Measure — Baselining Current Performance
Measure converts the Define phase's intentions into trustworthy numbers: it establishes how the process performs today, using data you can defend. Green Belts classify the data they collect, map the process as it truly runs, prove the measurement system itself is reliable, and compute the defect and capability metrics that fix the baseline. A wrong baseline sinks the whole project, so this phase is as much about the integrity of the measurement as about the measurement itself.
Data types: continuous vs. attribute, and the four scales
The single most tested distinction here is continuous (variable) versus attribute (discrete) data. Continuous data comes from measuring on a scale that can, in principle, take any value between two points — length in centimeters, time in seconds, temperature, weight. Attribute data comes from counting or categorizing — pass/fail, number of scratches, defect type. Continuous data is preferred for analysis because it carries far more information per observation, so it reveals process behavior and detects change with much smaller sample sizes than attribute data. Beneath this sits the nominal-ordinal-interval-ratio hierarchy: nominal values only name categories with no order (defect type, machine ID); ordinal values rank without equal spacing (poor/fair/good/excellent); interval scales have equal spacing but no true zero, so ratios are meaningless (°C — 20° is not 'twice as hot' as 10°); ratio scales have equal spacing and a true zero, so ratios hold (length in cm, where 20 cm is genuinely twice 10 cm). Knowing the scale dictates which statistics and charts are legitimate.
Process mapping: flowcharts, value stream maps, and more
Before measuring, the team documents how the process actually runs. A detailed flowchart shows the real sequence of steps and decisions, exposing rework loops, redundant checks, and hidden hand-offs; on standard flowchart notation a diamond is a decision point and a rectangle is an activity. A swim-lane (deployment) flowchart adds columns for each person or department, making hand-offs and ownership gaps visible — useful when the problem lives in the transitions between groups. A value stream map (VSM) goes further than a basic flowchart by overlaying data — cycle times, wait times, inventory, and the information flow that triggers each step — so the team can separate value-added from non-value-added time and compute lead time. A spaghetti diagram traces the physical path of a person, part, or document to reveal wasted motion and transport. Each map answers a different question, and the exam expects you to match the tool to the need.
Sampling: representative data without measuring everything
Measuring an entire population is usually impractical, so teams sample. In simple random sampling every item has an equal and independent chance of selection, which minimizes selection bias and lets the sample stand in for the population. Stratified sampling divides the population into meaningful groups (shifts, machines, regions) and samples within each to guarantee representation of every group. Systematic sampling takes every k-th item, which is efficient but dangerous if the process has a cycle that lines up with the interval. The goals are to avoid bias and to collect enough data that estimates are stable; a biased sample produces a confident but wrong baseline, which is worse than no baseline at all.
Measurement System Analysis and Gage R&R
Every observed value is the true value plus measurement error, so before trusting any baseline you must prove the measurement system is adequate — this is Measurement System Analysis (MSA). For continuous data the workhorse is Gage Repeatability and Reproducibility (Gage R&R), which partitions measurement variation into two parts. Repeatability is the variation when one appraiser measures the same item multiple times with the same gage — the equipment's own inconsistency. Reproducibility is the variation between different appraisers measuring the same items — the human/method inconsistency. A common rule of thumb reads the percentage of total variation consumed by measurement error: under 10% is acceptable, 10-30% is marginal (accept depending on cost and application), and over 30% means the measurement system must be fixed before the data can be used. For attribute data (pass/fail judgments) an attribute agreement analysis checks whether appraisers agree with each other and with a known standard. Beyond R&R, a good system must also be accurate (low bias against a reference), linear (bias constant across the range), and stable (unchanging over time).
Defect metrics: DPU, DPO, and DPMO worked
These metrics let processes of different complexity be compared on one scale. Defects Per Unit is DPU = total defects ÷ total units. Defects Per Opportunity divides by the chances to err: DPO = defects ÷ (units × opportunities per unit). Defects Per Million Opportunities scales that to a million: DPMO = DPO × 1,000,000. Worked example from the bank: inspect 500 units, each with 8 opportunities, and find 20 defects. Total opportunities = 500 × 8 = 4,000. DPO = 20 ÷ 4,000 = 0.005. DPMO = 0.005 × 1,000,000 = 5,000. The 'opportunity count' is a modeling choice the team must define consistently, because inflating opportunities artificially lowers DPMO — a common way teams flatter their sigma level. Note the difference between a defect (any single instance of nonconformance) and a defective (a unit with one or more defects); a single defective unit can carry several defects.
Yield, sigma level, and the 1.5-sigma shift
Yield is the fraction of output that is good. First-pass (or first-time) yield counts only units that pass without any rework, while rolled throughput yield (RTY) multiplies the first-pass yields of every step — RTY = Y₁ × Y₂ × … × Yₙ — exposing the 'hidden factory' of rework that final-inspection yield conceals. A four-step process each at 0.95 yield has RTY = 0.95⁴ ≈ 0.815, far below the 0.95 any single step reports. Process sigma expresses how many standard deviations fit between the process mean and the nearest specification limit; more sigmas mean fewer defects. By convention Six Sigma quotes long-term sigma with a 1.5σ shift built in to account for drift over time, which is why the familiar landmarks are ≈690,000 DPMO at 1σ, ≈66,807 DPMO at 3σ, ≈6,210 at 4σ, ≈233 at 5σ, and ≈3.4 DPMO at 6σ. Memorize 3.4 DPMO = six sigma and ~66,807 DPMO ≈ three sigma; the exam converts between DPMO and sigma using this table.
Keep going: the full Lean Six Sigma Green Belt (IASSC ICGB) guide covers every section of the exam. Lean Six Sigma Green Belt — Complete Study Guide (2026) — PDF + EPUB, $19.99 · 14-day refund →
Practice stays free. The full Lean Six Sigma Green Belt (IASSC ICGB) study guide is the material itself, taught start to finish — a downloadable PDF + EPUB you keep.