Study guide
6.7 Regression & Correlation Analysis
Certified Quality Engineer CQE
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Management and Leadership
1.1 Quality Philosophies & Foundations 1.2 Leadership & Organizational Systems 1.3 ASQ Code of Ethics & Professional Standards 1.4 Strategic Planning & Deployment 1.5 Management Infrastructure & Project Management 1.6 Quality Information Systems & Communication 1.7 Financial & Cost of Quality
2 Chapter 2: The Quality System
2.1 Quality Management System Elements & Standards 2.2 Quality Documentation Architecture 2.3 Quality Auditing Principles & Types 2.4 Audit Planning, Execution & Reporting 2.5 Corrective Action & Preventive Action Systems 2.6 Supplier Quality Management 2.7 Customer Relations & Customer Value 2.8 Quality Training Program Design & Effectiveness
3 Chapter 3: Product, Process, and Service Design
3.1 Design Input & Requirements 3.2 Design Review & Verification 3.3 Failure Mode and Effects Analysis 3.4 Design of Experiments Foundations 3.5 Robust Design & Taguchi Methods 3.6 Reliability & Maintainability Concepts 3.7 Reliability Modeling & System Evaluation 3.8 Technical Drawings, Specifications & GD&T
4 Chapter 4: Product and Process Control
4.1 Statistical Process Control Principles 4.2 Variable Control Charts 4.3 Attribute Control Charts 4.4 Process Capability & Performance Analysis 4.5 Acceptance Sampling Plans & Standards 4.6 Metrology, Calibration & Measurement Uncertainty 4.7 Measurement System Analysis (MSA) 4.8 Material Control, Traceability & MRB
5 Chapter 5: Continuous Improvement
5.1 Continuous Improvement Methodologies 5.2 Classic Quality Tools 5.3 Management & Planning Tools 5.4 Lean Tools & Waste Elimination 5.5 Problem-Solving Methodologies & Root Cause Analysis 5.6 Risk Management & Assessment 5.7 Change Management & Team Facilitation
6 Chapter 6: Quantitative Methods and Tools
6.1 Descriptive Statistics & Data Summarization 6.2 Probability Concepts & Distributions 6.3 Sampling Distributions & Central Limit Theorem 6.4 Point & Interval Estimation 6.5 Hypothesis Testing Fundamentals 6.6 Specific Statistical Tests 6.7 Regression & Correlation Analysis
7 Chapter 7: Risk Management & Advanced Quality Engineering
7.1 Risk Identification & Hazard Analysis 7.2 Quantitative Risk Analysis & Modeling 7.3 Risk Mitigation & Control Planning 7.4 Special Process Control & Advanced SPC 7.5 Advanced Statistical Design & Analysis 7.6 Computer Applications & Quality Engineering Software
6.7 Regression & Correlation Analysis
Key Takeaways Explain This Topic AI
Pearson's correlation coefficient (r) measures the strength and direction of a linear relationship between two continuous variables, bounded strictly between -1.0 and +1.0.
The coefficient of determination (R²) quantifies the proportion of total variation in the response variable (y) explained by the linear regression model (R² = SSR / SST = 1 - SSE / SST).
Simple linear regression models the functional relationship y = β0 + β1 x + ε, estimating parameters via Ordinary Least Squares (OLS) to minimize the Sum of Squared Errors (SSE).
Model assumptions (LINE: Linearity, Independence, Normality, Equal variance/Homoscedasticity) must be validated through systematic residual diagnostics (e_i = y_i - ŷ_i).
A statistically significant correlation does not imply causation; confounding process variables, non-linearities, and influential leverage points must be thoroughly investigated.
Last updated: July 2026
6.7 Regression & Correlation Analysis
In continuous process improvement, quality engineers frequently need to investigate relationships between process input variables ( x x x , independent predictor) and product output parameters ( y y y , dependent response). Correlation analysis quantifies the strength and direction of linear association, while regression analysis develops a mathematical model to predict y y y from x x x and optimize process settings.
1. Pearson Correlation Coefficient ( r r r )
The sample Pearson correlation coefficient ( r r r ) measures the strength and direction of a linear relationship between two continuous variables: r = S x y S x x S y y = ∑ ( x i − x ˉ ) ( y i − y ˉ ) ∑ ( x i − x ˉ ) 2 ∑ ( y i − y ˉ ) 2 r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum (x_i - \bar{x})^2 \sum (y_i - \bar{y})^2}} r = S xx S yy S x y = ∑ ( x i − x ˉ ) 2 ∑ ( y i − y ˉ ) 2 ∑ ( x i − x ˉ ) ( y i − y ˉ )
Properties of r r r :
Strictly bounded: − 1.0 ≤ r ≤ + 1.0 -1.0 \le r \le +1.0 − 1.0 ≤ r ≤ + 1.0 .
r = + 1.0 r = +1.0 r = + 1.0 : Perfect positive linear association.
r = − 1.0 r = -1.0 r = − 1.0 : Perfect negative linear association.
r = 0.0 r = 0.0 r = 0.0 : No linear association (nonlinear patterns may still exist).
