Study guide

9.4 Analysis of Variance (One-Way & Two-Way ANOVA) & Chi-Square Contingency Tests

CSSC Certified Six Sigma Green Belt Exam

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Six Sigma Overview & Exam Fundamentals

1.1 CSSC Certification Overview, Exam Structure & Body of Knowledge 1.2 History & Evolution of Six Sigma 1.3 Six Sigma Organizational Roles & Belts Hierarchy 1.4 The DMAIC Methodology Architecture vs. DMADV / DFSS 1.5 Other Process Improvement & Quality Methods, and When to Choose Six Sigma

2 Chapter 2: Quality Fundamentals & Defect Metrics

2.1 Critical to Quality (CTQ) Characteristics & VOC Translation 2.2 Defect Metrics: DPU, DPO, DPMO, & the 1.5-Sigma Shift 2.3 First Time Yield (FTY) & Rolled Throughput Yield (RTY) 2.4 Cost of Quality (COQ) & Cost of Poor Quality (COPQ)

3 Chapter 3: Lean Principles & Waste Elimination

3.1 Foundations of Lean & The Toyota Production System (TPS) 3.2 The Eight Wastes: CSSC's Seven Muda, TIMWOODS, and Muri/Mura 3.3 The 5S Visual Workplace Methodology 3.4 Value Stream Mapping (VSM), Cycle Time & Takt Time

4 Chapter 4: Project Selection & Team Management

4.1 Six Sigma Project Selection & Business Case Development 4.2 Project Charter Architecture 4.3 Stakeholder Management & Communication Planning 4.4 Team Dynamics & Tuckman's Team Development Stages 4.5 The 15-Point Project Viability Model & Process-Level Selection

5 Chapter 5: Define Phase & Process Mapping

5.1 High-Level Process Mapping with SIPOC 5.2 Flowcharts, Swimlane Diagrams & Process Value Analysis 5.3 Quality Function Deployment (QFD) & The House of Quality 5.4 Critical to Quality (CTQ) Flowdown & Operational Definitions 5.5 The CSSC Define Toolset: Ground Rules, In/Out of the Box, Is–Is Not & Tollgate Checklists

6 Chapter 6: Measure Phase & Measurement Systems

6.1 Data Types: Continuous (Variable) vs. Discrete (Attribute) Data 6.2 Data Collection Planning & Sampling Strategies 6.3 Measurement System Analysis (MSA) Foundations & Calibration 6.4 Variable & Attribute Gage R&R Studies

7 Chapter 7: Analyze Phase & Root Cause Analysis

7.1 Qualitative Root Cause Tools: 5 Whys & Cause-and-Effect (Fishbone) Diagrams 7.2 Failure Mode and Effects Analysis (FMEA) & RPN Calculation 7.3 Pareto Analysis & The 80/20 Rule 7.4 Graphical Data Exploration: Histograms, Box Plots, & Multi-Vari Charts 7.5 The Root Cause Verification Matrix

8 Chapter 8: Descriptive Statistics & Distributions

8.1 Measures of Central Tendency & Dispersion 8.2 The Normal Distribution & The Empirical Rule (68-95-99.7) 8.3 Standard Normal Distribution, Z-Scores & Probability Calculations 8.4 Common Non-Normal Distributions: Binomial, Poisson & Chi-Square

9 Chapter 9: Inferential Statistics & Hypothesis Testing

9.1 Fundamentals of Hypothesis Testing: Null vs. Alternative, Alpha & Beta Risks 9.2 One-Sample and Two-Sample t-Tests for Means 9.3 Paired t-Tests & Non-Parametric Alternatives 9.4 Analysis of Variance (One-Way & Two-Way ANOVA) & Chi-Square Contingency Tests 9.5 Sample Size Determination for Hypothesis Tests

10 Chapter 10: Correlation, Regression, & Experimentation

10.1 Linear Correlation & Pearson's Correlation Coefficient (r) 10.2 Simple Linear Regression & Coefficient of Determination (R^2) 10.3 Introduction to Design of Experiments (DOE) & Factorial Designs 10.4 Analyzing Experimental Results: Main Effects & Interaction Plots

11 Chapter 11: Improve Phase & Solution Implementation

11.1 Brainstorming, Creative Thinking & Solution Selection Matrices 11.2 Piloting Solutions & Conducting Risk Assessments 11.3 Implementation Planning & Change Management (ADKAR) 11.4 Mistake-Proofing (Poka-Yoke) & Standard Work

