (Bessel's correction) corrects this. The lost degree of freedom accounts for the fact that the sample mean was estimated from the same data. For large samples, the difference between N and N−1 becomes negligible.

z = (x − mean) ÷ standard deviation. It tells you how many standard deviations a value is from the mean. Use the Z-Score tab above. A z-score of +2 means 2 SD above average; −1.5 means 1.5 SD below. Values with |z| > 3 are considered potential outliers.

There is no universally good SD — it depends on context and scale. Use CV = (SD/mean × 100%) to assess relative variability: below 15% is low, above 30% is high. For a 100-point test, SD of 10–15 is typical. In quality control, smaller SD = better precision. In A/B testing, SD determines required sample size for statistical significance.

Related Calculators