Study guide

7.2 Hypothesis Testing: Means, Proportions, and Variances

FE Industrial and Systems Study Guide Exam

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Exam Overview, CBT Strategy & Handbook Navigation

1.1 FE Industrial Exam Structure, CBT Logistics, and Pacing 1.2 Using the NCEES FE Reference Handbook and Approved Calculators

2 Chapter 2: Mathematics for Industrial Engineers

2.1 Analytic Geometry, Coordinate Systems, and Trigonometry 2.2 Calculus: Differential, Integral, and Series Approximations 2.3 Linear Algebra: Matrix Operations, Systems of Equations, and Vectors

3 Chapter 3: Engineering Sciences Core

3.1 Thermodynamics Fundamentals and Fluid Flow Principles 3.2 Statics, Dynamics, and Mechanics of Materials Core 3.3 Electrical Circuits, DC/AC Analysis, and Power in Industrial Systems

4 Chapter 4: Ethics and Professional Practice

4.1 NCEES Model Rules of Professional Conduct and Technical Society Ethics 4.2 Engineering Licensure, Public Welfare, and Professional Liability 4.3 Engineering Contracts, Intellectual Property, and Conflicts of Interest

5 Chapter 5: Engineering Economics

5.1 Time Value of Money, Discount Factors, and Equivalence 5.2 Evaluating Alternatives: Present Worth, Equivalent Annual Cost, and IRR 5.3 Incremental Rate of Return, Replacement Analysis, and Break-Even 5.4 Depreciation Methods, MACRS, Income Taxes, and Inflation

6 Chapter 6: Probability Foundations & Distributions

6.1 Basic Probability Concepts, Conditional Probability, and Bayes' Theorem 6.2 Discrete Probability Distributions 6.3 Continuous Probability Distributions

7 Chapter 7: Statistical Inference, Regression & Design of Experiments

7.1 Sampling Distributions, Central Limit Theorem, and Confidence Intervals 7.2 Hypothesis Testing: Means, Proportions, and Variances 7.3 Linear Regression, Correlation Analysis, and Design of Experiments (ANOVA)

8 Chapter 8: Data, Logic, Modeling & Linear Programming

8.1 Optimization Modeling: Decision Variables, Objectives, and Constraints 8.2 Linear Programming: Graphical Solutions and Simplex Fundamentals 8.3 Duality, Sensitivity Analysis, and Integer Linear Programming 8.4 Data, Logic Development, and Analytics

9 Chapter 9: Stochastic Models & Simulation

9.1 Queuing Theory: Little's Law, M/M/1, and Multi-Server Systems 9.2 Discrete-Time Markov Chains: Transition Matrices and Steady-State Probabilities 9.3 Discrete-Event Simulation Principles and Monte Carlo Modeling

10 Chapter 10: Engineering Management, Project Planning & Performance Measurement

10.1 Organizational Structure, Leadership, and Worker Motivation Theories 10.2 Project Management: WBS, Critical Path Method (CPM), PERT, and Crashing 10.3 Earned Value Management (EVM), Decision Trees, and Risk Preference 10.4 Performance Measurement: Productivity, KPIs, Wage Systems, and the Balanced Scorecard

11 Chapter 11: Manufacturing & Production Systems

11.1 Manufacturing Processes, Material Processing, and Tolerancing 11.2 Assembly Line Balancing, Takt Time, Cycle Time, and Throughput 11.3 Sustainable Manufacturing, Waste Minimization, and Life-Cycle Assessment

12 Chapter 12: Production Planning, Inventory Control & Sequencing

12.1 Demand Forecasting Methods and Forecast Error Tracking 12.2 Aggregate Planning, Master Production Scheduling (MPS), and MRP 12.3 Deterministic Inventory Models: EOQ, EPQ, and Safety Stock 12.4 Job Sequencing, Dispatching Rules, and Flow Shop Scheduling

13 Chapter 13: Lean Systems & Theory of Constraints

13.1 Lean Foundations: Value Stream Mapping, Seven Wastes, and 5S 13.2 Pull Systems, Kanban, Just-In-Time (JIT), and SMED Setup Reduction 13.3 Theory of Constraints: Bottleneck Identification and Drum-Buffer-Rope

14 Chapter 14: Facilities Planning & Material Handling

14.1 Facility Location Analysis: Qualitative Factor Rating and Centroid Method 14.2 Facility Layout Design and Flow Analysis 14.3 Material Handling Equipment, Warehousing, and AS/RS

15 Chapter 15: Supply Chain Design & Logistics

15.1 Supply Chain Network Architecture and Transportation Modes 15.2 Distribution Strategies, Cross-Docking, and Vehicle Routing 15.3 The Bullwhip Effect, Risk Pooling, and Supply Chain Coordination

