FREE Six Sigma Green Belt Study Guide 2026

This free Six Sigma Green Belt study guide walks through everything the ASQ Certified Six Sigma Green Belt (CSSGB) exam tests, organized to ASQ’s official Body of Knowledge.^[2]

It’s interactive, not a wall of text: every module has built-in checkpoint quizzes, flashcards, and practice questions, so you learn by doing — not just reading.

The exam covers six sections — an Overview of Six Sigma and the organization, plus the five DMAIC phases (Define, Measure, Analyze, Improve, Control).^[3] We teach one module per section, starting with the fundamentals — sigma level, DPMO, and lean — that the rest of the methodology builds on.

Read a module, test yourself at each checkpoint, then drill gaps with our free practice test and flashcards. This is a high-yield overview mapped to the official outline — not a replacement for the full ASQ handbook.

Studying for the more advanced credential instead? See our Six Sigma Black Belt practice test, which covers full design of experiments, advanced and multivariate statistics, and deployment leadership — the depth beyond Green Belt.

Six Sigma Green Belt Exam Snapshot

Study by weight. Define and Measure are the two largest sections at 20% each, and the four DMAIC analysis-and-action phases together are about 74% of the exam, so that is where most of your time goes — but the Overview section ties the fundamentals together and shows up throughout:

Module 1 · Six Sigma & the Organization (11%)

Before the DMAIC phases, you need the mental model the whole methodology rests on: what “Six Sigma” means statistically, how performance is measured, and how lean fits in. About 11% of the exam (≈11 questions) is this overview — fundamentals, lean principles, and the basics of Design for Six Sigma.^[2]

1.1 Six Sigma, sigma level & DPMO

Six Sigma is a disciplined, data-driven approach to reducing process variation and defects. Its name is literal: a “Six Sigma” process is so capable that six standard deviations fit between the process mean and the nearest specification limit, leaving almost no room for defects. Performance is reported on a universal scale — the sigma level — so a billing process and a machining process can be compared on the same yardstick.^[6]

Defects are counted as defects per million opportunities. Where an opportunity is any chance for a defect to occur, DPMO normalizes by complexity so processes of different sizes compare fairly:

• Defects per unit: $DPU = \dfrac{D}{U}$ — total defects over total units.
• Defects per million opportunities: $DPMO = \dfrac{D}{U \times O} \times 1{,}000{,}000$ where $O$ is opportunities per unit.
• Process yield: $Y = 1 - \dfrac{\text{defects}}{\text{opportunities}}$.

The headline figure — a Six Sigma process produces just 3.4 DPMO — includes a 1.5σ shift for long-term drift: a process that is 6σ in the short term is treated as 4.5σ over the long term. Memorize the conversion table; it appears in many forms:

1.2 Lean principles & DFSS

Lean is Six Sigma’s partner: where Six Sigma attacks variation, lean attacks waste — any activity that consumes resources without adding value the customer would pay for. A Green Belt is expected to know the core lean ideas and tools. The classic frame is the eight wastes, remembered by the acronym DOWNTIME:

Two lean ideas underpin most improvements. The theory of constraints says throughput is limited by a single bottleneck, so improve that constraint first. And a value stream map distinguishes value-added steps (the customer would pay for them) from non-value-added steps (candidates for elimination).

Finally, know the difference between the two Six Sigma roadmaps. DMAIC improves an existing process; DMADV, or Design for Six Sigma, is used to design a new process or product to meet Six Sigma quality from the start. A basic FMEA (failure mode and effects analysis) is introduced here as a design-risk tool.^[2]

Module 2 · Define (20%)

One of the two largest sections — 20% of the exam, about 20 questions. Define is where a project is framed: what problem you’re solving, for which customer, and within what boundaries. Get Define wrong and the whole project drifts, so the exam tests it heavily.

