Study guide

2.4 Hypothesis Testing, Goodness of Fit, and Linear Regression

FE Other Disciplines Fundamentals of Engineering

Track your progress as you read

Finished sections get a check mark.

1 Introduction & Exam Strategy

About the FE Other Disciplines Exam How to Use This Study Guide

2 Chapter 1: Mathematics

1.1 Analytic Geometry, Trigonometry, and Vector Operations 1.2 Differential and Integral Calculus 1.3 Ordinary Differential Equations and Laplace Transforms 1.4 Linear Algebra, Matrices, and Systems of Equations 1.5 Numerical Methods, Root-Finding, and Numerical Integration

3 Chapter 2: Probability and Statistics

2.1 Probability Distributions, PMFs, and PDFs 2.2 Expected Value, Mean, Variance, and Descriptive Statistics 2.3 Estimation, Confidence Intervals, and Sample Sizes 2.4 Hypothesis Testing, Goodness of Fit, and Linear Regression

4 Chapter 3: Chemistry

3.1 Stoichiometry, Chemical Reactions, and Solutions 3.2 Chemical Equilibrium, Acids/Bases, and Ideal Gas Laws 3.3 Oxidation-Reduction, Electrochemistry, and Corrosion Basics

5 Chapter 4: Instrumentation and Controls

4.1 Sensors, Transducers, and Signal Conditioning 4.2 Data Sampling, Accuracy/Precision, and A/D Conversion 4.3 Logic Diagrams, Boolean Algebra, and Control Logic

6 Chapter 5: Engineering Ethics and Societal Impacts

5.1 NCEES Model Rules of Professional Conduct and Codes of Ethics 5.2 Public Protection, Licensure Boards, and Professional Liability 5.3 Societal Impacts: Sustainability, Life-Cycle Analysis, and Public Safety

7 Chapter 6: Safety, Health, and Environment

6.1 OSHA Regulations, Industrial Hygiene, and PPE 6.2 Hazard Assessment, Risk Management, and Environmental Rules 6.3 Gas Detection and Monitoring 6.4 Electrical Safety 6.5 Confined Space Entry and Ventilation Rates

8 Chapter 7: Engineering Economics

7.1 Time Value of Money, Interest Formulas, and Cash Flow Diagrams 7.2 Capital Budgeting: Present Worth, Annual Worth, IRR, and B/C Ratio 7.3 Depreciation Methods, Income Taxes, and Breakeven Analysis 7.4 Cost Analysis, Uncertainty, Expected Value, and Decision Trees

9 Chapter 8: Statics

8.1 Statics Equilibrium, Free-Body Diagrams, and Force Systems 8.2 Distributed Loads, Resultants, and Equivalent Force-Couple Systems 8.3 Internal Forces in Rigid Bodies: Trusses, Frames, Machines, and Friction 8.4 Three-Dimensional Equilibrium, Cables, and Pulley Systems 8.5 Area Properties: Centroids, Moments of Inertia, and Radius of Gyration 8.6 Weight, Mass, and Force Units: slug, lbm, lbf, and g_c

10 Chapter 9: Dynamics

9.1 Kinematics of Particles and Rigid Bodies 9.2 Newton's Second Law for Particles and Rigid Bodies 9.3 Angular Motion, Torque, and Mass Moment of Inertia 9.4 Work-Energy Principle, Power, and Efficiency 9.5 Impulse-Momentum, Impact, and Coefficient of Restitution 9.6 Dynamic Friction and Vibrations: Natural Frequency and Damping

11 Chapter 10: Strength of Materials

10.1 Stress Types, Axial Stress-Strain, and Thermal Deformation 10.2 Torsion of Circular Shafts, Polar Moment, and Angle of Twist 10.3 Shear and Moment Diagrams 10.4 Flexural Bending Stress and Transverse Shear Stress 10.5 Stress Transformations, Principal Stresses, and Mohr's Circle 10.6 Beam Deflections and Combined Axial/Bending Loading 10.7 Column Buckling (Euler Formula) and Yield/Failure Criteria