16 Chapter 16: Human Factors & Cognitive Ergonomics

16.1 Cognitive Engineering, Human-Machine Interface, Displays, and Controls 16.2 Anthropometric Principles, Percentile Design, and Biomechanical Posture 16.3 Environmental Ergonomics: Industrial Noise, Lighting, and Heat Stress

17 Chapter 17: Physical Ergonomics & Workplace Safety

17.1 Physical Ergonomics: The NIOSH Manual Lifting Equation 17.2 Cumulative Trauma Disorders and Postural Assessment 17.3 Workplace Safety: OSHA Standards, Hazard Analysis, and Control Hierarchy

18 Chapter 18: Work Design & Measurement

18.1 Methods Engineering, Process Charting, and Principles of Motion Economy 18.2 Direct Stopwatch Time Study: Observed Time, Performance Rating, and Allowances 18.3 Predetermined Time Systems, Work Sampling, and Learning Curves

19 Chapter 19: Quality Management, Six Sigma & Analytical Tools

19.1 Quality Philosophies, Total Quality Management, and Cost of Quality 19.2 Six Sigma DMAIC Methodology and Quality Function Deployment 19.3 The Seven Basic Quality Tools, Root Cause Analysis, and Taguchi Loss Function

20 Chapter 20: Statistical Process Control

20.1 Statistical Process Control Principles, Variation, and Run Rules 20.2 Variable Control Charts: X-bar & R, X-bar & S, and Individual-Moving Range 20.3 Attribute Control Charts: Defectives (p, np) and Defects (c, u)

21 Chapter 21: Process Capability & Acceptance Sampling

21.1 Process Capability Indices and Defect Levels 21.2 Acceptance Sampling Concepts and Operating Characteristic Curves 21.3 Sampling Risks, AQL, LTPD, and Standard Sampling Plans

22 Chapter 22: Systems Engineering & Life-Cycle Management

22.1 The Systems Engineering V-Model, Stakeholder Requirements, and Functional Architecture 22.2 Configuration Management, Interface Control, and Verification & Validation 22.3 Life-Cycle Costing, Maintainability, and System Availability

23 Chapter 23: Reliability Engineering & Risk Analysis

23.1 Reliability Mathematics, Hazard Rates, and the Bathtub Curve 23.2 System Reliability Configurations: Series, Parallel, and k-out-of-n Systems 23.3 Failure Modes and Effects Analysis and Fault Tree Analysis

7.2 Hypothesis Testing: Means, Proportions, and Variances

Key Takeaways Explain This Topic AI

Statistical hypothesis testing contrasts a null hypothesis H₀ (status quo, equality: =, ≤, ≥) against an alternative hypothesis H₁ (claim, difference: ≠, >, <); equality must strictly reside in H₀.

Type I error (α, significance level, producer's risk) is rejecting a true null hypothesis; Type II error (β, consumer's risk) is failing to reject a false null hypothesis; statistical power is 1 - β.

The p-value decision rule is universal across all tests: reject H₀ if p-value ≤ α; fail to reject H₀ if p-value > α.

When conducting a one-sample Z-test for a proportion, the standard error in the test statistic denominator must be calculated using the hypothesized null proportion p₀, not the observed sample proportion p̂.

Testing equality of two variances employs Snedecor's F-test (F₀ = s₁² / s₂² with ν₁ = n₁ - 1 and ν₂ = n₂ - 1); placing the larger sample variance in the numerator simplifies analysis to an upper-tail critical comparison.

Last updated: September 2026

Hypothesis testing is the formalized mathematical procedure by which industrial and systems engineers make objective, data-driven decisions regarding processes, machinery, materials, and human workflows. Rather than relying on subjective intuition or raw numerical differences that could simply be artifacts of random sampling variability, hypothesis testing quantifies the probability that an observed effect is genuine. The NCEES FE Reference Handbook provides the complete inventory of test statistics under the Engineering Probability and Statistics section. Understanding how to set up hypotheses, avoid decision errors, calculate test statistics, and interpret p-values is essential for the FE examination.

  • 1. The Hypothesis Testing Framework
  • Null ( H 0 H_0 H 0 ​ ) and Alternative ( H 1 H_1 H 1 ​ or H a H_a H a ​ ) Hypotheses

Every statistical test establishes two mutually exclusive and exhaustive propositions:

Null Hypothesis ( H 0 H_0 H 0 ​ ) : Represents the baseline status quo, default assumption, historical standard, or claim of no effect or no difference. The null hypothesis must always contain the condition of equality ( = = = , ≤ \le ≤ , or ≥ \ge ≥ ).