2.1 Project charter, VOC & CTQ

Every project starts with a project charter: a one-page contract capturing the business case, a fact-based problem statement (what is wrong, where, when, how big — no causes, no solutions), a SMART goal statement, the scope, the team, and the timeline. The charter aligns the team and the sponsor and authorizes the work.^[2]

You can’t improve quality you haven’t defined, so Define captures the Voice of the Customer (VOC) — what customers actually need, gathered through surveys, interviews, complaints, and observation — and translates it into measurable Critical-to-Quality requirements. A CTQ turns vague language (“it should be fast”) into a specification (“delivered within 24 hours”). The Kano model then helps prioritize which requirements matter most:

2.2 SIPOC, project management & teams

To scope the process at a high level, Green Belts build a SIPOC diagram — Suppliers, Inputs, Process, Outputs, Customers. Read left to right, it shows who supplies the inputs, the five-to-seven high-level process steps, what the process produces, and who receives it. SIPOC keeps everyone agreed on where the process starts and stops.^[6]

Define also draws on project-management basics: a Gantt chart for the schedule, an activity network (CPM/PERT) to find the critical path, project metrics, risk analysis, and a formal project closure. The seven management and planning tools — affinity diagram, interrelationship digraph, tree diagram, prioritization matrix, matrix diagram, PDPC, and activity network — organize ideas and decisions.

Because projects run through teams, the exam covers team dynamics: the stages a team passes through (forming, storming, norming, performing, adjourning), clear roles (sponsor/champion, team leader, members, process owner), and idea tools like brainstorming, nominal group technique, and multivoting.

Module 3 · Measure (20%)

The other largest section — 20% of the exam, about 20 questions — and the most quantitative. Measure establishes a fact-based baseline: you map the process, decide what data to collect, make sure the measurement system is trustworthy, and quantify how capable the process is today.

3.1 Data, statistics & distributions

First, classify your data. Continuous (variable) data is measured on a scale — time, weight, length — and carries the most information; discrete (attribute) data is counted in categories such as pass/fail or defect counts. The data type drives which charts and tests you can use, so getting it right matters.

Summarize data with measures of central tendency — the mean, median, and mode — and measures of dispersion — the range, variance, and standard deviation. The standard deviation is the heart of Six Sigma:

• Sample standard deviation: $s = \sqrt{\dfrac{\sum (x_i - \bar{x})^2}{n-1}}$.
• A standardized value (z-score): $z = \dfrac{x - \mu}{\sigma}$ — how many standard deviations a point sits from the mean.

Most Six Sigma analysis assumes the normal distribution, the symmetric bell curve. The empirical rule says about 68%, 95%, and 99.7% of normal data fall within ±1, ±2, and ±3 standard deviations of the mean.

The central limit theorem is why this matters so broadly: averages of samples tend toward normality even when the raw data isn’t normal. You should also recognize the binomial (counts of pass/fail) and Poisson (counts of rare events) distributions.^[2]

3.2 Measurement systems & process capability

Before trusting any data, check the measurement system itself with a Measurement System Analysis (MSA). A Gage R&R study separates two sources of measurement error: Repeatability (one operator, same part, repeated) and Reproducibility (different operators, same part).

If too much variation comes from the gage, the numbers can’t support good decisions — fix the measurement system first. A common rule of thumb: under 10% measurement variation is acceptable, over 30% is not.^[6]

With trustworthy data, quantify process capability— how well the process fits inside the customer’s specification limits. The two Green-Belt indices are Cp and Cpk:

• Potential capability: $C_p = \dfrac{USL - LSL}{6\sigma}$ — spec width over the process spread, assuming the process is centered.
• Actual capability: $C_{pk} = \min\left[\dfrac{USL - \mu}{3\sigma},\ \dfrac{\mu - LSL}{3\sigma}\right]$ — also penalizes an off-center mean.

Because Cpk accounts for centering, $C_{pk} \le C_p$ always, and they are equal only when the process is perfectly centered. The long-term versions, Pp and Ppk, use the overall standard deviation instead of the within-subgroup one. A capability of 1.33 (about 4σ) is a common minimum target.^[4]

Finally, measure yield. First Time Yield (FTY) is the fraction of units that pass a step the first time; Rolled Throughput Yield (RTY) multiplies every step’s first-time yield to give the probability a unit passes the whole process defect-free —$RTY = \prod_{i=1}^{n} Y_i$. RTY exposes the “hidden factory” of rework that a final-inspection yield hides.

Module 4 · Analyze (18%)

About 18% of the exam, roughly 18 questions. Analyze is where you find and verify the root cause(s) of the problem — not the symptoms, and not by opinion. It blends graphical root-cause tools with the first real inferential statistics.

