Study guide

15.3 Scatterplots, Correlation Coefficient r, Linear Regression & Residuals

FTCE Mathematics 6-12 026

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Exam Overview & Rational Number Systems

1.1 FTCE Mathematics 6-12 Overview, Interface & Test-Taking Strategies 1.2 Rational Numbers, Representations & Multi-Step Operations

2 Chapter 2: Real Number Approximations & Proportional Reasoning

2.1 Irrational Numbers, Radical Estimation & Scientific Notation 2.2 Ratios, Rates, Proportions & Dimensional Analysis 2.3 Percent Applications, Financial Models & Depreciation

3 Chapter 3: Algebraic Expressions & Polynomial Operations

3.1 Polynomial Arithmetic, Factoring Strategies & Special Products 3.2 Radical Operations, Rational Expressions & Equations 3.3 Field Properties, Exponent Laws & Literal Equations

4 Chapter 4: Linear Relationships, Systems & Inequalities

4.1 Linear Equations, Slope Concepts & Line Forms 4.2 Systems of Linear Equations & Real-World Applications 4.3 Single & Systems of Linear Inequalities on the Coordinate Plane 4.4 Absolute Value Equations & Inequalities in One and Two Variables

5 Chapter 5: Quadratic Equations, Functions & Complex Numbers

5.1 Quadratic Equations, Discriminant & Complex Number Arithmetic 5.2 Quadratic Functions: Vertex, Axis, Intercepts & Concavity 5.3 Quadratic Inequalities in One/Two Variables & Optimization

6 Chapter 6: Function Foundations, Transformations & Inverses

6.1 Function Definitions, Relations, Domain, Range & Notation 6.2 Function Families, Key Graphical Behaviors & Even/Odd Symmetry 6.3 Shifting, Scaling & Reflecting Functions Across Representations 6.4 Function Composition, Invertibility & Inverse Functions

7 Chapter 7: Geometric Foundations, Angles & Deductive Proofs

7.1 Angle Relationships, Parallel Lines & Transversals 7.2 Triangle Sum, Exterior Angles & Polygon Angle Formulas 7.3 Formal Proof Methods, CPCTC & Quadrilateral Properties

8 Chapter 8: Congruence, Similarity & Geometric Transformations

8.1 Triangle Congruence (SSS, SAS, ASA, AAS, HL) & Similarity (AA, SAS, SSS) 8.2 Rigid Motions: Translations, Reflections, Rotations & Symmetry 8.3 Dilations, Scale Factors & Proportional Reasoning in Geometry

9 Chapter 9: Coordinate Geometry & 3D Spatial Measurement

9.1 Distance, Midpoint, Slope Criteria & Coordinate Proofs 9.2 Perimeter, Area, Surface Area & Volume of 2D and 3D Figures 9.3 Cross Sections of 3D Figures & Solids of Revolution

10 Chapter 10: Circle Theorems & Geometric Constructions

10.1 Central, Inscribed, Tangent & Chord Angles and Arcs 10.2 Intersecting Chords, Secant/Tangent Lengths, Arc Length & Sector Area 10.3 Euclidean Constructions with Compass and Straightedge

11 Chapter 11: Conic Sections & Coordinate Modeling

11.1 Circles and Parabolas: Geometric Definitions, Equations & Modeling 11.2 Ellipses and Hyperbolas: Foci, Axes, Asymptotes & Real-World Applications

12 Chapter 12: Trigonometric Foundations & Oblique Triangles

12.1 Right-Triangle Trigonometry, Pythagorean Theorem & Special Triangles 12.2 The Unit Circle, Radian Measure & Exact Trigonometric Values 12.3 Law of Sines (Ambiguous Case), Law of Cosines & Oblique Triangle Area

13 Chapter 13: Trigonometric Functions, Polar & Parametric Systems

13.1 Graphs of the Six Trigonometric Functions & Periodic Modeling 13.2 Polar Coordinates, Polar-Rectangular Conversion & Polar Graphs 13.3 Parametric Equations, Elimination of Parameters & Projectile Motion

