Do you need Minitab for Six Sigma? To pass the exam, no. You cannot use it in the exam room anyway. To run real projects, you need some statistics software, and Excel can carry you further than most people think.
This page answers the three questions students ask me most about tools: what can I bring into the exam, which calculator should I use, and can I do the statistics in Excel or by hand?
What you can bring into the exam, by certifying body
Rules change. I checked each body’s own pages in October 2026. Confirm with your body before exam day.
| Body | Open or closed book | Calculator | Software (Minitab, Excel) |
|---|---|---|---|
| ASQ (test center or remote) | Open book. All references, including notes, must be bound and stay bound. Hand-stapled pages and loose notes are not allowed. | An on-screen basic scientific calculator is built into every computer-based exam. At a test center you may bring a silent, hand-held, battery-operated calculator with no programmable memory. ASQ names the TI-89 as not allowed. | No. |
| IASSC | Closed book. IASSC gives you its own Reference Document as a download at the start of the exam. Do not bring your own copy. | A non-programmable four-function or scientific calculator. IASSC lists the SHARP EL 501XBWH, TI-34, TI-30X and Casio FX-115 as examples. Graphing and programmable models (TI-83, TI-84, Casio FX9750, FX5800P) are not allowed. Neither are phone calculators. | No. IASSC names Minitab and SigmaXL as not permitted. |
| Villanova | Open book. Villanova confirmed to us in October 2026 that you may use course materials, notes and other resources. | Not published. Check with Villanova. | Not published. Check with Villanova. |
| ILSSI | Open book. ILSSI confirmed to us in October 2026 that printed materials, notes or a PDF of the course materials are allowed. | Not published. Check with ILSSI. | Not published. Check with ILSSI. |
Sources: ASQ’s Six Sigma Black Belt certification page (the same rules appear on the Green and Yellow Belt pages) and its remote proctoring FAQ; IASSC’s permitted resources and open book answers; Villanova’s certification requirements; ILSSI’s certification page.
ASQ remote exams have extra rules
If you take an ASQ exam at home through remote proctoring, three things change.
- No scratch paper and no pens. You use the Scratchpad inside the exam software instead. Practice your math without paper at least once before the day.
- Your references still have to be bound.
- ASQ’s FAQ allows “built-in calculators” and asks you to show your references and calculator at check-in. It does not say plainly whether a hand-held calculator is fine at home. If you want to use one, ask ASQ before exam day.
IASSC: learn the reference document before the exam
IASSC exams are closed book, but you are not on your own. The IASSC Reference Document holds the formulas and tables you get on screen. IASSC links to it from its permitted resources page. Download it now and study with it, so you know where everything sits when the clock is running.
Which calculator should you use for the Six Sigma exam?
Use a plain, non-programmable scientific calculator. ASQ and IASSC both allow one, and it does everything the exam needs.
Make sure it has these keys, and that you can find them without looking:
- ln and e^x for rolled throughput yield, Poisson and exponential questions
- x^y and √ for NPV, binomial and standard deviation work
- x² and memory keys (M+, MR) for sums of squares
- Σ or statistics mode, if your body allows it, for mean and standard deviation. Check that your model is non-programmable.
Buy it early. Do every practice exam with it. The middle of the exam is a bad time to learn where the inverse key is.
Some links on this page are affiliate links. If you buy through one, I may earn a small commission. It costs you nothing extra. As an Amazon Associate I earn from qualifying purchases.
I used an HP 10bII (HP sells the current model as the 10bII+) for my ASQ Six Sigma Black Belt exam. It is a financial calculator, not a programmable one, so it fits ASQ’s rule. It has the statistics keys (mean and standard deviation) and NPV and IRR built in, which saves time on the project selection questions. A plain scientific model works too: IASSC names the TI-30X and the Casio FX-115 as examples, and both meet ASQ’s rule. (Here is how I studied and passed.)
Do you need Minitab to pass the Six Sigma exam?
No. No certifying body above lets you use Minitab or any other software in the exam. The exam tests whether you can pick the right tool, run the numbers with a formula, a table and a calculator, and read the result.