12 Chapter 11: Materials

11.1 Crystal Structure, Phase Diagrams, and Material Processing 11.2 Mechanical Properties, Fatigue, Fracture, and Corrosion 11.3 Thermal Properties of Materials 11.4 Electrical and Chemical Properties of Materials 11.5 Material Selection for Engineering Design

13 Chapter 12: Fluid Mechanics

12.1 Fluid Properties, Hydrostatic Pressure, and Manometry 12.2 Dimensionless Numbers and Flow Regimes 12.3 Continuity Equation, Bernoulli Equation, and Energy Equation 12.4 Impulse-Momentum Equation and Forces on Control Volumes 12.5 Flow in Pipes, Friction Factors (Moody Diagram), and Head Losses 12.6 Turbomachinery: Pumps, Turbines, Fans, and Compressors 12.7 Open-Channel Flow, Manning's Equation, and Drag 12.8 Fluid Transport Systems and Flow Measurement 12.9 Ideal Gas Law, Gas Mixtures, and the Real Gas Law

14 Chapter 13: Basic Electrical Engineering

13.1 Electrical Fundamentals, Ohm's Law, and DC Circuits 13.2 AC Circuits, Phasors, Impedance, and Three-Phase Power 13.3 Magnetic Fields, Induction, Transformers, and Motor Efficiency 13.4 Electrical Measuring Devices: Voltmeter, Ammeter, and Wattmeter

15 Chapter 14: Thermodynamics and Heat Transfer

14.1 First & Second Laws of Thermodynamics and Pure Substance Properties 14.2 Thermodynamic Processes and Property Diagrams (T-s, P-h, P-v) 14.3 Heat Transfer: Conduction, Convection, and Radiation 14.4 Ideal Gas Processes and Power/Refrigeration Cycles 14.5 Mass and Energy Balances on Control Volumes 14.6 Combustion and Combustion Products 14.7 Psychrometrics: Relative Humidity, Wet-Bulb Temperature, and the Psychrometric Chart

2.4 Hypothesis Testing, Goodness of Fit, and Linear Regression

Key Takeaways Explain This Topic AI

A hypothesis test rejects the null hypothesis when the test statistic falls beyond the critical value, or equivalently when the p-value is below the significance level alpha.

A Type I error rejects a true null hypothesis and has probability alpha; a Type II error fails to reject a false null hypothesis and has probability beta.

The least-squares slope is Sxy/Sxx and the line always passes through the point (x-bar, y-bar).

The coefficient of determination R-squared is the fraction of variation in y explained by the model and equals the square of the correlation coefficient r in simple linear regression.

A high R-squared confirms goodness of fit, not causation, and says nothing about whether the chosen model form is physically correct.

Last updated: August 2026

2.4 Hypothesis Testing, Goodness of Fit, and Linear Regression

Hypothesis testing asks whether an observed difference is larger than sampling variation can plausibly explain, and regression asks how well a straight line describes a data set. Both are examined under the NCEES sub-topics sample distributions and sizes and goodness of fit , and both are formula-driven once the correct test or statistic is identified.

1. Hypothesis Testing Methodology

Hypothesis testing evaluates competing claims regarding a population parameter using sample evidence.

Step-by-Step Procedure

Formulate Hypotheses:

Null Hypothesis ( H 0 H_0 H 0 ​ ): Baseline status-quo claim (e.g., H 0 : μ = μ 0 H_0: \mu = \mu_0 H 0 ​ : μ = μ 0 ​ or H 0 : μ ≤ μ 0 H_0: \mu \le \mu_0 H 0 ​ : μ ≤ μ 0 ​ ).

Alternative Hypothesis ( H a H_a H a ​ ): Research hypothesis (e.g., H a : μ ≠ μ 0 H_a: \mu \ne \mu_0 H a ​ : μ  = μ 0 ​ , H a : μ > μ 0 H_a: \mu > \mu_0 H a ​ : μ > μ 0 ​ , or H a : μ < μ 0 H_a: \mu < \mu_0 H a ​ : μ < μ 0 ​ ).

Select Significance Level ( α \alpha α ): Common values are α = 0.05 \alpha = 0.05 α = 0.05 (5%) or α = 0.01 \alpha = 0.01 α = 0.01 (1%).