14 Chapter 14: Descriptive Statistics & Data Distributions

14.1 Measures of Central Tendency, Dispersion & Resistant Statistics 14.2 Univariate Displays (Histograms, Boxplots, Stemplots) & Distribution Shapes 14.3 Normal Distribution Properties, Empirical Rule & Standard Normal Z-Scores

15 Chapter 15: Statistical Studies, Bivariate Data & Linear Regression

15.1 Statistical Study Design, Random Sampling Techniques & Sources of Bias 15.2 Survey Sampling, Margin of Error & Population Estimation 15.3 Scatterplots, Correlation Coefficient r, Linear Regression & Residuals

16 Chapter 16: Probability Theory, Two-Way Tables & Expected Value

16.1 Fundamental Counting Principle, Permutations & Combinations 16.2 Addition Rule, Multiplication Rule, Independent Events & Conditional Probability 16.3 Two-Way Frequency Tables, Law of Large Numbers & Expected Value of Random Variables

17 Chapter 17: Precalculus: Polynomial, Rational & Transcendental Functions

17.1 Fundamental Theorem of Algebra, Remainder/Factor Theorems & Polynomial Roots 17.2 Rational Functions, Domain Restrictions, Holes & Asymptotes 17.3 Exponential & Logarithmic Functions, Base e & Logarithmic Properties

18 Chapter 18: Sequences, Series, Binomial Theorem & Vectors

18.1 Arithmetic & Geometric Sequences, Finite Sums & Infinite Series Convergence 18.2 Binomial Theorem, Pascal's Triangle & General Term Determination 18.3 Vector Representations, Operations, Magnitude, Direction & Dot Product

19 Chapter 19: Differential Calculus Foundations

19.1 Limits, One-Sided Limits, Indeterminate Forms & Continuity 19.2 Derivative Definition, Power, Product, Quotient & Transcendental Rules 19.3 Chain Rule, Tangent Line Equations & Rates of Change

20 Chapter 20: Applications of Derivatives & Integral Calculus

20.1 First & Second Derivative Tests, Extrema, Concavity & Optimization 20.2 Antiderivatives, Indefinite Integrals & Integration by U-Substitution 20.3 Fundamental Theorem of Calculus, Net Signed Area & Geometric Applications

21 Chapter 21: Mathematical Representations, Progressions & Tools

21.1 Concrete-Representational-Abstract (CPA) Progressions & Concept Continuity 21.2 The Rule of Four: Translating Among Representations to Support Fluency 21.3 Manipulatives, Dynamic Software & Technology in Secondary Mathematics

22 Chapter 22: Assessment, Error Analysis & Misconception Remediation

22.1 Student Misconceptions & Error Analysis Across Core Secondary Domains 22.2 Evaluating Validity of Student Mathematical Arguments & Reasonableness Checks 22.3 Formative Assessment Formats, Data-Driven Differentiation & Tiered Tasks

15.3 Scatterplots, Correlation Coefficient r, Linear Regression & Residuals

Key Takeaways Explain This Topic AI

Bivariate numerical associations on scatterplots are evaluated across four distinct dimensions: direction (positive or negative), form (linear or non-linear), strength (weak, moderate, or strong), and unusual features (outliers and influential leverage points).

Pearson's correlation coefficient r r r is a unitless index bounded by − 1 ≤ r ≤ 1 -1 \le r \le 1 − 1 ≤ r ≤ 1 that measures the direction and strength of linear association only; it is invariant under positive linear transformations of scale and symmetric with respect to x x x and y y y .

The coefficient of determination r 2 r^2 r 2 represents the proportion of total variation in the response variable y y y that is explained by the linear relationship with explanatory variable x x x in the least-squares model.