12 Chapter 12: Control Phase, SPC, & Process Capability

12.1 Statistical Process Control (SPC) & Shewhart Control Rules 12.2 Variable Control Charts: I-MR, X-bar & R, X-bar & S 12.3 Attribute Control Charts: p, np, c, and u Charts 12.4 Process Capability & Performance Indices: Cp, Cpk, Pp, Ppk 12.5 Communicating Statistical Results to Business Audiences

9.4 Analysis of Variance (One-Way & Two-Way ANOVA) & Chi-Square Contingency Tests

Key Takeaways Explain This Topic AI

One-Way ANOVA tests for differences in means across three or more groups (k ≥ 3) simultaneously; running multiple pairwise t-tests is prohibited because it rapidly inflates the family-wise Type I error rate (α_family = 1 - (1 - α)^c).

The ANOVA F-statistic partitions total variation into between-group variance (MS_Between / signal) and within-group error variance (MS_Within / noise): F = MS_Between / MS_Within; a significant F-test requires post-hoc testing (e.g., Tukey's HSD) to identify specific differing pairs.

Two-Way ANOVA evaluates two independent factors simultaneously, testing for Main Effects of Factor A, Main Effects of Factor B, and their Interaction Effect (A × B), which indicates whether the effect of one factor depends on the level of the other.

The Chi-Square (χ²) Test of Independence analyzes contingency tables to test for an association between two categorical variables by comparing observed frequencies (O) against expected frequencies (E), with df = (r - 1)(c - 1).

The Chi-Square test requires all expected cell counts E_ij ≥ 5; if expected frequencies fall below 5, categories must be combined or Fisher's Exact Test must be utilized.

Last updated: September 2026

9.4 Analysis of Variance (One-Way & Two-Way ANOVA) & Chi-Square Contingency Tests

Quick Summary: When continuous improvement investigations expand beyond two groups or involve categorical attribute variables, practitioners deploy advanced inferential tools. One-Way Analysis of Variance (ANOVA) compares continuous means across three or more groups ( k ≥ 3 k \ge 3 k ≥ 3 ) simultaneously using the F F F -ratio ( M S Between / M S Within MS_{\text{Between}} / MS_{\text{Within}} M S Between ​ / M S Within ​ ), preventing the massive Type I error inflation caused by multiple pairwise t t t -tests. When two factors influence the process, Two-Way ANOVA evaluates both main effects and their interaction effect . For categorical and discrete defect counts, the Chi-Square ( χ 2 \chi^2 χ 2 ) Test of Independence evaluates contingency tables to detect associations between process attributes.

  • One-Way Analysis of Variance (ANOVA)
  • Purpose & Operational Context

One-Way ANOVA (Analysis of Variance) is a parametric hypothesis test used to compare the means of three or more independent groups ( k ≥ 3 k \ge 3 k ≥ 3 ) on a single continuous response variable ( Y Y Y ). Examples include:

Comparing tensile strength across four raw material resin suppliers.

Comparing order cycle time across three operating shifts (Day, Evening, Night).

Comparing dimensional accuracy across five CNC machining centers.

Hypotheses Formulation

H 0 : μ 1 = μ 2 = μ 3 = ⋯ = μ k H_0: \mu_1 = \mu_2 = \mu_3 = \dots = \mu_k H 0 ​ : μ 1 ​ = μ 2 ​ = μ 3 ​ = ⋯ = μ k ​ (All group population means are equal)

H a : H_a: H a ​ : At least one group mean is different from the others

Crucial Exam Rule: The alternative hypothesis in ANOVA is NOT "all means are different." H a H_a H a ​ simply states that at least one group mean is statistically distinguishable from at least one other group mean. In fact, three groups could have identical means while a fourth group is radically higher, satisfying H a H_a H a ​ .

Why Multiple t t t -Tests Are Strictly Prohibited: α \alpha α -Inflation

A frequent novice mistake in Six Sigma is comparing four groups by running separate two-sample t t t -tests on every pair (Group 1 vs 2, 1 vs 3, 1 vs 4, 2 vs 3, 2 vs 4, 3 vs 4).