Compute Test Statistic:

Z Z Z -test (Known σ \sigma σ or n ≥ 30 n \ge 30 n ≥ 30 ): Z c a l c = x ˉ − μ 0 σ / n Z_{calc} = \frac{\bar{x} - \mu_0}{\sigma / \sqrt{n}} Z c a l c ​ = σ / n ​ x ˉ − μ 0 ​ ​

t t t -test (Unknown σ \sigma σ , n < 30 n < 30 n < 30 ): t c a l c = x ˉ − μ 0 s / n t_{calc} = \frac{\bar{x} - \mu_0}{s / \sqrt{n}} t c a l c ​ = s / n ​ x ˉ − μ 0 ​ ​

Determine Critical Region / p p p -value:

Critical Value Method: Compare test statistic to z c r i t z_{crit} z cr i t ​ or t c r i t t_{crit} t cr i t ​ .

p p p -value Method: Compute p p p -value = P ( observing test statistic ∣ H 0 is true ) = P(\text{observing test statistic} \mid H_0 \text{ is true}) = P ( observing test statistic ∣ H 0 ​ is true ) .

State Decision Rule:

If ∣ Z c a l c ∣ > Z c r i t |Z_{calc}| > Z_{crit} ∣ Z c a l c ​ ∣ > Z cr i t ​ (or p -value < α p\text{-value} < \alpha p -value < α ), Reject H 0 H_0 H 0 ​ .

If ∣ Z c a l c ∣ ≤ Z c r i t |Z_{calc}| \le Z_{crit} ∣ Z c a l c ​ ∣ ≤ Z cr i t ​ (or p -value ≥ α p\text{-value} \ge \alpha p -value ≥ α ), Fail to Reject H 0 H_0 H 0 ​ .

Decision Errors in Hypothesis Testing

  • Truth \ 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 ( 1 − α 1 - \alpha 1 − α ) Type I Error ( α \alpha α ) (False Positive)

H 0 H_0 H 0 ​ is False Type II Error ( β \beta β ) (False Negative) Correct Decision ( 1 − β 1 - \beta 1 − β ) ( Power )

Worked Engineering Example: One-Tailed Z Z Z -Test

Problem: An environmental compliance regulation limits lead content in discharge water to μ 0 ≤ 15.0 ppb \mu_0 \le 15.0\text{ ppb} μ 0 ​ ≤ 15.0 ppb . An inspector tests n = 36 n = 36 n = 36 samples, finding a sample mean x ˉ = 16.2 ppb \bar{x} = 16.2\text{ ppb} x ˉ = 16.2 ppb . The known population standard deviation is σ = 3.6 ppb \sigma = 3.6\text{ ppb} σ = 3.6 ppb . Test at α = 0.05 \alpha = 0.05 α = 0.05 significance level whether the plant violates effluent standards ( Z c r i t = 1.645 Z_{crit} = 1.645 Z cr i t ​ = 1.645 for upper one-tailed test).

Solution: Step 1: Set Hypotheses: H 0 : μ ≤ 15.0 ppb vs. H a : μ > 15.0 ppb H_0: \mu \le 15.0\text{ ppb} \quad \text{vs.} \quad H_a: \mu > 15.0\text{ ppb} H 0 ​ : μ ≤ 15.0 ppb vs. H a ​ : μ > 15.0 ppb

Step 2: Calculate Test Statistic Z c a l c Z_{calc} Z c a l c ​ : Z c a l c = x ˉ − μ 0 σ / n = 16.2 − 15.0 3.6 / 36 = 1.2 3.6 / 6 = 1.2 0.60 = 2.00 Z_{calc} = \frac{\bar{x} - \mu_0}{\sigma / \sqrt{n}} = \frac{16.2 - 15.0}{3.6 / \sqrt{36}} = \frac{1.2}{3.6 / 6} = \frac{1.2}{0.60} = 2.00 Z c a l c ​ = σ / n ​ x ˉ − μ 0 ​ ​ = 3.6/ 36 ​ 16.2 − 15.0 ​ = 3.6/6 1.2 ​ = 0.60 1.2 ​ = 2.00

Step 3: Compare to Critical Value: Since Z c a l c = 2.00 > Z c r i t = 1.645 Z_{calc} = 2.00 > Z_{crit} = 1.645 Z c a l c ​ = 2.00 > Z cr i t ​ = 1.645 , the calculated value lies in the critical rejection region.