The least-squares regression line y ^ = a + b x \hat{y} = a + bx y ^ ​ = a + b x minimizes the sum of squared vertical residuals ∑ ( y − y ^ ) 2 \sum (y - \hat{y})^2 ∑ ( y − y ^ ​ ) 2 , with slope b = r s y s x b = r\frac{s_y}{s_x} b = r s x ​ s y ​ ​ , y-intercept a = y ˉ − b x ˉ a = \bar{y} - b\bar{x} a = y ˉ ​ − b x ˉ , and guaranteed passage through centroid ( x ˉ , y ˉ ) (\bar{x}, \bar{y}) ( x ˉ , y ˉ ​ ) .

Residuals e = y − y ^ e = y - \hat{y} e = y − y ^ ​ must exhibit a patternless, random horizontal scatter around zero in a residual plot to confirm linear model adequacy; systematic curvature reveals non-linearity, and predictions beyond the domain of x x x (extrapolation) are fundamentally untrustworthy.

Last updated: September 2026

  • 15.3 Scatterplots, Correlation Coefficient r, Linear Regression & Residuals
  • Bivariate Quantitative Data and Scatterplot Diagnostics

Bivariate data analysis investigates the relationship between two quantitative variables measured on the same individual entities. In most bivariate settings, researchers distinguish between:

The explanatory variable ( x x x , independent variable), which accounts for, explains, or predicts changes in the response.

The response variable ( y y y , dependent variable), which measures the primary outcome under investigation.

A scatterplot provides the primary graphical display for bivariate numerical data, plotting values of the explanatory variable on the horizontal axis and values of the response variable on the vertical axis. When analyzing a scatterplot on the FTCE Mathematics 6-12 examination, you must systematically evaluate four key attributes:

Direction: An association is positive if larger values of x x x correspond to larger values of y y y (the scatter drifts upward from left to right). An association is negative if larger values of x x x correspond to smaller values of y y y (the scatter drifts downward from left to right).

Form: Describes the geometric trajectory of the data points. Form is classified as linear if points cluster along a straight line, or non-linear (curvilinear, exponential, power, or periodic) if the trajectory bends.

Strength: Measures how tightly clustered the observations are around the underlying form. Strong relationships exhibit narrow dispersion around a central curve, whereas weak associations display broad, diffuse scatter.

Unusual Features: Includes isolated outliers (individual points that deviate substantially from the overall bivariate pattern) and clusters/gaps in the data space.

Pearson Correlation Coefficient r r r and Coefficient of Determination r 2 r^2 r 2

The Pearson product-moment correlation coefficient r r r provides a standardized numerical index measuring the direction and strength of the linear relationship between two quantitative variables. Mathematically, r r r is the average product of the standardized scores ( z z z -scores) of x x x and y y y : r = 1 n − 1 ∑ i = 1 n ( x i − x ˉ s x ) ( y i − y ˉ s y ) r = \frac{1}{n - 1} \sum_{i=1}^n \left(\frac{x_i - \bar{x}}{s_x}\right) \left(\frac{y_i - \bar{y}}{s_y}\right) r = n − 1 1 ​ i = 1 ∑ n ​ ( s x ​ x i ​ − x ˉ ​ ) ( s y ​ y i ​ − y ˉ ​ ​ )

Essential Mathematical Properties of r r r

Bounded Range: The correlation coefficient is strictly bounded: − 1 ≤ r ≤ 1 -1 \le r \le 1 − 1 ≤ r ≤ 1 .

r = 1 r = 1 r = 1 indicates a perfect positive linear relationship.

r = − 1 r = -1 r = − 1 indicates a perfect negative linear relationship.

r = 0 r = 0 r = 0 indicates the absence of any linear association.

Dimensionless (Unitless): Because r r r is calculated using standardized z z z -scores, it has no units of measurement. Converting measurements (e.g., changing height from inches to centimeters) leaves r r r completely unchanged.

Scale Invariance: Adding a constant or multiplying by a positive scalar ( x ′ = c 1 x + d 1 , y ′ = c 2 y + d 2 x' = c_1 x + d_1, y' = c_2 y + d_2 x ′ = c 1 ​ x + d 1 ​ , y ′ = c 2 ​ y + d 2 ​ with c 1 , c 2 > 0 c_1, c_2 > 0 c 1 ​ , c 2 ​ > 0 ) leaves r r r invariant. Multiplying by a negative scalar reverses the sign of r r r .

