Correlation and Linear Regression: Understanding the Linear Correlation Coefficient (r)

Correlation and Linear Relationships

Definitions and Basic Concepts

Correlation is a fundamental concept in statistics that describes the association between two variables. When the values of one variable are somehow associated with the values of another, a correlation exists. A linear correlation is a specific type of correlation where the plotted points of paired data result in a pattern that can be approximated by a straight line.

• Correlation: Exists between two variables when their values are associated.

• Linear correlation: Exists when the association forms a straight-line pattern in a scatterplot.

Interpreting Scatterplots

Scatterplots are graphical tools used to visualize the relationship between two quantitative variables. They help determine the type and strength of correlation present.

• Positive linear correlation: As x increases, y also increases. The points form an upward-sloping straight line.

• Negative linear correlation: As x increases, y decreases. The points form a downward-sloping straight line.

• No correlation: No distinct pattern; the points are scattered randomly.

• Nonlinear relationship: The association exists but is not linear; points may form a curve or other pattern.

The Linear Correlation Coefficient (r)

Definition and Interpretation

The linear correlation coefficient (r) measures the strength and direction of the linear correlation between paired quantitative x and y values in a sample. It is also known as the Pearson product moment correlation coefficient.

• r: Measures linear correlation in a sample.

• ρ (rho): Measures linear correlation in a population.

• r is computed using specific formulas and is typically calculated using statistical software.

Notation and Calculation

To calculate and interpret the linear correlation coefficient r, certain notations are used:

• n: Number of pairs of sample data.

• Σx, Σy: Sum of all x values and y values, respectively.

• Σx², Σy²: Sum of squared x values and squared y values.

• Σxy: Sum of products of paired x and y values.

• r: Linear correlation coefficient for sample data.

• ρ: Linear correlation coefficient for population data.

Requirements for Using r

Before calculating r, certain requirements must be met to ensure valid results:

• The sample of paired (x, y) data must be a simple random sample of quantitative data.

• Visual examination of the scatterplot must confirm that the points approximate a straight-line pattern.

• Outliers must be considered, as they can strongly affect the value of r.

• For formal inference, the paired data should have a bivariate normal distribution.

Formula for Calculating r

The formula for the linear correlation coefficient r is:

$r = \frac{n(\sum xy) - (\sum x)(\sum y)}{\sqrt{n(\sum x^2) - (\sum x)^2} \sqrt{n(\sum y^2) - (\sum y)^2}}$

Properties of the Linear Correlation Coefficient r

The linear correlation coefficient r has several important properties:

• The value of r is always between -1 and 1 inclusive ($-1 \leq r \leq 1$).

• Changing the scale of either variable does not affect r.

• Interchanging x and y does not change r.

• r measures only linear relationships, not nonlinear ones.

• r is sensitive to outliers; a single outlier can dramatically affect its value.

Examples: Matching Scatterplots to Correlation Coefficients

Scatterplots can be matched to their corresponding correlation coefficients based on the pattern and direction of the data:

• r = -0.90: Strong negative linear correlation.

• r = 1.00: Perfect positive linear correlation.

• r = -0.33: Weak negative linear correlation.

• r = 0.90: Strong positive linear correlation.

Interpreting r and Explained Variation

Explained Variation (r²)

If a linear correlation exists, a linear equation can be used to predict y from x. The value of r² (coefficient of determination) represents the proportion of variation in y explained by the linear relationship with x.

• $0 \leq r^2 \leq 1$

• When r² is close to 1, most of the variation in y is explained by the linear relationship.

• When r² is close to 0, most of the variation in y is not explained by the linear relationship.

• r² is often expressed as a percentage (0% to 100%).

Correlation Does Not Imply Causation

Even when a linear correlation is found, it does not imply causation. Other variables (lurking variables) may influence the relationship. For example, finding a correlation between chocolate consumption and Nobel Laureates does not mean chocolate causes Nobel Prizes.

• Lurking variables: Variables not included in the study that may affect the results.

• Common errors: Assuming causality, using averages (which may inflate r), and ignoring nonlinear relationships.

Summary Table: Types of Correlation and Their Characteristics

Type of Correlation	Scatterplot Pattern	r Value	Interpretation
Positive Linear	Upward-sloping straight line	r > 0 (close to 1)	As x increases, y increases
Negative Linear	Downward-sloping straight line	r < 0 (close to -1)	As x increases, y decreases
No Correlation	No distinct pattern	r ≈ 0	No association between x and y
Nonlinear Relationship	Curved or other pattern	r may be moderate	Association exists but is not linear

Additional info: The notes above expand on the original content by providing definitions, properties, examples, and a summary table for clarity. All included images directly reinforce the explanation of scatterplots and correlation coefficients.