Alternative Hypothesis ( H 1 H_1 H 1 ​ ) : Represents the experimental claim, departure from standard, new operating condition, or research hypothesis to be proven ( < < < , > > > , or ≠ \neq  = ).

Directionality: Two-Tailed vs. One-Tailed Tests

Test Type Alternative Hypothesis ( H 1 H_1 H 1 ​ ) Rejection Region Critical Values

Two-Tailed μ ≠ μ 0 \mu \neq \mu_0 μ  = μ 0 ​ Both tails (split α / 2 \alpha/2 α /2 in each tail) ± Z α / 2 \pm Z_{\alpha/2} ± Z α /2 ​ or ± t α / 2 , ν \pm t_{\alpha/2, \nu} ± t α /2 , ν ​

Upper One-Tailed (Right) μ > μ 0 \mu > \mu_0 μ > μ 0 ​ Upper tail (entire α \alpha α in right tail) + Z α +Z_\alpha + Z α ​ or + t α , ν +t_{\alpha, \nu} + t α , ν ​

Lower One-Tailed (Left) μ < μ 0 \mu < \mu_0 μ < μ 0 ​ Lower tail (entire α \alpha α in left tail) − Z α -Z_\alpha − Z α ​ or − t α , ν -t_{\alpha, \nu} − t α , ν ​ Two-Tailed (α/2 in each tail) One-Tailed Upper (α in right tail)

  • ┌───────┴───────┐ ┌───┴───┐
  • ┌─┐ ┌─┐ │ ┌─┐
  • ───┴─┴─────────────┴─┴─── ───┴───────┴─┴───
  • Z_crit +Z_crit +Z_crit
  • 2. Decision Errors, Risks, and Test Power

Because sample statistics are random variables, hypothesis testing inherently involves uncertainty. The two potential incorrect decisions define Type I and Type II errors.

Reality \ Decision Fail to Reject H 0 H_0 H 0 ​ Reject H 0 H_0 H 0 ​

H 0 H_0 H 0 ​ is True Correct Decision ; Confidence Level = 1 − α = 1 - \alpha = 1 − α Type I Error ( α \alpha α ) ; Significance Level / Producer's Risk

H 0 H_0 H 0 ​ is False Type II Error ( β \beta β ) ; Consumer's Risk Correct Decision ; Statistical Power = 1 − β = 1 - \beta = 1 − β

Industrial Engineering Analogs: Producer's vs. Consumer's Risk

Type I Error ( α \alpha α , Producer's Risk) : Rejecting a good batch of parts or stopping a production process that is actually functioning within specification. The manufacturer ("producer") incurs the unnecessary cost of scrap, sorting, or downtime.

Type II Error ( β \beta β , Consumer's Risk) : Failing to reject a defective batch of parts or allowing an out-of-control machine to continue operating. The customer ("consumer") receives non-conforming product, leading to field failures, warranty claims, or safety hazards.

Statistical Power ( 1 − β 1 - \beta 1 − β )

Power is the probability of correctly rejecting a false null hypothesis: Power = P ( Reject H 0 ∣ H 0 is False ) = 1 − β \text{Power} = P(\text{Reject } H_0 \mid H_0 \text{ is False}) = 1 - \beta Power = P ( Reject H 0 ​ ∣ H 0 ​ is False ) = 1 − β . Power can be increased by:

Increasing sample size n n n : Reduces standard error, narrowing distributions and decreasing overlap.

Increasing significance level α \alpha α : Moving the critical threshold inwards decreases β \beta β (at the cost of increasing Type I risk).

Larger effect size ( δ = ∣ μ − μ 0 ∣ \delta = |\mu - \mu_0| δ = ∣ μ − μ 0 ​ ∣ ) : Substantial real differences are easier to detect.

Decreasing process variance ( σ 2 \sigma^2 σ 2 ) : Improving process capability tightens the distributions.

  • 3. Decision Rules: Critical Value Approach vs. p-Value Approach
  • The Critical Value Method

Compute the test statistic from the sample data ( Z 0 , t 0 , F 0 , χ 0 2 Z_0, t_0, F_0, \chi_0^2 Z 0 ​ , t 0 ​ , F 0 ​ , χ 0 2 ​ ). If the test statistic falls into the designated rejection region defined by the critical value at significance level α \alpha α , reject H 0 H_0 H 0 ​ ; otherwise, fail to reject H 0 H_0 H 0 ​ .