Symmetry: Variable roles are mathematically symmetric: r ( x , y ) = r ( y , x ) r(x, y) = r(y, x) r ( x , y ) = r ( y , x ) . Swapping which variable is assigned to the x x x -axis and y y y -axis has zero effect on r r r .

Linearity Restriction: The coefficient r r r measures linear strength only . A bivariate relationship can exhibit a perfect deterministic curved relationship (such as y = x 2 y = x^2 y = x 2 symmetric over [ − 3 , 3 ] [-3, 3] [ − 3 , 3 ] ) while yielding r = 0 r = 0 r = 0 .

Non-Resistant Statistic: Like the mean and standard deviation, r r r is strongly influenced by extreme outliers.

The Coefficient of Determination r 2 r^2 r 2

The coefficient of determination r 2 r^2 r 2 is the square of Pearson's correlation coefficient. It quantifies the proportion of total variation in the response variable y y y that is accounted for by the linear relationship with explanatory variable x x x : r 2 = SS regression SS total = 1 − SS residuals SS total r^2 = \frac{\text{SS}_{\text{regression}}}{\text{SS}_{\text{total}}} = 1 - \frac{\text{SS}_{\text{residuals}}}{\text{SS}_{\text{total}}} r 2 = SS total ​ SS regression ​ ​ = 1 − SS total ​ SS residuals ​ ​

Contextual Interpretation: 'Approximately r 2 × 100 % r^2 \times 100\% r 2 × 100% of the variation in the response variable ( y y y ) is explained by the linear relationship with the explanatory variable ( x x x ).'

The remaining ( 1 − r 2 ) × 100 % (1 - r^2) \times 100\% ( 1 − r 2 ) × 100% represents residual variation attributable to unmeasured lurking variables or random noise.

The Least-Squares Regression Line (LSRL)

When a scatterplot displays a reasonably linear form, we model the relationship using a straight line. The mathematical model is expressed as: y ^ = a + b x \hat{y} = a + bx y ^ ​ = a + b x

where y ^ \hat{y} y ^ ​ (y-hat) represents the predicted value of the response variable for a given value of x x x , b b b is the slope, and a a a is the y-intercept.

The least-squares regression line (LSRL) is the unique straight line that minimizes the sum of the squared vertical deviations (residuals) between the observed data points and the line: Minimize ∑ i = 1 n ( y i − y ^ i ) 2 \text{Minimize } \sum_{i=1}^n (y_i - \hat{y}_i)^2 Minimize i = 1 ∑ n ​ ( y i ​ − y ^ ​ i ​ ) 2

Formulating the Regression Equation from Summary Statistics

When given summary statistics (means x ˉ , y ˉ \bar{x}, \bar{y} x ˉ , y ˉ ​ , standard deviations s x , s y s_x, s_y s x ​ , s y ​ , and correlation r r r ), the parameters of the LSRL are calculated directly:

Slope ( b b b ): b = r ( s y s x ) b = r \left(\frac{s_y}{s_x}\right) b = r ( s x ​ s y ​ ​ ) Because standard deviations are strictly positive, the sign of the slope b b b is identical to the sign of the correlation coefficient r r r .

Y-Intercept ( a a a ): a = y ˉ − b x ˉ a = \bar{y} - b\bar{x} a = y ˉ ​ − b x ˉ This algebraic formula guarantees that the regression line always passes through the centroid (point of averages), ( x ˉ , y ˉ ) (\bar{x}, \bar{y}) ( x ˉ , y ˉ ​ ) .

Contextual Interpretation of Regression Parameters

Slope ( b b b ): 'For each 1-unit increase in the explanatory variable x x x , the model predicts that the response variable y y y will increase (or decrease) by approximately ∣ b ∣ |b| ∣ b ∣ units.'

Y-Intercept ( a a a ): 'The predicted value of the response variable y y y when the explanatory variable x x x equals 0.' The intercept is physically meaningful only if x = 0 x = 0 x = 0 is plausible within the context and lies within or near the domain of observed data.