Testing Statistical Significance of r r r :
To test whether a sample correlation r r r indicates a non-zero population correlation ( ρ ≠ 0 \rho \neq 0 ρ = 0 ), calculate the t t t -statistic with d f = n − 2 df = n - 2 df = n − 2 : t c a l c = r n − 2 1 − r 2 t_{calc} = \frac{r \sqrt{n - 2}}{\sqrt{1 - r^2}} t c a l c = 1 − r 2 r n − 2
2. Coefficient of Determination ( R 2 R^2 R 2 )
The coefficient of determination ( R 2 R^2 R 2 ) represents the proportion of total variability in response y y y explained by the linear model: R 2 = r 2 = S S R S S T = 1 − S S E S S T R^2 = r^2 = \frac{SSR}{SST} = 1 - \frac{SSE}{SST} R 2 = r 2 = SST SSR = 1 − SST SSE
S S T = ∑ ( y i − y ˉ ) 2 SST = \sum (y_i - \bar{y})^2 SST = ∑ ( y i − y ˉ ) 2 : Total Sum of Squares
S S R = ∑ ( y ^ i − y ˉ ) 2 SSR = \sum (\hat{y}_i - \bar{y})^2 SSR = ∑ ( y ^ i − y ˉ ) 2 : Regression Sum of Squares (Explained variation)
S S E = ∑ ( y i − y ^ i ) 2 SSE = \sum (y_i - \hat{y}_i)^2 SSE = ∑ ( y i − y ^ i ) 2 : Error Sum of Squares (Unexplained variation)
Interpretation Example: If r = 0.90 r = 0.90 r = 0.90 , then R 2 = 0.81 R^2 = 0.81 R 2 = 0.81 . This means 81 % 81\% 81% of the variation in the response variable is explained by the linear relationship with x x x , while 19 % 19\% 19% is due to unexplained random noise or unmeasured process variables.
3. Simple Linear Regression Model & OLS Parameter Estimation
Simple linear regression models a linear relationship between predictor x x x and response y y y : y = β 0 + β 1 x + ε y = \beta_0 + \beta_1 x + \varepsilon y = β 0 + β 1 x + ε
where β 0 \beta_0 β 0 is the population y-intercept, β 1 \beta_1 β 1 is the population slope, and ε ∼ N ( 0 , σ 2 ) \varepsilon \sim N(0, \sigma^2) ε ∼ N ( 0 , σ 2 ) is the random error term.
Ordinary Least Squares (OLS) Estimators
OLS estimates parameters β ^ 0 \hat{\beta}_0 β ^ 0 and β ^ 1 \hat{\beta}_1 β ^ 1 by minimizing the sum of squared residuals ∑ e i 2 = ∑ ( y i − y ^ i ) 2 \sum e_i^2 = \sum (y_i - \hat{y}_i)^2 ∑ e i 2 = ∑ ( y i − y ^ i ) 2 : Estimated Slope: β ^ 1 = S x y S x x = ∑ ( x i − x ˉ ) ( y i − y ˉ ) ∑ ( x i − x ˉ ) 2 \text{Estimated Slope:} \quad \hat{\beta}_1 = \frac{S_{xy}}{S_{xx}} = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sum (x_i - \bar{x})^2} Estimated Slope: β ^ 1 = S xx S x y = ∑ ( x i − x ˉ ) 2 ∑ ( x i − x ˉ ) ( y i − y ˉ ) Estimated Intercept: β ^ 0 = y ˉ − β ^ 1 x ˉ \text{Estimated Intercept:} \quad \hat{\beta}_0 = \bar{y} - \hat{\beta}_1 \bar{x} Estimated Intercept: β ^ 0 = y ˉ − β ^ 1 x ˉ Fitted Regression Line: y ^ = β ^ 0 + β ^ 1 x \text{Fitted Regression Line:} \quad \hat{y} = \hat{\beta}_0 + \hat{\beta}_1 x Fitted Regression Line: y ^ = β ^ 0 + β ^ 1 x
4. Model Assumptions & Residual Diagnostics (LINE)
For OLS regression estimates and hypothesis tests to be valid, four underlying assumptions regarding residuals ( e i = y i − y ^ i e_i = y_i - \hat{y}_i e i = y i − y ^ i ) must be satisfied ( LINE acronym):
Linearity: The true relationship between x x x and y y y is linear.