Step 4: Conclusion: Reject H 0 H_0 H 0 ​ at the α = 0.05 \alpha = 0.05 α = 0.05 level ( p p p -value = 1 − Φ ( 2.00 ) = 0.0228 < 0.05 = 1 - \Phi(2.00) = 0.0228 < 0.05 = 1 − Φ ( 2.00 ) = 0.0228 < 0.05 ). There is sufficient evidence to conclude that the plant's mean discharge lead concentration significantly exceeds 15.0 ppb.

2. Simple Linear Regression and Correlation

Simple linear regression models the relationship between an independent predictor variable x x x and a dependent response variable y y y .

Least-Squares Regression Equations Model: y ^ = a + b x \text{Model: } \hat{y} = a + b x Model: y ^ ​ = a + b x

Where b b b is the slope and a a a is the y y y -intercept.

NCEES Formulas for Sum of Squares S x x = ∑ i = 1 n ( x i − x ˉ ) 2 = ∑ x i 2 − ( ∑ x i ) 2 n S_{xx} = \sum_{i=1}^n (x_i - \bar{x})^2 = \sum x_i^2 - \frac{(\sum x_i)^2}{n} S xx ​ = i = 1 ∑ n ​ ( x i ​ − x ˉ ) 2 = ∑ x i 2 ​ − n ( ∑ x i ​ ) 2 ​ S y y = ∑ i = 1 n ( y i − y ˉ ) 2 = ∑ y i 2 − ( ∑ y i ) 2 n S_{yy} = \sum_{i=1}^n (y_i - \bar{y})^2 = \sum y_i^2 - \frac{(\sum y_i)^2}{n} S yy ​ = i = 1 ∑ n ​ ( y i ​ − y ˉ ​ ) 2 = ∑ y i 2 ​ − n ( ∑ y i ​ ) 2 ​ S x y = ∑ i = 1 n ( x i − x ˉ ) ( y i − y ˉ ) = ∑ x i y i − ( ∑ x i ) ( ∑ y i ) n S_{xy} = \sum_{i=1}^n (x_i - \bar{x})(y_i - \bar{y}) = \sum x_i y_i - \frac{(\sum x_i)(\sum y_i)}{n} S x y ​ = i = 1 ∑ n ​ ( x i ​ − x ˉ ) ( y i ​ − y ˉ ​ ) = ∑ x i ​ y i ​ − n ( ∑ x i ​ ) ( ∑ y i ​ ) ​

Least-Squares Parameters Slope: b = S x y S x x \text{Slope: } b = \frac{S_{xy}}{S_{xx}} Slope: b = S xx ​ S x y ​ ​ Intercept: a = y ˉ − b x ˉ \text{Intercept: } a = \bar{y} - b \bar{x} Intercept: a = y ˉ ​ − b x ˉ

Sample Correlation Coefficient ( r r r ) and Coefficient of Determination ( R 2 R^2 R 2 )

Sample Correlation Coefficient ( r r r ): r = S x y S x x S y y , − 1 ≤ r ≤ 1 r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}}, \quad -1 \le r \le 1 r = S xx ​ S yy ​ ​ S x y ​ ​ , − 1 ≤ r ≤ 1

Coefficient of Determination ( R 2 R^2 R 2 ): R 2 = r 2 R^2 = r^2 R 2 = r 2 R 2 R^2 R 2 represents the proportion of total variation in y y y explained by the linear relationship with x x x . For example, R 2 = 0.81 R^2 = 0.81 R 2 = 0.81 means 81% of the variation in y y y is explained by x x x .