The p-Value Method

The p p p -value is the probability, assuming H 0 H_0 H 0 ​ is true, of obtaining a test statistic at least as extreme as the observed value. Golden Decision Rule: { p -value ≤ α ⟹ Reject H 0 (Statistically Significant) p -value > α ⟹ Fail to Reject H 0 \text{Golden Decision Rule:} \quad \begin{cases} p\text{-value} \le \alpha & \implies \text{Reject } H_0 \text{ (Statistically Significant)} \\ p\text{-value} > \alpha & \implies \text{Fail to Reject } H_0 \end{cases} Golden Decision Rule: { p -value ≤ α p -value > α ​ ⟹ Reject H 0 ​ (Statistically Significant) ⟹ Fail to Reject H 0 ​ ​

For a right-tailed Z Z Z -test: p -value = P ( Z ≥ Z 0 ) p\text{-value} = P(Z \ge Z_0) p -value = P ( Z ≥ Z 0 ​ ) .

For a left-tailed Z Z Z -test: p -value = P ( Z ≤ Z 0 ) p\text{-value} = P(Z \le Z_0) p -value = P ( Z ≤ Z 0 ​ ) .

For a two-tailed Z Z Z -test: p -value = 2 × P ( Z ≥ ∣ Z 0 ∣ ) p\text{-value} = 2 \times P(Z \ge |Z_0|) p -value = 2 × P ( Z ≥ ∣ Z 0 ​ ∣ ) .

  • 4. Test Statistics for a Single Population Mean (μ)
  • Case 1: Variance Known (Z-Test)

When σ \sigma σ is known and either the population is normal or n ≥ 30 n \ge 30 n ≥ 30 : Z 0 = x ˉ − μ 0 σ / n Z_0 = \frac{\bar{x} - \mu_0}{\sigma / \sqrt{n}} Z 0 ​ = σ / n ​ x ˉ − μ 0 ​ ​

Case 2: Variance Unknown (One-Sample t-Test)

When σ \sigma σ is unknown, estimate with s s s : t 0 = x ˉ − μ 0 s / n t_0 = \frac{\bar{x} - \mu_0}{s / \sqrt{n}} t 0 ​ = s / n ​ x ˉ − μ 0 ​ ​

with degrees of freedom ν = n − 1 \nu = n - 1 ν = n − 1 . Compare t 0 t_0 t 0 ​ against critical values t α , ν t_{\alpha, \nu} t α , ν ​ or t α / 2 , ν t_{\alpha/2, \nu} t α /2 , ν ​ .

  • 5. Two-Sample Tests for Comparing Population Means
  • Independent Samples with Equal Variances (Pooled t-Test)

Testing H 0 : μ 1 − μ 2 = Δ 0 H_0: \mu_1 - \mu_2 = \Delta_0 H 0 ​ : μ 1 ​ − μ 2 ​ = Δ 0 ​ (usually Δ 0 = 0 \Delta_0 = 0 Δ 0 ​ = 0 ) under the assumption σ 1 2 = σ 2 2 \sigma_1^2 = \sigma_2^2 σ 1 2 ​ = σ 2 2 ​ : t 0 = ( x ˉ 1 − x ˉ 2 ) − Δ 0 s p 1 n 1 + 1 n 2 , where s p = ( n 1 − 1 ) s 1 2 + ( n 2 − 1 ) s 2 2 n 1 + n 2 − 2 t_0 = \frac{(\bar{x}_1 - \bar{x}_2) - \Delta_0}{s_p \sqrt{\frac{1}{n_1} + \frac{1}{n_2}}}, \quad \text{where } s_p = \sqrt{\frac{(n_1 - 1)s_1^2 + (n_2 - 1)s_2^2}{n_1 + n_2 - 2}} t 0 ​ = s p ​ n 1 ​ 1 ​ + n 2 ​ 1 ​ ​ ( x ˉ 1 ​ − x ˉ 2 ​ ) − Δ 0 ​ ​ , where s p ​ = n 1 ​ + n 2 ​ − 2 ( n 1 ​ − 1 ) s 1 2 ​ + ( n 2 ​ − 1 ) s 2 2 ​ ​ ​

Degrees of freedom: ν = n 1 + n 2 − 2 \nu = n_1 + n_2 - 2 ν = n 1 ​ + n 2 ​ − 2 .