The number of pairwise comparisons ( c c c ) required for k k k groups is: c = k ( k − 1 ) 2 c = \frac{k(k - 1)}{2} c = 2 k ( k − 1 ) ​

If each individual test is conducted at significance level α = 0.05 \alpha = 0.05 α = 0.05 , the family-wise Type I error rate (the overall probability of committing at least one false alarm across the family of tests) inflates exponentially: α family = 1 − ( 1 − α ) c \alpha_{\text{family}} = 1 - (1 - \alpha)^c α family ​ = 1 − ( 1 − α ) c

Number of Groups ( k k k ) Number of Pairs ( c c c ) Family-Wise Error Rate ( α family \alpha_{\text{family}} α family ​ at α = 0.05 \alpha = 0.05 α = 0.05 )

k = 3 k = 3 k = 3 3 3 3 1 − ( 0.95 ) 3 = 1 − 0.857 = 14.3 % 1 - (0.95)^3 = 1 - 0.857 = \mathbf{14.3\%} 1 − ( 0.95 ) 3 = 1 − 0.857 = 14.3%

k = 4 k = 4 k = 4 6 6 6 1 − ( 0.95 ) 6 = 1 − 0.735 = 26.5 % 1 - (0.95)^6 = 1 - 0.735 = \mathbf{26.5\%} 1 − ( 0.95 ) 6 = 1 − 0.735 = 26.5%

k = 5 k = 5 k = 5 10 10 10 1 − ( 0.95 ) 10 = 1 − 0.599 = 40.1 % 1 - (0.95)^{10} = 1 - 0.599 = \mathbf{40.1\%} 1 − ( 0.95 ) 10 = 1 − 0.599 = 40.1%

k = 6 k = 6 k = 6 15 15 15 1 − ( 0.95 ) 15 = 1 − 0.463 = 53.7 % 1 - (0.95)^{15} = 1 - 0.463 = \mathbf{53.7\%} 1 − ( 0.95 ) 15 = 1 − 0.463 = 53.7%

For 5 groups, running separate t t t -tests produces a 40.1 % 40.1\% 40.1% chance of falsely claiming a significant difference when none exists! ANOVA solves this problem by conducting a single, omnibus test that evaluates all means simultaneously while strictly holding α = 0.05 \alpha = 0.05 α = 0.05 .

The Mechanics of the F-Statistic & The ANOVA Table

ANOVA evaluates differences in means by analyzing and partitioning variances . The total variation in the data is split into two independent components: Partitioning Total Variance

  • TOTAL VARIATION
  • (SS_Total)
  • │
  • ┌────────────────┴────────────────┐
  • ▼ ▼
  • BETWEEN-GROUP VARIATION WITHIN-GROUP VARIATION
  • (SS_Between) (SS_Within)
  • Variation caused by Random, unexplained
  • Factor Treatment background noise / error

Between-Group Variance ( S S Between SS_{\text{Between}} S S Between ​ or S S Factor SS_{\text{Factor}} S S Factor ​ ) : Measures how far the individual group sample means deviate from the overall grand mean ( x ˉ ˉ \bar{\bar{x}} x ˉ ˉ ). This represents the signal generated by process differences.

Within-Group Variance ( S S Within SS_{\text{Within}} S S Within ​ or S S Error SS_{\text{Error}} S S Error ​ ) : Measures how individual observations deviate from their respective group sample mean. This represents inherent, unexplained random noise .

Mean Squares & The F F F -Ratio

Dividing Sum of Squares ( S S SS SS ) by their respective Degrees of Freedom ( d f df df ) yields Mean Squares ( M S MS MS ) : F calc = M S Between M S Within = Between-Group Variance (Signal) Within-Group Variance (Noise) F_{\text{calc}} = \frac{MS_{\text{Between}}}{MS_{\text{Within}}} = \frac{\text{Between-Group Variance (Signal)}}{\text{Within-Group Variance (Noise)}} F calc ​ = M S Within ​ M S Between ​ ​ = Within-Group Variance (Noise) Between-Group Variance (Signal) ​

If F ≈ 1 F \approx 1 F ≈ 1 : Between-group differences are no larger than random process noise ( p > 0.05 p > 0.05 p > 0.05 ; Fail to reject H 0 H_0 H 0 ​ ).

If F ≫ 1 F \gg 1 F ≫ 1 : The variance between group averages is substantially larger than random background noise ( p ≤ 0.05 p \le 0.05 p ≤ 0.05 ; Reject H 0 H_0 H 0 ​ ).