Residual Analysis and Linear Appropriateness

A residual e i e_i e i ​ is the directed vertical distance between an observed data point and the regression line: Residual = Observed y − Predicted y = y i − y ^ i \text{Residual } = \text{Observed } y - \text{Predicted } y = y_i - \hat{y}_i Residual = Observed y − Predicted y = y i ​ − y ^ ​ i ​

A positive residual ( e i > 0 e_i > 0 e i ​ > 0 ) means the observed point lies above the line; the model underestimated the actual value.

A negative residual ( e i < 0 e_i < 0 e i ​ < 0 ) means the observed point lies below the line; the model overestimated the actual value.

A fundamental algebraic property of the LSRL is that the sum and mean of the residuals are always zero: ∑ e i = 0 \sum e_i = 0 ∑ e i ​ = 0 and e ˉ = 0 \bar{e} = 0 e ˉ = 0 .

A residual plot is a scatterplot displaying explanatory values x i x_i x i ​ (or predicted values y ^ i \hat{y}_i y ^ ​ i ​ ) on the horizontal axis and residuals e i e_i e i ​ on the vertical axis.

Linear Model is Appropriate: The residual plot displays a random, uniform, horizontal band of scatter around the reference line e = 0 e = 0 e = 0 , with no discernible pattern and approximately constant vertical spread (homoscedasticity).

Linear Model is Inappropriate: The residual plot exhibits a clear curved or U-shaped pattern, indicating that the true relationship is non-linear and requiring a polynomial or non-linear transformation.

Heteroscedasticity: A fan-shaped or megaphone pattern (where vertical spread expands as x x x increases) indicates non-constant variance.

Outliers, High Leverage, Influential Points & Extrapolation

Regression Outlier: A data point with an unusually large residual that deviates far vertically from the fitted line.

High-Leverage Point: A data point whose explanatory coordinate x x x is substantially smaller or larger than the mean x ˉ \bar{x} x ˉ . High-leverage points possess the physical potential to act as a fulcrum, heavily tilting the regression line.

Influential Point: An observation that substantially alters the slope, y-intercept, or correlation coefficient when removed from the dataset. High-leverage points that deviate from the linear trend of the remaining data are strongly influential.

The Danger of Extrapolation: Extrapolation is the practice of using a regression model to predict values of y y y for explanatory values far outside the domain of observed x x x -data. Extrapolation is highly unreliable because researchers have no empirical evidence that the linear pattern persists beyond the observed range.

Worked Exemplar: Deriving the LSRL from Summary Statistics

Problem: An educational psychologist investigates the relationship between weekly time spent reading for pleasure in hours ( x x x ) and vocabulary assessment scores on a 100-point scale ( y y y ) among a random sample of 30 middle school students. The sample summary statistics are:

Mean reading hours: x ˉ = 5.0 hours \bar{x} = 5.0\text{ hours} x ˉ = 5.0 hours , Standard deviation: s x = 2.0 hours s_x = 2.0\text{ hours} s x ​ = 2.0 hours

Mean vocabulary score: y ˉ = 70.0 points \bar{y} = 70.0\text{ points} y ˉ ​ = 70.0 points , Standard deviation: s y = 8.0 points s_y = 8.0\text{ points} s y ​ = 8.0 points

Pearson correlation coefficient: r = 0.75 r = 0.75 r = 0.75

Compute the Regression Slope ( b b b ): b = r ( s y s x ) = 0.75 ( 8.0 2.0 ) = 0.75 × 4.0 = 3.0 points per hour b = r \left(\frac{s_y}{s_x}\right) = 0.75 \left(\frac{8.0}{2.0}\right) = 0.75 \times 4.0 = 3.0\text{ points per hour} b = r ( s x ​ s y ​ ​ ) = 0.75 ( 2.0 8.0 ​ ) = 0.75 × 4.0 = 3.0 points per hour