Worked Engineering Example: Stress-Strain Calibration Curve

Problem: An experiment measures force x x x (kN) and resulting deflection y y y (mm) for a structural bracket ( n = 5 n = 5 n = 5 data points):

  • S x x = 20.0 kN 2 S_{xx} = 20.0\text{ kN}^2 S xx ​ = 20.0 kN 2
  • S y y = 80.0 mm 2 S_{yy} = 80.0\text{ mm}^2 S yy ​ = 80.0 mm 2
  • S x y = 36.0 kN ⋅ mm S_{xy} = 36.0\text{ kN}\cdot\text{mm} S x y ​ = 36.0 kN ⋅ mm

x ˉ = 10.0 kN , y ˉ = 25.0 mm \bar{x} = 10.0\text{ kN}, \quad \bar{y} = 25.0\text{ mm} x ˉ = 10.0 kN , y ˉ ​ = 25.0 mm

Find slope b b b , intercept a a a , regression equation y ^ \hat{y} y ^ ​ , correlation coefficient r r r , coefficient of determination R 2 R^2 R 2 , and predicted deflection for x = 15.0 kN x = 15.0\text{ kN} x = 15.0 kN .

Solution: Step 1: Compute Slope b b b and Intercept a a a : b = S x y S x x = 36.0 20.0 = 1.80 mm/kN b = \frac{S_{xy}}{S_{xx}} = \frac{36.0}{20.0} = 1.80\text{ mm/kN} b = S xx ​ S x y ​ ​ = 20.0 36.0 ​ = 1.80 mm/kN a = y ˉ − b x ˉ = 25.0 − ( 1.80 × 10.0 ) = 25.0 − 18.0 = 7.0 mm a = \bar{y} - b \bar{x} = 25.0 - (1.80 \times 10.0) = 25.0 - 18.0 = 7.0\text{ mm} a = y ˉ ​ − b x ˉ = 25.0 − ( 1.80 × 10.0 ) = 25.0 − 18.0 = 7.0 mm Regression Line: y ^ = 7.0 + 1.80 x \text{Regression Line: } \hat{y} = 7.0 + 1.80 x Regression Line: y ^ ​ = 7.0 + 1.80 x

Step 2: Compute Correlation r r r and R 2 R^2 R 2 : r = S x y S x x S y y = 36.0 20.0 × 80.0 = 36.0 1600 = 36.0 40.0 = 0.900 r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} = \frac{36.0}{\sqrt{20.0 \times 80.0}} = \frac{36.0}{\sqrt{1600}} = \frac{36.0}{40.0} = 0.900 r = S xx ​ S yy ​ ​ S x y ​ ​ = 20.0 × 80.0 ​ 36.0 ​ = 1600 ​ 36.0 ​ = 40.0 36.0 ​ = 0.900 R 2 = r 2 = ( 0.900 ) 2 = 0.8100 ( 81.0 % of variance explained ) R^2 = r^2 = (0.900)^2 = 0.8100 \quad (81.0\% \text{ of variance explained}) R 2 = r 2 = ( 0.900 ) 2 = 0.8100 ( 81.0% of variance explained )

Step 3: Predict Response at x = 15.0 kN x = 15.0\text{ kN} x = 15.0 kN : y ^ ( 15.0 ) = 7.0 + ( 1.80 × 15.0 ) = 7.0 + 27.0 = 34.0 mm \hat{y}(15.0) = 7.0 + (1.80 \times 15.0) = 7.0 + 27.0 = 34.0\text{ mm} y ^ ​ ( 15.0 ) = 7.0 + ( 1.80 × 15.0 ) = 7.0 + 27.0 = 34.0 mm

Goodness of Fit: What Each Statistic Actually Tells You

NCEES lists goodness of fit as its own sub-topic and names three measures: the correlation coefficient, standard errors, and R 2 R^2 R 2 . They answer different questions, and the exam tests whether you know which is which.

Statistic Range What it measures What it does not tell you

r r r (correlation coefficient) − 1 -1 − 1 to + 1 +1 + 1 Strength and direction of a linear association Whether the relationship is causal, or whether a nonlinear model would fit better

R 2 R^2 R 2 (coefficient of determination) 0 0 0 to 1 1 1 Fraction of the variance in y y y explained by the regression Direction of the relationship (the sign is squared away)