The Standard ANOVA Source Table

Source of Variation Sum of Squares ( S S SS SS ) Degrees of Freedom ( d f df df ) Mean Square ( M S MS MS ) F F F -Statistic p p p -value

Between Groups (Factor) S S Factor SS_{\text{Factor}} S S Factor ​ k − 1 k - 1 k − 1 M S Factor = S S Factor k − 1 MS_{\text{Factor}} = \frac{SS_{\text{Factor}}}{k - 1} M S Factor ​ = k − 1 S S Factor ​ ​ F = M S Factor M S Error F = \frac{MS_{\text{Factor}}}{MS_{\text{Error}}} F = M S Error ​ M S Factor ​ ​ From F F F -dist

Within Groups (Error) S S Error SS_{\text{Error}} S S Error ​ N − k N - k N − k M S Error = S S Error N − k MS_{\text{Error}} = \frac{SS_{\text{Error}}}{N - k} M S Error ​ = N − k S S Error ​ ​ — —

Total S S Total SS_{\text{Total}} S S Total ​ N − 1 N - 1 N − 1 — — —

(Where $k = $ number of groups, $N = $ total observations across all groups).

Step-by-Step Worked Example: Interpreting an ANOVA Table

Scenario: A Green Belt compares the yield strength of metal brackets across k = 4 k = 4 k = 4 independent heat-treatment recipes. A total of N = 24 N = 24 N = 24 brackets were tested ( n = 6 n = 6 n = 6 brackets per recipe). The partial ANOVA table shows:

  • S S Factor = 180.0 SS_{\text{Factor}} = 180.0 S S Factor ​ = 180.0
  • S S Error = 120.0 SS_{\text{Error}} = 120.0 S S Error ​ = 120.0

Calculate Degrees of Freedom, Mean Squares, and F F F -Statistic:

  • d f Factor = k − 1 = 4 − 1 = 3 df_{\text{Factor}} = k - 1 = 4 - 1 = 3 d f Factor ​ = k − 1 = 4 − 1 = 3
  • d f Error = N − k = 24 − 4 = 20 df_{\text{Error}} = N - k = 24 - 4 = 20 d f Error ​ = N − k = 24 − 4 = 20
  • d f Total = N − 1 = 24 − 1 = 23 df_{\text{Total}} = N - 1 = 24 - 1 = 23 d f Total ​ = N − 1 = 24 − 1 = 23
  • M S Factor = 180.0 3 = 60.0 MS_{\text{Factor}} = \frac{180.0}{3} = 60.0 M S Factor ​ = 3 180.0 ​ = 60.0
  • M S Error = 120.0 20 = 6.0 MS_{\text{Error}} = \frac{120.0}{20} = 6.0 M S Error ​ = 20 120.0 ​ = 6.0

F calc = M S Factor M S Error = 60.0 6.0 = 10.00 F_{\text{calc}} = \frac{MS_{\text{Factor}}}{MS_{\text{Error}}} = \frac{60.0}{6.0} = 10.00 F calc ​ = M S Error ​ M S Factor ​ ​ = 6.0 60.0 ​ = 10.00

At α = 0.05 \alpha = 0.05 α = 0.05 , the critical F 0.05 , 3 , 20 = 3.10 F_{0.05, 3, 20} = 3.10 F 0.05 , 3 , 20 ​ = 3.10 . Because F calc = 10.00 > 3.10 F_{\text{calc}} = 10.00 > 3.10 F calc ​ = 10.00 > 3.10 ( p < 0.001 p < 0.001 p < 0.001 ), we Reject H 0 H_0 H 0 ​ . Heat-treatment recipe significantly affects yield strength.

Post-Hoc Analysis: Tukey's Honestly Significant Difference (HSD)

A statistically significant ANOVA test is an omnibus test : it proves that a difference exists somewhere among the means, but it does not identify which specific groups differ.

To identify which groups differ without inflating Type I error, practitioners run a post-hoc multiple comparison test . The gold standard in Six Sigma is Tukey's Honestly Significant Difference (HSD) test.

