Six Sigma · Variation

Standard Deviation

Also known as: Sigma (σ), Standard Dev, SD

Methodology
Six Sigma Lean
Belt Level
Yellow Green Black
Type
Metric

Standard deviation measures how spread out a set of data is around its average. A small standard deviation means the values cluster tightly around the mean; a large one means they are scattered widely.

In Six Sigma, it is the fundamental unit of variation: nearly every capability metric, control limit, and sigma level is built on top of it.

The Formula

How Standard Deviation Is Calculated

The formula changes slightly depending on whether you have every value (a population) or a representative subset (a sample). Use the sample form for almost all real-world Six Sigma work.

Population

Population Standard Deviation

Use when your data covers every member of the entire group.

\sigma = \sqrt{\dfrac{1}{N}\sum_{i=1}^{N}\left(x_i - \mu\right)^2}

Sample

Sample Standard Deviation

Use when your data is a subset of a larger group. This is the most common case in Six Sigma work.

s = \sqrt{\dfrac{1}{n-1}\sum_{i=1}^{n}\left(x_i - \bar{x}\right)^2}

Formula Key

\sigma\,/\,s
Standard deviation — σ for a population, s for a sample
\mu\,/\,\bar{x}
The mean — μ for a population, x̄ for a sample (also known as X-Bar)
\Sigma
Sum — Add up the results of all the calculations for the items listed in the parentheses
N
Count — The number of data elements for which you calculated standard deviation
x
Each value — A placeholder for each data element

In plain English: find the average, measure how far each value sits from it, square those distances so they don't cancel out, average them, then take the square root to return to the original units. The result is the typical distance between any single data point and the mean.

Reading the Numbers

What Each Band of Standard Deviation Contains

For normally distributed data, the empirical rule tells you what share of values fall within a given number of standard deviations of the mean. This is the backbone of control charts and sigma levels.

±1σ
68.3% of data

About two-thirds of all values fall within one standard deviation of the mean.

±2σ
95.4% of data

Nearly all typical variation lives within two standard deviations.

±3σ
99.7% of data

Control limits sit here; points beyond ±3σ signal a special cause.

±6σ
99.99966%

Six Sigma quality: just 3.4 defects per million opportunities.

How It Differs

Standard Deviation vs. Related Metrics

Standard Deviation vs Variance

Variance is the average of the squared deviations, so it is literally standard deviation squared. Taking the square root returns standard deviation to the original units (grams, seconds, millimeters), which makes it far easier to interpret. Use variance for the math; report standard deviation to people.

Standard Deviation vs Range

Range is simply the largest value minus the smallest, so it depends on only two points and one outlier can distort it completely. Standard deviation uses every value in the data set, giving a stable picture of spread that does not swing on a single extreme reading.

Standard Deviation vs Sigma Level

Standard deviation is the raw unit of spread. A sigma level counts how many standard deviations fit between the process mean and the nearest specification limit. In short, standard deviation measures variation; sigma level measures how much of it your process can absorb before producing a defect.

Real-World Example

A coffee roaster fills 340 g bags of beans with a tolerance of ±10 g (330–350 g). Sampling 30 bags off the line, the team finds a mean fill of 341 g and a standard deviation of 4.1 g. At ±3σ that is ±12.3 g of spread, which pushes past the 10 g tolerance, so some bags come out underweight.

4.1 → 2.5 g
Standard deviation (s)
39%
Less variation
±7.5 g
New 3σ spread, inside spec

What They Changed

After recalibrating the filler hopper and stabilizing the bean flow, standard deviation dropped to 2.5 g. Now ±3σ spans only 7.5 g, comfortably inside the 10 g tolerance, and nearly every bag meets spec without the team chasing the average up or down.

Where It's Applied

Where Standard Deviation Is Used in Lean & Six Sigma

  • Measure phase of DMAIC, baselining current process variation
  • Process capability analysis (Cp, Cpk, Pp, Ppk)
  • Control charts, setting the ±3σ upper and lower control limits
  • Hypothesis testing and confidence intervals

Related Blog Posts

Six Sigma Tools & Templates

Six Sigma Tools & Templates

These Six Sigma tools and templates are intended for the benefit of students enrolled in our free promotional training program.   These tools and templates are provided “as-is” (i.e. we do not offer additional support for them). Six Sigma Templates: Communication Plan Template (PowerPoint) Control Plan Template (Excel) Countermeasures Matrix Template (Excel) Discrete Control Chart Template (PowerPoint) DMADV Project Planning […]

Read article →
Black Belt

War-Winning Weapons In A Black Belt’s Arsenal

Six Sigma professionals such as Black Belts certainly know how to make the best possible use of Six Sigma concepts and methodologies, but more often than not, the going still gets tough for them because if you look at Six Sigma implementations, you will realize that it’s often a full-blown war out there. You will […]

Read article →
Business Optimization, Six Sigma Certification & Training, Six Sigma Implementation

Is Your Organization Ready for Six Sigma Training?

Six Sigma Training can be used successfully across all industries.  The leading companies that developed and implemented Six Sigma methodology, such as GE, Motorola and AlliedSignal, have made sure their efforts bring success.  Although it isn’t easy, continual work toward improvement is key. The Six Sigma Methodology has been proven to increase savings on production […]

Read article →