2026.03.31｜MiDFUN Editorial Team
What Is Standard Deviation?
Standard deviation (abbreviated SD) is the statistical metric used to measure the degree of dispersion in a set of data. Put simply, standard deviation tells us how far each data point lies from the mean.
The symbol for standard deviation is usually the Greek letter σ (lowercase sigma) for the population standard deviation, and s for the sample standard deviation. In quality control and SPC (Statistical Process Control), σ is one of the most frequently cited statistics of all.
When the value of the standard deviation is small, it means the data is concentrated near the mean and the distribution is tightly clustered; conversely, the larger the standard deviation, the more dispersed the data and the greater the differences among the points. For manufacturing, this directly reflects how stable a process is.
How to Calculate Standard Deviation
Standard deviation is calculated in two forms — the population standard deviation and the sample standard deviation — the difference lying in the choice of denominator:
Population Standard Deviation
σ = √[ (1/N) × Σ(xi − μ)2 ]
where N = population size, xi = each observed value, μ = population mean
Sample Standard Deviation
s = √[ 1/(n−1) × Σ(xi − x̄)2 ]
where n = sample size, xi = each observed value, x̄ = sample mean
The denominator of the sample standard deviation uses n−1 (known as Bessel’s correction), the purpose being to compensate for the underestimation bias that arises when inferring the population variance from a sample. In practical quality control, because it is almost impossible to measure every product, in most cases the sample standard deviation s is what is used.
The Role of Standard Deviation in Quality Control
In quality control, the most important application of standard deviation is setting the control limits. The formulas for the upper control limit (UCL) and lower control limit (LCL) of an SPC control chart are:
UCL = x̄ + 3σ
CL = x̄ (center line, i.e. the mean)
LCL = x̄ − 3σ
This is the well-known ±3σ rule. According to the theory of the normal distribution, about 99.73% of the data will fall within the range of three standard deviations above and below the mean. When a measured value exceeds the UCL or LCL, it is judged as Out of Control (OOC), indicating that the process may have a special cause and requires immediate investigation and correction.
Beyond a single point exceeding the control limits, quality personnel also identify process anomalies through a variety of decision rules — consecutive-point trends, shifts, cyclical patterns, and more — in accordance with international standards such as AIAG-VDA. The basis of all these rules is built upon the statistical properties of the standard deviation.
Standard Deviation vs. Six Sigma
Many people easily confuse “standard deviation” with “Six Sigma.” Although the two are closely related, they are different concepts:
| Comparison Item | Standard Deviation (σ) | Six Sigma (6σ) |
|---|---|---|
| Nature | A statistic that measures the degree of data dispersion | A management methodology pursuing near-zero defects |
| Objective | To quantify the degree of variation | To reduce the defect rate below 3.4 parts per million |
| Level of Application | A foundational tool for data analysis | An organization-level quality improvement strategy |
| Relationship | Six Sigma takes “6 times the standard deviation” as its quality target, with σ as its core calculation basis | |
The name Six Sigma comes precisely from the standard deviation symbol σ: when process capability reaches the 6σ level, it means the specification limits are six standard deviations away from the mean, and the defect rate can be driven down to 3.4 parts per million (3.4 DPMO). To explore the Six Sigma methodology in depth, see the Complete Introduction to Six Sigma.
Application of Standard Deviation in SPC Control Charts
In SPC control charts, standard deviation is the core parameter that determines the sensitivity of the chart. Whether it is an X̄-R chart, an X̄-S chart, or an I-MR chart, the width of the control limits depends directly on the magnitude of σ.
The smaller the σ, the more stable the process. When the standard deviation shrinks, the control limits narrow accordingly, meaning the natural range of variation in the process becomes smaller and product consistency higher. This is the goal every quality engineer pursues: continuously shrinking the σ value by eliminating sources of special-cause variation.
In practical work, quality personnel periodically recalculate σ to update the control limits, ensuring the control chart correctly reflects the current process capability. In addition, σ is also an essential parameter for calculating the process capability indices Cpk and Ppk, directly affecting the assessment of process capability.
The MiDFUN SPC system can automatically calculate standard deviation, plot control charts, detect OOC anomalies, and raise real-time alerts based on AIAG-VDA decision rules, helping enterprises transform from manual spot checks to data-driven automated quality control. For more on the principles and practical application of SPC control charts, see SPC Modern Quality Automated Control Charts and Analysis.
Frequently Asked Questions (FAQ)
Q1: What does standard deviation mean?
Standard deviation is a statistical metric that measures the degree of dispersion in a set of data. It represents the average distance between each data point and the mean. A smaller standard deviation indicates more concentrated data, while a larger one indicates more dispersed data. In quality control, standard deviation directly reflects the degree of process stability.
Q2: How is the symbol σ pronounced?
σ is the lowercase Greek letter sigma, pronounced “sigma.” In the field of quality control, σ refers specifically to the population standard deviation. The sample standard deviation is denoted by the letter s. When “3σ” is mentioned, it is read “three sigma,” referring to three times the standard deviation.
Q3: How is standard deviation used in an SPC control chart?
The control limits of an SPC control chart are set at the mean ±3σ. When measured data exceeds this range (the UCL or LCL), it is judged as Out of Control (OOC), indicating the process may contain a special cause. In addition, standard deviation is used to calculate the process capability indices (Cpk / Ppk) to assess whether a process is capable of stably producing conforming products.
Q4: What is the relationship between standard deviation and Six Sigma?
Standard deviation (σ) is a statistic, while Six Sigma is a quality management methodology built on σ. The goal of Six Sigma is to have the specification limits of a process reach six standard deviations away from the mean, keeping the defect rate below 3.4 parts per million. In other words, σ is the calculation tool, and Six Sigma is the management strategy.
Q5: How does the MiDFUN SPC system use standard deviation for quality control?
The MiDFUN SPC Statistical Process Control system can automatically calculate the standard deviation and control limits in real time, plot control charts such as X̄-R, X̄-S, and I-MR, and automatically detect OOC anomalies based on the 7 AIAG-VDA decision rules. The system supports process capability analysis (Cpk / Ppk), helping enterprises monitor process stability in real time and transform from passive inspection to a proactive, preventive quality management model.
Want standard deviation to safeguard your process?
Discover how MiDFUN uses its SPC system to automatically monitor σ and detect anomalies in real time.
Further reading: Six Sigma｜SPC Control Charts and Analysis｜Cpk / Ppk Process Capability Indices｜MSA Measurement System Analysis
Copyright © 2026 MiDFUN Co., Ltd. Some rights reserved
Author: Pei-Chi Chiu. First published: 2026-03-31. Type: Quality Management Column
Original link: https://www.midfun.com.tw/qc/glossary-sigma-standard-deviation/
This work is released under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). You are welcome to share it freely, provided that you credit the original author, include the original link, do not use it commercially, and do not modify the content.
Suggested citation format: Chiu, P.-C. (2026). “What Is Standard Deviation? σ Definition, Formulas, and SPC Applications.” MiDFUN Quality Management Column.
For reprint permission and content inquiries: midfun@midfun.com.tw