Compute the Y-Intercept ( a a a ): a = y ˉ − b x ˉ = 70.0 − ( 3.0 ) ( 5.0 ) = 70.0 − 15.0 = 55.0 points a = \bar{y} - b\bar{x} = 70.0 - (3.0)(5.0) = 70.0 - 15.0 = 55.0\text{ points} a = y ˉ ​ − b x ˉ = 70.0 − ( 3.0 ) ( 5.0 ) = 70.0 − 15.0 = 55.0 points

State the Regression Equation: y ^ = 55.0 + 3.0 x \hat{y} = 55.0 + 3.0x y ^ ​ = 55.0 + 3.0 x

Evaluate the Coefficient of Determination ( r 2 r^2 r 2 ): r 2 = ( 0.75 ) 2 = 0.5625 = 56.25 % r^2 = (0.75)^2 = 0.5625 = 56.25\% r 2 = ( 0.75 ) 2 = 0.5625 = 56.25%

Interpretation: Exactly 56.25 % 56.25\% 56.25% of the variation in vocabulary scores is explained by the linear relationship with weekly reading hours.

Predict Score and Calculate Residual: A student in the sample reads 7.0 7.0 7.0 hours per week and achieves an observed vocabulary score of 79.0 79.0 79.0 points.

Predicted score: y ^ = 55.0 + 3.0 ( 7.0 ) = 55.0 + 21.0 = 76.0 points \hat{y} = 55.0 + 3.0(7.0) = 55.0 + 21.0 = 76.0\text{ points} y ^ ​ = 55.0 + 3.0 ( 7.0 ) = 55.0 + 21.0 = 76.0 points .

Residual: e = y − y ^ = 79.0 − 76.0 = + 3.0 points e = y - \hat{y} = 79.0 - 76.0 = +3.0\text{ points} e = y − y ^ ​ = 79.0 − 76.0 = + 3.0 points .

Interpretation: Because the residual is positive ( + 3.0 +3.0 + 3.0 ), the observed score lies above the regression line; the model underestimated the student's actual performance by 3.0 3.0 3.0 points.

Regression Parameters and Bivariate Diagnostics Guide

Statistic / Parameter Mathematical Formula Contextual Meaning Diagnostic Criterion / Pitfall

Correlation Coefficient r r r 1 n − 1 ∑ z x z y \frac{1}{n-1}\sum z_x z_y n − 1 1 ​ ∑ z x ​ z y ​ Direction and strength of linear association ( − 1 ≤ r ≤ 1 -1 \le r \le 1 − 1 ≤ r ≤ 1 ) Only measures linear patterns; highly non-resistant to outliers

Coefficient of Determination r 2 r^2 r 2 r 2 = 1 − SS resid SS total r^2 = 1 - \frac{\text{SS}_{\text{resid}}}{\text{SS}_{\text{total}}} r 2 = 1 − SS total ​ SS resid ​ ​ Proportion of response variance explained by linear model Does not imply causation; unaffected by non-linear relationships

Regression Slope b b b r ( s y s x ) r \left(\frac{s_y}{s_x}\right) r ( s x ​ s y ​ ​ ) Rate of change in predicted y ^ \hat{y} y ^ ​ per 1-unit increase in x x x Asymmetric: regression of y y y on x x x differs from x x x on y y y

Y-Intercept a a a y ˉ − b x ˉ \bar{y} - b\bar{x} y ˉ ​ − b x ˉ Predicted value of response variable y ^ \hat{y} y ^ ​ when x = 0 x = 0 x = 0 Meaningless if x = 0 x = 0 x = 0 is outside observed domain or physically impossible

Residual e e e y − y ^ y - \hat{y} y − y ^ ​ Vertical error of model prediction for an observed point Sums to zero ( ∑ e = 0 \sum e = 0 ∑ e = 0 ); negative residual indicates model overestimation

Residual Plot Scatter of ( x , e ) (x, e) ( x , e ) or ( y ^ , e ) (\hat{y}, e) ( y ^ ​ , e ) Validates linearity and constant error variance (homoscedasticity) Curvature reveals non-linearity; expanding funnel indicates heteroscedasticity