S e S_e S e ​ (standard error of estimate) ≥ 0 \ge 0 ≥ 0 , in units of y y y Typical size of a residual — the scatter about the fitted line Anything about the slope's own uncertainty R 2 = r 2 = 1 − S S E S S T , S e = S S E n − 2 R^2 = r^2 = 1 - \frac{SSE}{SST}, \qquad S_e = \sqrt{\frac{SSE}{n-2}} R 2 = r 2 = 1 − SST SSE ​ , S e ​ = n − 2 SSE ​ ​

where S S E = ∑ ( y i − y ^ i ) 2 SSE = \sum(y_i - \hat{y}_i)^2 SSE = ∑ ( y i ​ − y ^ ​ i ​ ) 2 is the residual sum of squares and S S T = ∑ ( y i − y ˉ ) 2 SST = \sum(y_i - \bar{y})^2 SST = ∑ ( y i ​ − y ˉ ​ ) 2 is the total sum of squares. The n − 2 n-2 n − 2 divisor reflects the two parameters (slope and intercept) already estimated from the data.

The Three Traps NCEES Builds Into Regression Items

Sign confusion. R 2 = 0.81 R^2 = 0.81 R 2 = 0.81 is consistent with r = + 0.9 r = +0.9 r = + 0.9 or r = − 0.9 r = -0.9 r = − 0.9 . You must read the slope's sign to know the direction. If a question gives you R 2 R^2 R 2 and asks for r r r , it must also tell you whether the trend rises or falls.

"High R 2 R^2 R 2 means the model is right." It does not. Fitting a straight line to genuinely curved data can still yield R 2 > 0.9 R^2 > 0.9 R 2 > 0.9 over a narrow range while extrapolating disastrously. Goodness of fit is a statement about residuals inside the sampled range only .

Extrapolation. A calibration curve fit over 0 0 0 – 50 kPa 50\ \text{kPa} 50 kPa says nothing about behavior at 200 kPa 200\ \text{kPa} 200 kPa .

Interpretation script for the exam: " R 2 = 0.94 R^2 = 0.94 R 2 = 0.94 means 94% of the observed variation in the response is accounted for by the linear model; the remaining 6% is residual scatter." That sentence, with the right number substituted, is the correct answer to most conceptual regression items.

3. NCEES Reference Summary Table

Statistical Tool Formula / Statistic Conditions / Critical Values

Z-Confidence Interval x ˉ ± z α / 2 ( σ / n ) \bar{x} \pm z_{\alpha/2} (\sigma / \sqrt{n}) x ˉ ± z α /2 ​ ( σ / n ​ ) Known σ \sigma σ or n ≥ 30 n \ge 30 n ≥ 30 ; z 0.025 = 1.960 z_{0.025} = 1.960 z 0.025 ​ = 1.960

t-Confidence Interval x ˉ ± t α / 2 , n − 1 ( s / n ) \bar{x} \pm t_{\alpha/2, n-1} (s / \sqrt{n}) x ˉ ± t α /2 , n − 1 ​ ( s / n ​ ) Unknown σ \sigma σ , n < 30 n < 30 n < 30 , d f = n − 1 df = n-1 df = n − 1

Z-Test Statistic Z = x ˉ − μ 0 σ / n Z = \frac{\bar{x} - \mu_0}{\sigma / \sqrt{n}} Z = σ / n ​ x ˉ − μ 0 ​ ​ Compare to Z c r i t Z_{crit} Z cr i t ​ or Φ ( Z ) \Phi(Z) Φ ( Z ) for p p p -value

t-Test Statistic t = x ˉ − μ 0 s / n t = \frac{\bar{x} - \mu_0}{s / \sqrt{n}} t = s / n ​ x ˉ − μ 0 ​ ​ Compare to t c r i t t_{crit} t cr i t ​ with d f = n − 1 df = n-1 df = n − 1

Regression Slope b = S x y S x x b = \frac{S_{xy}}{S_{xx}} b = S xx ​ S x y ​ ​ Least-squares fit minimizing ∑ e i 2 \sum e_i^2 ∑ e i 2 ​

Regression Intercept a = y ˉ − b x ˉ a = \bar{y} - b\bar{x} a = y ˉ ​ − b x ˉ Line passes through ( x ˉ , y ˉ ) (\bar{x}, \bar{y}) ( x ˉ , y ˉ ​ )

Correlation Coefficient r = S x y S x x S y y r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} r = S xx ​ S yy ​ ​ S x y ​ ​ Scale-invariant measure of linear association