What you do need:
- The formulas. Get my free Six Sigma Formula Reference Sheet from the ASQ Green Belt or ASQ Black Belt study guide page. Members of the Pass Your Six Sigma Green Belt and Black Belt courses also get my Black Belt Formula Sheet PDF in the course downloads: the business and process metrics the statistics sheets skip (NPV, DPMO to sigma level, RTY, takt time, OEE and the distributions), each with a worked example.
- The tables: z, t, chi-square and F. Learn to read each one. Our Z table guide explains the two common z table layouts.
- Practice reading results. Exam-style questions can show an ANOVA table or a regression printout and ask what it means. Learn what each column is.
- Knowing which test to use. The Hypothesis Tests Study Guide lays out the choices.
Software helps you study in one way: you can check your hand calculations. Run a practice problem by hand, then in Excel. If the answers match, you understand it.
Do you need Minitab for Six Sigma projects?
For real projects, you need some statistics software. Real data sets are too big for a calculator. Many Six Sigma courses teach with Minitab. If your employer has a license, use it. Minitab offers a free trial.
If you do not have a license, these options cost little or nothing:
| Tool | Cost | Good for | Limits |
|---|---|---|---|
| Excel with the Analysis ToolPak | Included with desktop Excel (Windows and Mac). You switch it on as an add-in. | t-tests, F-test, one- and two-way ANOVA, regression, correlation, descriptive statistics, histograms | No control charts, capability, Gage R&R or DOE menus. You build those from formulas. |
| Real Statistics Resource Pack for Excel | Free download | Adds many tests the ToolPak lacks, including more ANOVA and regression options and non-parametric tests | An add-in written by one author; check its version against your Excel version. |
| JASP | Free, open source | Point-and-click statistics with a Quality Control module: control charts, process capability, Gage R&R and other measurement system studies, DOE worksheets | A separate program, not Excel. Some menus are named differently from Minitab. |
| R | Free, open source | Anything, if you are willing to write code. The qcc package covers control charts and capability. | A learning curve. Best if you or your team already code. |
Commercial Excel add-ins built for Six Sigma also exist. SigmaXL is one; IASSC names it in its rules. Compare price and features against what your projects need.
To turn on the Analysis ToolPak in Excel for Windows: File, Options, Add-ins, choose Excel Add-ins in the Manage box, click Go, tick Analysis ToolPak. On a Mac: Tools, Excel Add-ins. Microsoft’s Analysis ToolPak guide lists every tool. Once it is on, look for Data Analysis on the Data tab.
Free templates for Pareto charts, box plots and more are on our Six Sigma Excel templates page.
How to run the main Six Sigma tests in Excel or by hand
Each row shows the Excel route and the by-hand formula. The by-hand version is what the exam tests.
| Test | In Excel | By hand (exam) | Learn it |
|---|---|---|---|
| One-sample t-test | No ToolPak tool. Compute t yourself, then p = T.DIST.2T(ABS(t), n-1) | t = (x̄ – μ₀) / (s / √n), df = n – 1, compare to the t table | One sample t test |
| Two-sample t-test | ToolPak: t-Test: Two-Sample Assuming Equal Variances (or Unequal). Or =T.TEST(A, B, 2, 2) for equal variances, type 3 for unequal | Pooled: t = (x̄₁ – x̄₂) / (s_p √(1/n₁ + 1/n₂)) | Two sample t tests |
| Paired t-test | ToolPak: t-Test: Paired Two Sample for Means. Or =T.TEST(A, B, 2, 1) | t = d̄ / (s_d / √n) on the differences | Paired t test |
| F-test for two variances | ToolPak: F-Test Two-Sample for Variances. =F.TEST(A, B) gives a two-tailed p | F = s₁² / s₂² (larger on top), compare to the F table | F-test |