Independent Samples with Unequal Variances (Welch's t-Test)

When σ 1 2 ≠ σ 2 2 \sigma_1^2 \neq \sigma_2^2 σ 1 2 ​  = σ 2 2 ​ : t 0 = ( x ˉ 1 − x ˉ 2 ) − Δ 0 s 1 2 n 1 + s 2 2 n 2 t_0 = \frac{(\bar{x}_1 - \bar{x}_2) - \Delta_0}{\sqrt{\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}}} t 0 ​ = n 1 ​ s 1 2 ​ ​ + n 2 ​ s 2 2 ​ ​ ​ ( x ˉ 1 ​ − x ˉ 2 ​ ) − Δ 0 ​ ​

with Satterthwaite degrees of freedom ν \nu ν .

Paired Samples (Paired t-Test)

For paired observations with differences d i = x 1 i − x 2 i d_i = x_{1i} - x_{2i} d i ​ = x 1 i ​ − x 2 i ​ , testing H 0 : μ d = Δ d 0 H_0: \mu_d = \Delta_{d0} H 0 ​ : μ d ​ = Δ d 0 ​ : t 0 = d ˉ − Δ d 0 s d / n t_0 = \frac{\bar{d} - \Delta_{d0}}{s_d / \sqrt{n}} t 0 ​ = s d ​ / n ​ d ˉ − Δ d 0 ​ ​

with degrees of freedom ν = n − 1 \nu = n - 1 ν = n − 1 , where d ˉ \bar{d} d ˉ is the sample mean of differences and s d s_d s d ​ is the sample standard deviation of differences.

  • 6. Hypothesis Testing for Proportions
  • Single-Sample Proportion Z-Test

Testing H 0 : p = p 0 H_0: p = p_0 H 0 ​ : p = p 0 ​ versus H 1 : p ≠ p 0 H_1: p \neq p_0 H 1 ​ : p  = p 0 ​ (or p > p 0 p > p_0 p > p 0 ​ , p < p 0 p < p_0 p < p 0 ​ ): Z 0 = p ^ − p 0 p 0 ( 1 − p 0 ) n Z_0 = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1 - p_0)}{n}}} Z 0 ​ = n p 0 ​ ( 1 − p 0 ​ ) ​ ​ p ^ ​ − p 0 ​ ​

Exam Watchout: Standard Error in Single Proportion Tests Under the null hypothesis, the true proportion is assumed to equal p 0 p_0 p 0 ​ . Therefore, the standard error in the denominator must be evaluated using p 0 p_0 p 0 ​ , not p ^ \hat{p} p ^ ​ . In contrast, confidence intervals use p ^ \hat{p} p ^ ​ because no hypothesized value exists. Substituting p ^ \hat{p} p ^ ​ into the denominator of a hypothesis test is a classic distractor on the FE exam.

Two-Sample Proportion Z-Test

Testing H 0 : p 1 − p 2 = 0 H_0: p_1 - p_2 = 0 H 0 ​ : p 1 ​ − p 2 ​ = 0 against H 1 : p 1 ≠ p 2 H_1: p_1 \neq p_2 H 1 ​ : p 1 ​  = p 2 ​ . Because the null hypothesis posits that p 1 = p 2 = p p_1 = p_2 = p p 1 ​ = p 2 ​ = p , data from both samples are pooled to form the best single estimate of the common proportion p ˉ \bar{p} p ˉ ​ : p ˉ = x 1 + x 2 n 1 + n 2 \bar{p} = \frac{x_1 + x_2}{n_1 + n_2} p ˉ ​ = n 1 ​ + n 2 ​ x 1 ​ + x 2 ​ ​

The test statistic is: Z 0 = p ^ 1 − p ^ 2 p ˉ ( 1 − p ˉ ) ( 1 n 1 + 1 n 2 ) Z_0 = \frac{\hat{p}_1 - \hat{p}_2}{\sqrt{\bar{p}(1 - \bar{p})\left(\frac{1}{n_1} + \frac{1}{n_2}\right)}} Z 0 ​ = p ˉ ​ ( 1 − p ˉ ​ ) ( n 1 ​ 1 ​ + n 2 ​ 1 ​ ) ​ p ^ ​ 1 ​ − p ^ ​ 2 ​ ​

  • 7. Tests on Variances: Chi-Square and F-Tests
  • Single Population Variance (Chi-Square Test)

Testing H 0 : σ 2 = σ 0 2 H_0: \sigma^2 = \sigma_0^2 H 0 ​ : σ 2 = σ 0 2 ​ (or σ 2 ≤ σ 0 2 \sigma^2 \le \sigma_0^2 σ 2 ≤ σ 0 2 ​ , σ 2 ≥ σ 0 2 \sigma^2 \ge \sigma_0^2 σ 2 ≥ σ 0 2 ​ ): χ 0 2 = ( n − 1 ) s 2 σ 0 2 \chi_0^2 = \frac{(n - 1)s^2}{\sigma_0^2} χ 0 2 ​ = σ 0 2 ​ ( n − 1 ) s 2 ​

with degrees of freedom ν = n − 1 \nu = n - 1 ν = n − 1 . Compare χ 0 2 \chi_0^2 χ 0 2 ​ against critical values χ α , ν 2 \chi^2_{\alpha, \nu} χ α , ν 2 ​ or χ α / 2 , ν 2 \chi^2_{\alpha/2, \nu} χ α /2 , ν 2 ​ .