Interpreting Tukey's Grouping Output

Statistical software groups means using letters. Groups that do not share a common letter are statistically significantly different: Tukey HSD Output Interpretation

  • Recipe Sample Size Mean Yield Grouping
  • ──────────────────────────────────────────────────────
  • Recipe C 6 142.5 MPa A
  • Recipe A 6 138.0 MPa A
  • Recipe D 6 124.2 MPa B
  • Recipe B 6 112.8 MPa C

Recipes C and A share letter A : No significant difference between C and A.

Recipe D has letter B : Recipe D is significantly lower than A and C, but significantly higher than B.

Recipe B has letter C : Recipe B is significantly lower than all other recipes.

Two-Way ANOVA: Main Effects & Interactions

While One-Way ANOVA examines a single independent factor, industrial processes are usually driven by multiple variables acting simultaneously. Two-Way ANOVA evaluates two independent factors (Factor A A A and Factor B B B ) on a continuous response Y Y Y , testing three distinct hypotheses:

Main Effect of Factor A A A : Does Factor A A A significantly alter Y Y Y , averaged across all levels of Factor B B B ?

Main Effect of Factor B B B : Does Factor B B B significantly alter Y Y Y , averaged across all levels of Factor A A A ?

Interaction Effect ( A × B A \times B A × B ): Does the effect of Factor A A A depend upon the specific level of Factor B B B ?

Interpreting Interaction Plots

Interaction plots graph the response mean on the vertical axis against Factor A A A on the horizontal axis, with separate lines for each level of Factor B B B : Interpreting Interaction Plots

  • NO INTERACTION (Parallel Lines) STRONG INTERACTION (Crossing Lines)
  • Y ┌ B2 (High) Y ┌ B1 (High) B2
  • │ ─────── │ ╲ ╱
  • │ ╱ │ ╲ ╱
  • │ ╱ │ ╲ ╱
  • │ ╱ B1 (Low) │ ╲ ╱
  • │ ─────── │ ╳
  • │ ╱ │ ╱ ╲
  • │ ╱ │ ╱ ╲
  • └──────────────────────── └─────────────────────────────
  • A1 (Low) A2 (High) A1 (Low) A2 (High)

Effect of A is same across B Effect of A REVERSES across B!

Parallel Lines : Zero interaction. The effect of Factor A A A is consistent regardless of Factor B B B . The factors act independently.

Non-Parallel or Crossing Lines : Strong interaction present. The effect of Factor A A A depends entirely on what level Factor B B B is set to. When an interaction is significant, main effects cannot be interpreted in isolation.

Chi-Square ( χ 2 \chi^2 χ 2 ) Tests for Categorical Data

When data is categorical or discrete attribute counts (e.g., defect type, region, supplier, pass/fail), ANOVA cannot be applied. Practitioners deploy Chi-Square ( χ 2 \chi^2 χ 2 ) tests .

1. Chi-Square Test of Independence (Contingency Tables)

Evaluates whether two categorical variables are statistically independent or whether an association exists between them (e.g., Defect Category vs. Manufacturing Shift).

Data is tabulated in an r × c r \times c r × c contingency table ( r r r rows, c c c columns): χ calc 2 = ∑ i = 1 r ∑ j = 1 c ( O i j − E i j ) 2 E i j \chi^2_{\text{calc}} = \sum_{i=1}^r \sum_{j=1}^c \frac{(O_{ij} - E_{ij})^2}{E_{ij}} χ calc 2 ​ = i = 1 ∑ r ​ j = 1 ∑ c ​ E ij ​ ( O ij ​ − E ij ​ ) 2 ​

Where:

$O_{ij} = $ Observed frequency count in row i i i , column j j j

$E_{ij} = $ Expected frequency count under the null hypothesis of independence: E i j = ( Row i Total ) × ( Column j Total ) Grand Total N E_{ij} = \frac{(\text{Row } i \text{ Total}) \times (\text{Column } j \text{ Total})}{\text{Grand Total } N} E ij ​ = Grand Total N ( Row i Total ) × ( Column j Total ) ​

D e g r e e s o f F r e e d o m : d f = ( r − 1 ) ( c − 1 ) \mathbf{Degrees\ of\ Freedom:}\ df = (r - 1)(c - 1) Degrees of Freedom : df = ( r − 1 ) ( c − 1 )

The Expected Cell Frequency Rule: The asymptotic Chi-Square distribution is valid only if all expected cell frequencies satisfy E i j ≥ 5 E_{ij} \ge 5 E ij ​ ≥ 5 . If any expected cell count is < 5 < 5 < 5 , categories must be combined, or Fisher's Exact Test must be used for 2 × 2 2 \times 2 2 × 2 tables.