4.1 Root cause analysis

Root cause analysis drills past symptoms to the underlying cause, so a fix actually prevents recurrence. Three tools do most of the work. The Fishbone (Ishikawa) diagram organizes candidate causes into categories around a spine — the 6 Ms: Methods, Machines, Materials, Measurement, Manpower (people), and Mother Nature (environment).^[6]

The 5 Whys drills down by asking “why?” of each answer about five times until you reach an actionable root cause. And the Pareto chart applies the 80/20 rule — that most of a problem comes from a few causes — by ordering causes from most to least frequent so you attack the vital few first. These tools generate and prioritize causes; you still confirm with data.

4.2 Hypothesis testing, correlation & FMEA

Hypothesis testing is how you confirm a cause statistically. You state a null hypothesis (no effect) and an alternative (there is an effect), then compute a p-value: the probability of seeing data this extreme if the null were true.

If $p \le \alpha$ (the significance level, often 0.05), you reject the null. Guard against two errors: a Type I error (rejecting a true null — false positive, probability α) and a Type II error (failing to reject a false null — false negative, probability β).^[2]

Green Belts use a handful of tests — a t-test to compare means, single-factor ANOVA to compare three or more means, and a chi-square test for categorical data. To study relationships between variables, correlation measures how strongly two variables move together (a coefficient from −1 to +1), and simple linear regression models one as a function of the other. Remember: correlation is not causation.

Finally, FMEA prioritizes risks before they bite. Each failure mode is rated 1–10 for Severity, Occurrence, and Detection; their product is the Risk Priority Number —$RPN = S \times O \times D$ — and the highest RPNs get addressed first.

Module 5 · Improve (16%)

About 16% of the exam, roughly 16 questions. Improve is where you generate, test, and implement solutions that remove the verified root cause. It pairs experimentation (DOE) with lean improvement tools.

5.1 Design of experiments (DOE) basics

Design of Experiments (DOE) is a structured way to learn how input factors affect an output response by changing the factors deliberately. Unlike changing one factor at a time (which is inefficient and misses interactions), a designed experiment varies factors together. Key vocabulary: a factor is an input you change, a level is a setting of that factor, the response is the measured output, a main effect is a factor’s average effect, and an interaction is when one factor’s effect depends on another’s level.^[2]

At Green-Belt level you mainly interpret simple experiments — reading main-effects and interaction plots from a full or fractional factorial design. Designing complex experiments and response-surface methods is Black-Belt work, which our Six Sigma Black Belt practice test covers.

5.2 Lean improvement tools

Many Improve-phase gains come from lean. Kaizen is continuous improvement through small, team-driven changes, often run as a focused, week-long kaizen event. 5S organizes the workplace — Sort, Set in order, Shine, Standardize, Sustain — so waste from searching and motion disappears and abnormalities are obvious.

Poka-yoke (mistake-proofing) designs the process so errors are prevented or caught instantly — a connector that only fits one way, a sensor that stops a line when a part is missing — removing reliance on human vigilance. Kanban and pull systems trigger work only on real demand, cutting overproduction and inventory.

Module 6 · Control (15%)

About 15% of the exam, roughly 15 questions. Control locks in the gains so the process doesn’t drift back to its old performance. Its centerpiece is statistical process control, backed by a control plan and standard work.

6.1 Statistical process control & charts

Statistical Process Control (SPC) monitors a process over time with a control chart: a time-ordered plot with a center line (the average) and control limits set at ±3σ. Its job is to tell two kinds of variation apart. Common-cause variation is inherent, random “noise” in a stable process; Special-cause variation is an assignable disturbance that makes the process unstable.^[5]

The cardinal rule: react only to special causes. Adjusting a stable process in response to common-cause variation is tampering, and it actually increases variation. Note too that control limits come from the process data (the voice of the process), while specification limits come from the customer — they are not the same thing.