| One-way ANOVA | ToolPak: Anova: Single Factor. Two-factor versions are there too | F = MS between / MS within, from the sums of squares table | ANOVA |
| Chi-square test of independence | No ToolPak tool. Build the expected table (row total x column total / grand total), then =CHISQ.TEST(observed, expected) gives p | χ² = Σ (O – E)² / E, df = (rows – 1)(columns – 1) | Chi-square test of independence |
| Simple linear regression | ToolPak: Regression (gives the ANOVA table, R², coefficients and p-values). Or =SLOPE, =INTERCEPT, =RSQ, =LINEST | b₁ = Sxy / Sxx, b₀ = ȳ – b₁x̄ | Linear regression |
| Control limits (I-MR) | No built-in chart. Compute the mean, the moving ranges and their average, then the limits with the formulas | UCL/LCL = x̄ ± 2.66 x MR̄; UCL for the MR chart = 3.267 x MR̄ | I-MR chart |
| Control limits (X-bar R) | Same approach with the subgroup means and ranges | x̿ ± A₂R̄; R chart: D₃R̄ to D₄R̄ (constants from the table for your subgroup size) | X-bar R charts |
| Cp and Cpk | =(USL-LSL)/(6*s) and =MIN(USL-mean, mean-LSL)/(3*s). Use the within-subgroup sigma (R̄/d₂) for Cp and Cpk; =STDEV.S of all the data gives Pp and Ppk | Same formulas | Process capability (Cp and Cpk) |
| Normal probabilities | =NORM.S.DIST(z, TRUE) for the area to the left; =NORM.S.INV(p) for the reverse | z = (x – μ) / σ and the z table | Z scores and the z table |
Worked example: a two-sample t-test both ways
Two lines, five cycle times each (minutes).
- Line A: 12, 14, 11, 13, 15. Mean 13.0. Standard deviation 1.581.
- Line B: 15, 17, 16, 14, 18. Mean 16.0. Standard deviation 1.581.
By hand. The standard deviations are equal, so the pooled s_p is 1.581. The standard error is 1.581 x √(1/5 + 1/5) = 1.581 x 0.632 = 1.000. So t = (13.0 – 16.0) / 1.000 = -3.00. With 8 degrees of freedom, the two-tailed critical t at α = 0.05 is 2.306. Since |t| = 3.00 is bigger than 2.306, reject the null. The lines differ.
In Excel. Put A in A2:A6 and B in B2:B6. =T.TEST(A2:A6, B2:B6, 2, 2) returns 0.017. Below 0.05. Same conclusion.
Worked example: I-MR control limits by hand
Eight readings: 10.2, 9.8, 10.5, 10.1, 9.9, 10.4, 10.0, 10.3.
- Mean = 81.2 / 8 = 10.15
- Moving ranges: 0.4, 0.7, 0.4, 0.2, 0.5, 0.4, 0.3. Average MR̄ = 2.9 / 7 = 0.414
- Individuals chart: UCL = 10.15 + 2.66 x 0.414 = 11.25. LCL = 10.15 – 2.66 x 0.414 = 9.05
- Moving range chart: UCL = 3.267 x 0.414 = 1.35. LCL = 0.
In Excel, the same numbers come from =AVERAGE, =ABS(B3-B2) filled down, and the two formulas. Add a line chart and draw the limits as extra series. That is an I-MR chart without Minitab.
Worked example: chi-square by hand
A survey asks 100 people two yes/no questions. The counts are 30 and 20 in the first row, 20 and 30 in the second. Every expected count is 50 x 50 / 100 = 25. Each cell adds (5)² / 25 = 1, so χ² = 4.0. The critical value at α = 0.05 with 1 degree of freedom is 3.841. Reject independence. In Excel, =CHISQ.TEST on the observed and expected tables returns 0.046.
The short version
- Exam: a non-programmable scientific calculator, the formulas, the tables, and practice reading output. No Minitab.
- Projects: Minitab if your employer has it. Otherwise Excel with the Analysis ToolPak, plus JASP or R for control charts and capability.
- Rules: check your body’s current page before exam day. They change.
If arithmetic is the part that worries you, the Math Confidence for Six Sigma course starts from square roots, logs and lookup tables.
External resources
- Microsoft: Use the Analysis ToolPak: every tool in the add-in, and how to load it.
- NIST/SEMATECH e-Handbook: control chart constants: A₂, D₃ and D₄ by subgroup size.
- JASP features: the full list of analyses, including the Quality Control module.