Two Population Variances (Snedecor's F-Test)

Testing equality of variances between two independent normal populations: H 0 : σ 1 2 = σ 2 2 H_0: \sigma_1^2 = \sigma_2^2 H 0 ​ : σ 1 2 ​ = σ 2 2 ​ versus H 1 : σ 1 2 ≠ σ 2 2 H_1: \sigma_1^2 \neq \sigma_2^2 H 1 ​ : σ 1 2 ​  = σ 2 2 ​ (or σ 1 2 > σ 2 2 \sigma_1^2 > \sigma_2^2 σ 1 2 ​ > σ 2 2 ​ ).

The test statistic is the ratio of sample variances: F 0 = s 1 2 s 2 2 F_0 = \frac{s_1^2}{s_2^2} F 0 ​ = s 2 2 ​ s 1 2 ​ ​

with numerator degrees of freedom ν 1 = n 1 − 1 \nu_1 = n_1 - 1 ν 1 ​ = n 1 ​ − 1 and denominator degrees of freedom ν 2 = n 2 − 1 \nu_2 = n_2 - 1 ν 2 ​ = n 2 ​ − 1 .

Standard Calculation Convention : Always designate the sample with the larger sample variance as Sample 1, so that s 1 2 ≥ s 2 2 s_1^2 \ge s_2^2 s 1 2 ​ ≥ s 2 2 ​ and F 0 ≥ 1 F_0 \ge 1 F 0 ​ ≥ 1 . For a two-tailed test at significance level α \alpha α , compare F 0 F_0 F 0 ​ to the upper-tail critical value F α / 2 , ν 1 , ν 2 F_{\alpha/2, \, \nu_1, \, \nu_2} F α /2 , ν 1 ​ , ν 2 ​ ​ . If F 0 > F α / 2 , ν 1 , ν 2 F_0 > F_{\alpha/2, \, \nu_1, \, \nu_2} F 0 ​ > F α /2 , ν 1 ​ , ν 2 ​ ​ , reject H 0 H_0 H 0 ​ and conclude the variances are unequal.

8. Practical vs. Statistical Significance in Industrial Manufacturing

In high-volume manufacturing environments equipped with automated sensing (e.g., in-line vision inspection logging n = 100,000 n = 100{,}000 n = 100 , 000 parts per day), standard error σ / n \sigma / \sqrt{n} σ / n ​ approaches zero. As a result, even microscopic, trivial deviations—such as a shaft measuring 25.0002 mm 25.0002\text{ mm} 25.0002 mm versus a 25.0000 mm 25.0000\text{ mm} 25.0000 mm target—will yield an astronomical test statistic ( Z 0 > 10 Z_0 > 10 Z 0 ​ > 10 ) and a tiny p p p -value ( p < 0.00001 p < 0.00001 p < 0.00001 ).

Industrial engineers must distinguish between:

Statistical Significance : The observed difference is unlikely to be explained by chance alone ( p ≤ α p \le \alpha p ≤ α ).

Practical (Engineering) Significance : The observed difference is large enough to impact product function, assembly fit, customer satisfaction, or unit economics. A change of 0.0002 mm 0.0002\text{ mm} 0.0002 mm on a part with a ± 0.100 mm \pm 0.100\text{ mm} ± 0.100 mm tolerance band is statistically significant but practically meaningless.

  • 9. Step-by-Step Worked Engineering Examples
  • Worked Example 7.2.1: Single-Sample t-Test on Polymer Tensile Strength

Problem : An additive manufacturing facility prints tooling fixtures. The nominal tensile yield strength must meet or exceed μ 0 = 45.0 MPa \mu_0 = 45.0\text{ MPa} μ 0 ​ = 45.0 MPa . A random sample of n = 16 n = 16 n = 16 test dogbones yields a sample mean of x ˉ = 43.8 MPa \bar{x} = 43.8\text{ MPa} x ˉ = 43.8 MPa and a sample standard deviation of s = 2.0 MPa s = 2.0\text{ MPa} s = 2.0 MPa . Test at significance level α = 0.05 \alpha = 0.05 α = 0.05 whether the fixtures fail to meet specification.