Step-by-Step Worked Example: Chi-Square Contingency Test

Scenario: A Green Belt investigates whether defect classification (Scratch vs Dent) depends on production shift (Day vs Night). A sample of N = 200 N = 200 N = 200 defective brackets yields the following observed counts ( O O O ):

  • Shift Scratch Dent Row Total
  • Day Shift 60 40 100
  • Night Shift 30 70 100
  • Column Total 90 110 Grand Total N = 200 N = 200 N = 200

Test for independence at α = 0.05 \alpha = 0.05 α = 0.05 .

Step 1: Compute Expected Frequencies ( E i j E_{ij} E ij ​ )

E Day, Scratch = 100 × 90 200 = 45.0 E_{\text{Day, Scratch}} = \frac{100 \times 90}{200} = 45.0 E Day, Scratch ​ = 200 100 × 90 ​ = 45.0

E Day, Dent = 100 × 110 200 = 55.0 E_{\text{Day, Dent}} = \frac{100 \times 110}{200} = 55.0 E Day, Dent ​ = 200 100 × 110 ​ = 55.0

E Night, Scratch = 100 × 90 200 = 45.0 E_{\text{Night, Scratch}} = \frac{100 \times 90}{200} = 45.0 E Night, Scratch ​ = 200 100 × 90 ​ = 45.0

E Night, Dent = 100 × 110 200 = 55.0 E_{\text{Night, Dent}} = \frac{100 \times 110}{200} = 55.0 E Night, Dent ​ = 200 100 × 110 ​ = 55.0 (All E i j ≥ 5 E_{ij} \ge 5 E ij ​ ≥ 5 , satisfying the cell frequency rule).

Step 2: Calculate Chi-Square Statistic χ calc 2 = ( 60 − 45 ) 2 45 + ( 40 − 55 ) 2 55 + ( 30 − 45 ) 2 45 + ( 70 − 55 ) 2 55 \chi^2_{\text{calc}} = \frac{(60 - 45)^2}{45} + \frac{(40 - 55)^2}{55} + \frac{(30 - 45)^2}{45} + \frac{(70 - 55)^2}{55} χ calc 2 ​ = 45 ( 60 − 45 ) 2 ​ + 55 ( 40 − 55 ) 2 ​ + 45 ( 30 − 45 ) 2 ​ + 55 ( 70 − 55 ) 2 ​ χ calc 2 = 225 45 + 225 55 + 225 45 + 225 55 = 5.000 + 4.091 + 5.000 + 4.091 = 18.182 \chi^2_{\text{calc}} = \frac{225}{45} + \frac{225}{55} + \frac{225}{45} + \frac{225}{55} = 5.000 + 4.091 + 5.000 + 4.091 = 18.182 χ calc 2 ​ = 45 225 ​ + 55 225 ​ + 45 225 ​ + 55 225 ​ = 5.000 + 4.091 + 5.000 + 4.091 = 18.182

Step 3: Determine Degrees of Freedom and Decision

d f = ( r − 1 ) ( c − 1 ) = ( 2 − 1 ) ( 2 − 1 ) = 1 × 1 = 1 df = (r - 1)(c - 1) = (2 - 1)(2 - 1) = 1 \times 1 = 1 df = ( r − 1 ) ( c − 1 ) = ( 2 − 1 ) ( 2 − 1 ) = 1 × 1 = 1

Critical value at α = 0.05 \alpha = 0.05 α = 0.05 with d f = 1 df = 1 df = 1 : χ crit 2 = 3.841 \chi^2_{\text{crit}} = 3.841 χ crit 2 ​ = 3.841

Since χ calc 2 = 18.182 > 3.841 \chi^2_{\text{calc}} = 18.182 > 3.841 χ calc 2 ​ = 18.182 > 3.841 ( p < 0.0001 p < 0.0001 p < 0.0001 ), Reject H 0 H_0 H 0 ​ .

Conclusion: Defect type is strongly associated with shift; Night Shift produces significantly more dents, while Day Shift produces significantly more scratches.