Choose the chart by data type. Variables charts (X-bar & R, X-bar & S, and I-MR) plot continuous data; attribute charts (p, np, c, and u) plot counts. Picking the right chart is one of the highest-yield Control-phase skills on the exam:

6.2 Control plans & sustaining gains

A chart alone doesn’t sustain a gain — a control plan does. The control plan documents, for each critical process input and output, what is monitored, how (the measurement method), the sample size and frequency, the control limits, and the reaction plan to follow when the process goes out of control.^[2]

Gains stick only with the surrounding system: Standard work (the documented best method), training so operators follow it, updated SOPs, ongoing monitoring, and total productive maintenance to keep equipment from causing defects. At project close, the process and its documentation are handed off to the process owner, who is accountable for keeping the improvement in place — and the team records lessons learned for future projects.

How to Use This Six Sigma Green Belt Study Guide

This guide is built to be worked, not just read. The most efficient path to a pass:

• Lock in the fundamentals first. The Overview module is short but high-leverage — sigma level, DPMO, and lean show up across every later phase.
• Study by weight. Define and Measure are 20% each and Measure is the most quantitative — give the formulas (DPMO, Cp/Cpk, RTY) real practice.
• Check off as you go. Use the Study Guide Contents to mark each section done; it raises your exam-readiness score.
• Take every checkpoint. The end-of-module quizzes show you exactly which phases need another pass.
• Drill the weak phase. Send your weak area into the flashcards and a practice test until the score climbs.
• Build your open-book references. Because the exam is open-book, assemble and tab a bound reference (formulas, tables, chart-selection guide) as you study.

Six Sigma Green Belt Glossary

The high-yield Six Sigma Green Belt terms in one place — hover any dotted term in the guide, or flip the whole deck here as a self-grading flashcard set.

1.5 sigma shift

An allowance for long-term process drift; a 6σ short-term process is treated as 4.5σ long-term, giving 3.4 DPMO.

5 Whys

Asking 'why?' repeatedly (about five times) to drill from a symptom down to the root cause.

5S

Sort, Set in order, Shine, Standardize, Sustain — a lean method for an organized, efficient workplace.

8 wastes (DOWNTIME)

Defects, Overproduction, Waiting, Non-utilized talent, Transportation, Inventory, Motion, Excess processing.

Central limit theorem

The distribution of sample means approaches normal as sample size grows, whatever the population shape.

Common-cause variation

Inherent, random variation always present in a stable process; do not react to single points.

Continuous data

Variable data measured on a continuous scale (time, weight, length), carrying more information per point.

Control chart

A time-ordered plot with a center line and ±3σ control limits used to detect special-cause variation.

Control limits

Limits at ±3σ derived from the process data (the voice of the process), not from customer specs.

Control plan

A document specifying how each key process input and output is monitored, with a reaction plan.

Correlation

A statistical relationship in which two variables move together; it does not by itself prove causation.

Cp

Potential capability = (USL − LSL) ÷ 6σ; compares spec width to process spread, ignoring centering.

Cpk

Actual capability = min[(USL − mean), (mean − LSL)] ÷ 3σ; accounts for both spread and centering. Cpk ≤ Cp.

Critical to Quality (CTQ)

A measurable characteristic whose performance standard must be met to satisfy the customer.

Defect

Any output that fails to meet a customer requirement or specification.

Defective

A unit that contains one or more defects.

Design of Experiments (DOE)

A structured method of deliberately changing input factors to learn their effect on the output.

Discrete data

Attribute data counted in categories (pass/fail, number of defects).

DMADV (DFSS)

Define, Measure, Analyze, Design, Verify — Design for Six Sigma, used to create a new process or product.

DMAIC

The five-phase improvement roadmap for an existing process: Define, Measure, Analyze, Improve, Control.

DPMO

Defects Per Million Opportunities — a standardized defect rate scaled to one million chances for a defect.

DPU

Defects Per Unit — total defects divided by total units inspected.

First Time Yield (FTY)

The fraction of units that complete a step correctly the first time, without rework.

Fishbone (Ishikawa) diagram

A cause-and-effect diagram organizing potential causes by category (the 6 Ms) around a spine.

FMEA

Failure Mode and Effects Analysis — identifying failure modes and prioritizing them by a Risk Priority Number.

Gage R&R

A study of measurement-system Repeatability (one operator) and Reproducibility (different operators).

Hypothesis testing

A method to decide, with a stated risk, whether sample data supports a claim about a population.

Interaction (DOE)

When the effect of one factor on the response depends on the level of another factor.

Kaizen

Continuous improvement through small, incremental, team-driven changes.