Solution :

Formulate Hypotheses :

H 0 : μ ≥ 45.0 MPa H_0: \mu \ge 45.0\text{ MPa} H 0 ​ : μ ≥ 45.0 MPa (compliant tensile strength)

H 1 : μ < 45.0 MPa H_1: \mu < 45.0\text{ MPa} H 1 ​ : μ < 45.0 MPa (undersized / non-compliant tensile strength, lower one-tailed test)

Select Test Statistic and Identify Degrees of Freedom :

Since σ \sigma σ is unknown, use the one-sample t t t -test with ν = n − 1 = 16 − 1 = 15 \nu = n - 1 = 16 - 1 = 15 ν = n − 1 = 16 − 1 = 15 .

Compute the Test Statistic : t 0 = x ˉ − μ 0 s / n = 43.8 − 45.0 2.0 / 16 = − 1.2 2.0 / 4 = − 1.2 0.50 = − 2.40 t_0 = \frac{\bar{x} - \mu_0}{s / \sqrt{n}} = \frac{43.8 - 45.0}{2.0 / \sqrt{16}} = \frac{-1.2}{2.0 / 4} = \frac{-1.2}{0.50} = -2.40 t 0 ​ = s / n ​ x ˉ − μ 0 ​ ​ = 2.0/ 16 ​ 43.8 − 45.0 ​ = 2.0/4 − 1.2 ​ = 0.50 − 1.2 ​ = − 2.40

Determine Critical Value and Rejection Region :

Significance level: α = 0.05 \alpha = 0.05 α = 0.05 .

Degrees of freedom: ν = 15 \nu = 15 ν = 15 .

For a lower one-tailed test, the critical value is − t 0.05 , 15 = − 1.753 -t_{0.05, 15} = -1.753 − t 0.05 , 15 ​ = − 1.753 .

Rejection region: Reject H 0 H_0 H 0 ​ if t 0 < − 1.753 t_0 < -1.753 t 0 ​ < − 1.753 .

Statistical Decision and Engineering Interpretation :

Since t 0 = − 2.40 < − 1.753 t_0 = -2.40 < -1.753 t 0 ​ = − 2.40 < − 1.753 , we reject H 0 H_0 H 0 ​ at α = 0.05 \alpha = 0.05 α = 0.05 .

Conclusion: There is statistically significant evidence at the 5 % 5\% 5% level that the 3D-printed polymer fixtures fail to meet the required 45.0 MPa 45.0\text{ MPa} 45.0 MPa tensile specification. The production batch must be quarantined.

Worked Example 7.2.2: Two-Sample Proportion Test on Solder Bridging

Problem : A surface-mount electronics line experiences solder bridging defects. An engineer tests a new nitrogen-inerting reflow process against the baseline air process:

Baseline Air ( 1 1 1 ): x 1 = 36 x_1 = 36 x 1 ​ = 36 defectives out of n 1 = 400 n_1 = 400 n 1 ​ = 400 boards ( p ^ 1 = 0.090 \hat{p}_1 = 0.090 p ^ ​ 1 ​ = 0.090 ).

Nitrogen Inerting ( 2 2 2 ): x 2 = 18 x_2 = 18 x 2 ​ = 18 defectives out of n 2 = 400 n_2 = 400 n 2 ​ = 400 boards ( p ^ 2 = 0.045 \hat{p}_2 = 0.045 p ^ ​ 2 ​ = 0.045 ).

At α = 0.05 \alpha = 0.05 α = 0.05 , has the nitrogen process significantly reduced the defect rate?

Solution :

Formulate Hypotheses :

H 0 : p 1 − p 2 ≤ 0 H_0: p_1 - p_2 \le 0 H 0 ​ : p 1 ​ − p 2 ​ ≤ 0 (nitrogen does not reduce defects)

H 1 : p 1 − p 2 > 0 H_1: p_1 - p_2 > 0 H 1 ​ : p 1 ​ − p 2 ​ > 0 (nitrogen significantly reduces defects, right-tailed test)

Compute Pooled Proportion p ˉ \bar{p} p ˉ ​ : p ˉ = x 1 + x 2 n 1 + n 2 = 36 + 18 400 + 400 = 54 800 = 0.0675 \bar{p} = \frac{x_1 + x_2}{n_1 + n_2} = \frac{36 + 18}{400 + 400} = \frac{54}{800} = 0.0675 p ˉ ​ = n 1 ​ + n 2 ​ x 1 ​ + x 2 ​ ​ = 400 + 400 36 + 18 ​ = 800 54 ​ = 0.0675 1 − p ˉ = 1 − 0.0675 = 0.9325 1 - \bar{p} = 1 - 0.0675 = 0.9325 1 − p ˉ ​ = 1 − 0.0675 = 0.9325