Master Inferential Hypothesis Testing Roadmap

Use this comprehensive master roadmap to select the exact hypothesis test required for any Six Sigma examination scenario:

Operational Question / Objective Input ( X X X ) Output ( Y Y Y ) Distribution / Conditions Recommended Statistical Test

Compare 1 mean to benchmark Constant 1 Continuous Normal ( σ \sigma σ known) One-Sample Z Z Z -test

Compare 1 mean to specification Constant 1 Continuous Normal ( σ \sigma σ unknown) One-Sample t t t -test ( d f = n − 1 df = n-1 df = n − 1 )

Compare 1 median to target Constant 1 Continuous Non-normal / Skewed 1-Sample Wilcoxon / Sign Test

Compare 2 independent means 1 Discrete (2 groups) 1 Continuous Normal (Equal variances) Two-Sample Pooled t t t -test

Compare 2 independent means 1 Discrete (2 groups) 1 Continuous Normal (Unequal variances) Welch's Unpooled t t t -test

Compare 2 paired/matched means 1 Discrete (2 conditions) 1 Continuous Normal differences Paired t t t -test ( d f = n pairs − 1 df = n_{\text{pairs}}-1 df = n pairs ​ − 1 )

Compare 2 independent medians 1 Discrete (2 groups) 1 Continuous Non-normal Mann-Whitney U Test

Compare 2 paired medians 1 Discrete (2 conditions) 1 Continuous Non-normal differences Wilcoxon Signed-Rank Test

Compare ≥ 3 \ge 3 ≥ 3 group means 1 Discrete ( k ≥ 3 k \ge 3 k ≥ 3 groups) 1 Continuous Normal, equal variance One-Way ANOVA ( F F F -test)

Compare ≥ 3 \ge 3 ≥ 3 groups with 2 factors 2 Discrete factors 1 Continuous Normal Two-Way ANOVA (Main + Interaction)

Compare ≥ 3 \ge 3 ≥ 3 group medians 1 Discrete ( k ≥ 3 k \ge 3 k ≥ 3 groups) 1 Continuous Non-normal Kruskal-Wallis / Mood's Median

Compare 1 proportion to target Constant 1 Discrete (Binary) n p 0 ≥ 5 , n ( 1 − p 0 ) ≥ 5 np_0 \ge 5, n(1-p_0) \ge 5 n p 0 ​ ≥ 5 , n ( 1 − p 0 ​ ) ≥ 5 1-Proportion Z Z Z -test

Compare 2 independent proportions 1 Discrete (2 groups) 1 Discrete (Binary) Binomial 2-Proportion Z Z Z -test (Pooled p ˉ \bar{p} p ˉ ​ )

Test association between 2 categories 1 Discrete ( r r r levels) 1 Discrete ( c c c levels) Expected counts ≥ 5 \ge 5 ≥ 5 Chi-Square ( χ 2 \chi^2 χ 2 ) Independence

Compare observed vs expected counts 1 Discrete ( k k k categories) Count frequency Expected counts ≥ 5 \ge 5 ≥ 5 Chi-Square Goodness-of-Fit

Critical CSSC Exam Traps

Trap 1: Running Multiple Pairwise t t t -Tests Instead of ANOVA — This severely inflates the family-wise Type I error rate ( α \alpha α -inflation), creating massive false alarm risks.

Trap 2: Believing H a H_a H a ​ in ANOVA Means "All Means Differ" — The alternative hypothesis in ANOVA only requires that at least one group mean differs from the others.

Trap 3: Calculating Degrees of Freedom for Contingency Tables as N − 1 N - 1 N − 1 — For a Chi-Square test of independence on an r × c r \times c r × c table, degrees of freedom is strictly d f = ( r − 1 ) ( c − 1 ) df = (r - 1)(c - 1) df = ( r − 1 ) ( c − 1 ) .

Trap 4: Ignoring the Expected Cell Frequency Requirement — Chi-Square tests are invalid if any expected cell count is < 5 < 5 < 5 . Always verify E i j = ( Row Total × Column Total ) / N ≥ 5 E_{ij} = (\text{Row Total} \times \text{Column Total}) / N \ge 5 E ij ​ = ( Row Total × Column Total ) / N ≥ 5 .

Trap 5: Interpreting Main Effects When Interaction is Significant — In Two-Way ANOVA, if the interaction p p p -value ≤ 0.05 \le 0.05 ≤ 0.05 , you must examine the interaction plot before drawing conclusions about main factors.

Loading diagram... ANOVA and Categorical Chi-Square Decision Architecture