Kanban

A visual signal that pulls work or replenishes inventory only when needed, limiting overproduction.

Kano model

A framework classifying requirements as basic needs, performance needs, and delighters to prioritize them.

Lean

A methodology focused on maximizing customer value while eliminating waste (non-value-added activity).

Measurement System Analysis (MSA)

A study quantifying how much of the observed variation comes from the measurement system itself.

Normal distribution

A symmetric, bell-shaped distribution defined by its mean and standard deviation.

Null hypothesis (H₀)

The default claim of no difference or no effect, which a test tries to disprove.

p-value

The probability of seeing data this extreme if the null hypothesis were true; if p ≤ α, reject the null.

Pareto chart

A bar chart ordering causes by frequency to highlight the 'vital few' that drive most of the problem.

Pareto principle

The 80/20 rule: roughly 80% of effects come from 20% of causes.

Poka-yoke

Mistake-proofing — designing a process so errors are prevented or caught immediately.

Pp and Ppk

Long-term performance indices, like Cp/Cpk but using the overall (long-term) standard deviation.

Process capability

How well a process meets its specification limits, comparing its natural spread to the spec width.

Process map

A flowchart of every step, decision, input, and output in a process as it actually runs.

Project charter

The document that authorizes a project, defining its problem, goal, scope, business case, and team.

Regression analysis

A method modeling how a response variable changes as one or more input variables change.

Repeatability

Variation when the same operator measures the same item multiple times with the same gage.

Reproducibility

Variation when different operators measure the same item with the same gage.

Rolled Throughput Yield (RTY)

The product of every step's first-time yield — the chance a unit passes the whole process defect-free.

Root cause analysis

Finding the fundamental cause of a problem rather than treating its symptoms.

RPN

Risk Priority Number = Severity × Occurrence × Detection in an FMEA; higher means higher priority.

Sigma level

The number of standard deviations between the process mean and the nearest specification limit; higher is better.

SIPOC

A high-level map of a process: Suppliers, Inputs, Process, Outputs, Customers — built in the Define phase.

Six Sigma

A data-driven methodology that reduces process variation and defects, targeting 3.4 defects per million opportunities.

Special-cause variation

Variation from an identifiable, assignable source that makes a process unstable.

Specification limits

The customer/engineering limits (USL, LSL) that define acceptable output; set by requirements, not the process.

Standard deviation

A measure of how spread out data is around the mean; the square root of the variance, written σ.

Standard work

The documented, current best way to perform a task, ensuring consistency and a baseline to improve.

Statistical Process Control (SPC)

Using control charts to monitor a process and tell common-cause from special-cause variation.

Takt time

The pace of customer demand = available production time ÷ demand; sets the rhythm of production.

Type I error

Rejecting a true null hypothesis — a false positive; its probability is α.

Type II error

Failing to reject a false null hypothesis — a false negative; its probability is β.

Value stream map

A lean diagram of material and information flow used to expose waste across a process.

Variance

The average of the squared deviations from the mean; the square of the standard deviation (σ²).

Voice of the Customer

The collected needs, wants, and expectations of customers, used to define what quality means.

Six Sigma Green Belt Study Guide FAQ

References

1. 1.American Society for Quality. “Certified Six Sigma Green Belt (CSSGB).” asq.org. ↑
2. 2.American Society for Quality. “Six Sigma Green Belt Body of Knowledge (BoK).” asq.org. ↑
3. 3.American Society for Quality. “DMAIC — The 5 Phases of Lean Six Sigma.” asq.org. ↑
4. 4.American Society for Quality. “Process Capability (Cp, Cpk).” asq.org. ↑
5. 5.American Society for Quality. “Control Chart.” asq.org. ↑
6. 6.American Society for Quality. “Quality Glossary.” asq.org. ↑
7. 100.American Society for Quality (ASQ). “SIPOC+CM Diagram.” asq.org, accessed 19 June 2026. ↑
8. 101.American Society for Quality (ASQ). “Fishbone (Cause-and-Effect) Diagram.” asq.org, accessed 19 June 2026. ↑

Career Employer

Career Employer is the ultimate resource to help you get started working the job of your dreams. We cover topics from general career information, career searching, exam preparation with free study materials, career interviewing, and becoming successful in your career of choice.

Follow Us:

All Posts