Compute Pooled Standard Error : SE = p ˉ ( 1 − p ˉ ) ( 1 n 1 + 1 n 2 ) = 0.0675 × 0.9325 × ( 1 400 + 1 400 ) \text{SE} = \sqrt{\bar{p}(1 - \bar{p})\left(\frac{1}{n_1} + \frac{1}{n_2}\right)} = \sqrt{0.0675 \times 0.9325 \times \left(\frac{1}{400} + \frac{1}{400}\right)} SE = p ˉ ​ ( 1 − p ˉ ​ ) ( n 1 ​ 1 ​ + n 2 ​ 1 ​ ) ​ = 0.0675 × 0.9325 × ( 400 1 ​ + 400 1 ​ ) ​ SE = 0.062944 × 0.0050 = 0.00031472 ≈ 0.01774 \text{SE} = \sqrt{0.062944 \times 0.0050} = \sqrt{0.00031472} \approx 0.01774 SE = 0.062944 × 0.0050 ​ = 0.00031472 ​ ≈ 0.01774

Compute the Test Statistic Z 0 Z_0 Z 0 ​ : Z 0 = p ^ 1 − p ^ 2 SE = 0.090 − 0.045 0.01774 = 0.045 0.01774 ≈ + 2.537 Z_0 = \frac{\hat{p}_1 - \hat{p}_2}{\text{SE}} = \frac{0.090 - 0.045}{0.01774} = \frac{0.045}{0.01774} \approx +2.537 Z 0 ​ = SE p ^ ​ 1 ​ − p ^ ​ 2 ​ ​ = 0.01774 0.090 − 0.045 ​ = 0.01774 0.045 ​ ≈ + 2.537

Decision and Conclusion :

For an upper one-tailed test at α = 0.05 \alpha = 0.05 α = 0.05 , critical value Z 0.05 = 1.645 Z_{0.05} = 1.645 Z 0.05 ​ = 1.645 .

Because Z 0 = 2.537 > 1.645 Z_0 = 2.537 > 1.645 Z 0 ​ = 2.537 > 1.645 (corresponding p -value ≈ 0.0056 < 0.05 p\text{-value} \approx 0.0056 < 0.05 p -value ≈ 0.0056 < 0.05 ), we reject H 0 H_0 H 0 ​ .

Conclusion: The nitrogen reflow process significantly reduces the solder bridge defect rate from 9.0 % 9.0\% 9.0% to 4.5 % 4.5\% 4.5% .

10. NCEES Reference Handbook Tips & Realistic Exam Traps

Equality Belongs in the Null : On conceptual setup questions, look for the choice where the equals sign ( = , ≤ , ≥ =, \le, \ge = , ≤ , ≥ ) is in H 0 H_0 H 0 ​ . Any choice placing an inequality without an equals sign ( < , > , ≠ <, >, \neq < , > ,  = ) in H 0 H_0 H 0 ​ is automatically invalid.

Two-Tailed p-Value Doubling : When given a test statistic Z 0 = 2.10 Z_0 = 2.10 Z 0 ​ = 2.10 for a two-tailed test, finding P ( Z ≥ 2.10 ) = 0.0179 P(Z \ge 2.10) = 0.0179 P ( Z ≥ 2.10 ) = 0.0179 from standard normal tables is only half the answer. You must double it for a two-tailed test: p -value = 2 × 0.0179 = 0.0358 p\text{-value} = 2 \times 0.0179 = 0.0358 p -value = 2 × 0.0179 = 0.0358 .

F-Test Degree of Freedom Ordering : When looking up F α , ν 1 , ν 2 F_{\alpha, \nu_1, \nu_2} F α , ν 1 ​ , ν 2 ​ ​ , make sure ν 1 \nu_1 ν 1 ​ is the numerator degrees of freedom (columns in the NCEES handbook table) and ν 2 \nu_2 ν 2 ​ is the denominator degrees of freedom (rows). Reversing them results in a completely wrong critical value because the F F F -distribution is not symmetric.

Pooled Proportion in Two-Sample Z-Tests : Never use unpooled sample variances when testing H 0 : p 1 = p 2 H_0: p_1 = p_2 H 0 ​ : p 1 ​ = p 2 ​ . The standard error must use p ˉ = ( x 1 + x 2 ) / ( n 1 + n 2 ) \bar{p} = (x_1 + x_2)/(n_1 + n_2) p ˉ ​ = ( x 1 ​ + x 2 ​ ) / ( n 1 ​ + n 2 ​ ) .