Independence: Residuals are independent of one another (no autocorrelation over time).
Normality: Residuals are normally distributed with mean zero ( e i ∼ N ( 0 , σ 2 ) e_i \sim N(0, \sigma^2) e i ∼ N ( 0 , σ 2 ) ).
Equal Variance (Homoscedasticity): Residual variance is constant across all predicted values y ^ \hat{y} y ^ and predictor values x x x .
Residual Plot Diagnostics:
Residuals vs. Fitted Values ( y ^ \hat{y} y ^ ): Should show a random scatter of points around zero. A funnel pattern indicates non-constant variance (heteroscedasticity); a curved pattern indicates non-linearity.
Normal Probability Plot of Residuals: Points should align closely along a straight diagonal line.
5. Worked Example: Regression of Adhesive Bond Strength on Cure Temperature
Scenario: A quality engineer evaluates how cure temperature ( x x x in ∘ C ^\circ\text{C} ∘ C ) affects adhesive bond strength ( y y y in MPa \text{MPa} MPa ). A sample of n = 5 n = 5 n = 5 experimental runs produces the following summary statistics:
x ˉ = 100 ∘ C , y ˉ = 50 MPa \bar{x} = 100^\circ\text{C}, \quad \bar{y} = 50\text{ MPa} x ˉ = 10 0 ∘ C , y ˉ = 50 MPa
- S x x = ∑ ( x i − x ˉ ) 2 = 250 S_{xx} = \sum (x_i - \bar{x})^2 = 250 S xx = ∑ ( x i − x ˉ ) 2 = 250
- S y y = ∑ ( y i − y ˉ ) 2 = 160 S_{yy} = \sum (y_i - \bar{y})^2 = 160 S yy = ∑ ( y i − y ˉ ) 2 = 160
S x y = ∑ ( x i − x ˉ ) ( y i − y ˉ ) = 180 S_{xy} = \sum (x_i - \bar{x})(y_i - \bar{y}) = 180 S x y = ∑ ( x i − x ˉ ) ( y i − y ˉ ) = 180
Step-by-Step Regression Solution:
Calculate Slope ( β ^ 1 \hat{\beta}_1 β ^ 1 ): β ^ 1 = S x y S x x = 180 250 = 0.72 MPa / ∘ C \hat{\beta}_1 = \frac{S_{xy}}{S_{xx}} = \frac{180}{250} = 0.72\text{ MPa}/^\circ\text{C} β ^ 1 = S xx S x y = 250 180 = 0.72 MPa / ∘ C
Calculate Intercept ( β ^ 0 \hat{\beta}_0 β ^ 0 ): β ^ 0 = y ˉ − β ^ 1 x ˉ = 50 − 0.72 ( 100 ) = 50 − 72 = − 22.0 MPa \hat{\beta}_0 = \bar{y} - \hat{\beta}_1 \bar{x} = 50 - 0.72(100) = 50 - 72 = -22.0\text{ MPa} β ^ 0 = y ˉ − β ^ 1 x ˉ = 50 − 0.72 ( 100 ) = 50 − 72 = − 22.0 MPa
Formulate Fitted Regression Equation: y ^ = − 22.0 + 0.72 x \hat{y} = -22.0 + 0.72 x y ^ = − 22.0 + 0.72 x
Calculate Pearson Correlation ( r r r ) and R 2 R^2 R 2 : r = S x y S x x S y y = 180 250 × 160 = 180 40000 = 180 200 = 0.90 r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} = \frac{180}{\sqrt{250 \times 160}} = \frac{180}{\sqrt{40000}} = \frac{180}{200} = 0.90 r = S xx S yy S x y = 250 × 160 180 = 40000 180 = 200 180 = 0.90 R 2 = ( 0.90 ) 2 = 0.81 ( 81 % of variation explained ) R^2 = (0.90)^2 = 0.81 \quad (81\% \text{ of variation explained}) R 2 = ( 0.90 ) 2 = 0.81 ( 81% of variation explained )
Prediction Example: For a cure temperature of x = 110 ∘ C x = 110^\circ\text{C} x = 11 0 ∘ C : y ^ = − 22.0 + 0.72 ( 110 ) = − 22.0 + 79.2 = 57.2 MPa \hat{y} = -22.0 + 0.72(110) = -22.0 + 79.2 = 57.2\text{ MPa} y ^ = − 22.0 + 0.72 ( 110 ) = − 22.0 + 79.2 = 57.2 MPa