Which Test, When? A Simple Guide to Picking the Right Statistics Tool for Six Sigma


Which Test, When? A Simple Guide to Picking the Right Statistics Tool for Six Sigma

Statistical questions on exams often feel like riddles because they dress a simple decision in formal language. The fastest way to stop guessing is to translate the prompt into one clear question, then pick the smallest, most direct test that answers it.

Translate the prompt into plain English first

When you see a question, underline these three things: the variable type (mean, proportion, variance), the number of groups (one, two, many), and the relationship scope (compare or relate). Translate those into a one-line instruction: “compare averages of two groups” or “compare proportions across groups” or “measure association between two numeric variables.” That one-line translation almost always points directly to the right test.

A compact decision map you can memorize

One numeric variable, compare to a target: single-sample test (is the average different from X?)

Two numeric groups, compare averages: two-sample t-test (paired if same subjects before/after)

More than two groups, compare averages: ANOVA (then post-hoc where necessary)

Comparing proportions (%): chi-square or two-proportion test

Relationship between two numeric variables: correlation/regression

Checking variability or spread: variance tests or simple capability metrics

Practical short-cuts that save time on test day

Numbers vs percentages: if the question is about percent defective or proportion, think proportions, not t-tests.

Paired vs independent: if the same units are measured twice (before/after), it’s paired. If samples are different groups, it’s independent.

Many groups: if the question mentions three or more treatments/groups, reach for ANOVA logic rather than multiple t-tests.

How to practice this so it sticks

Translation drill: read a question, write the one-line translation, then state the test, no calculation. Repeat 20 times.

Application drill: pick 5 questions and run the actual calculation or interpretation; review only those where your translation was wrong.

Five short translation practice items (reply with your 1-line translations; I’ll tell you if they’re correct)

-Did the new process reduce average cycle time?

-Did the proportion of defective units change after the process change?

-Are the average scores different across three training groups?

-Is there a relationship between hours studied and exam score?

-Has the variance in processing time increased after a change?

Reply with your one-line translation for any of the items above (e.g., "compare averages of two groups") and I’ll tell you whether that translation points to the right test.

Two tiny examples (translate, then test)

1) “Did the new process reduce average cycle time?” → translate: compare two averages → two-sample test (or paired if same lines before/after).

2) “Did the proportion of defective units change after the change?” → translate: compare two proportions → two-proportion test or chi-square.

One way to reduce second-guessing under time pressure

Give yourself one extra step during practice: before you touch a calculator, write the translation in 7 words or fewer. This habit prevents wasted time on the wrong calculation and is the single most efficient way to reduce errors.

Best, Ted

P.S. If you’d like to practice, reply with one of the five items above and your one-line translation; I’ll tell you whether you picked the right test.

When you're ready, there are a few ways I can help:

First, join 30,000+ other Six Sigma professionals by subscribing to my email newsletter. A short read every Monday to start your work week off correctly. Always free.

If you're looking to pass your Six Sigma Green Belt or Black Belt exams, I'd recommend starting with my affordable study guide:

1)→ 🟢Pass Your Six Sigma Green Belt

2)→ ⚫Pass Your Six Sigma Black Belt

You've spent so much effort learning Lean Six Sigma. Why leave passing your certification exam up to chance? This comprehensive study guide offers 1,000+ exam-like questions for Green Belts (2,000+ for Black Belts) with full answer walkthroughs, access to instructors, detailed study material, and more.

Join 10,000+ students here.

A few readers have said things like:

“I passed Villanova’s Black Belt exam on the first try. The sample tests were a huge help, and Ted’s responses kept me going.” — Sara M. (LinkedIn)

Welcome to the Six Fix!

Weekly newsletter to help you pass your Six Sigma exam and apply Six Sigma techniques in the real world.

Read more from Welcome to the Six Fix!

Most people don't have a time problem. They have a feedback problem. You sit down to study with every intention of making progress, and an hour later you look up wondering where the time went. Long sessions can feel productive. They rarely are. That gap between effort and learning is exactly what Pomodoro fixes. What I see over and over is this: capable people doing capable work, but without a tight feedback loop. Pomodoro converts unfocused time into short, measurable practice with constant...

How to Turn a Green Belt Project into Career Momentum Over the past few months, I’ve had more people reach out asking for career advice than I have in years. Building valuable skills still matters, but there’s another piece people miss: you have to be able to clearly show what you did and why it mattered. What’s interesting is this: most of the people I talk to are doing good work. They’ve led projects, improved processes, delivered real results. But when I ask them to explain what they...

Concrete is the backbone of modern construction. But what happens when the quality of cement begins to fail, and the concrete doesn’t set on time? This isn’t just a nuisance. Delays can cause costly project delays, quality issues, and even safety risks. Introduction: A Hypothetical DMAIC Case Study Inspired by Readers This article presents a hypothetical Six Sigma case study on delayed setting time of cement in concrete, structured using the DMAIC (Define, Measure, Analyze, Improve, Control)...