Understanding Correlation and Regression

Correlation measures the strength and direction of a linear relationship between two quantitative variables, while regression lines use this relationship to make predictions. It's crucial to remember that correlation does not imply causation.

Core Principles

• Correlation quantifies the linear association between two variables.
• Scatterplots visually represent bivariate data and potential correlations.
• The correlation coefficient (r) ranges from -1 to +1.
• A positive r indicates a positive linear relationship; a negative r indicates a negative linear relationship.
• An r close to 0 suggests a weak or no linear relationship.
• Statistical significance determines if a correlation is likely due to a real relationship or random chance.
• Regression lines (lines of best fit) model linear relationships for prediction.
• Correlation does not prove causation; other factors (lurking variables) may be involved.

Action Steps

• Create a scatterplot to visualize the relationship between two variables.
• Calculate the correlation coefficient (r) to quantify the linear association.
• Determine if the correlation is statistically significant using a critical values table.
• If significant, find the equation of the least-squares regression line.
• Use the regression line to make predictions, understanding its limitations.
• Always consider if correlation implies causation or if lurking variables are present.

Formulas

• Linear Correlation (sample): $r = \frac{\sum_{i=1}^{n} (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum_{i=1}^{n} (x_i - \bar{x})^2 \sum_{i=1}^{n} (y_i - \bar{y})^2}}$
• Least-Squares Regression Line: $\hat{y} = b_0 + b_1x$
• Slope ($b_1$): Represents the change in the predicted y-value for a one-unit increase in x.
• Y-intercept ($b_0$): Represents the predicted y-value when x is 0.

Key Terms

• Scatterplot: A two-dimensional graph plotting paired quantitative data points.
• Bivariate Data: Data consisting of two quantitative variables, expressed as ordered pairs (x, y).
• Correlation Coefficient (r): A statistic measuring the strength and direction of a linear relationship (-1 to +1).
• Statistically Significant: When there is enough sample evidence to conclude a linear relationship exists.
• Least-Squares Regression Line: The trend line that best models paired quantitative data with a significant correlation.
• Line of Best Fit: Another term for the least-squares regression line or trend line.
• Causation: When one variable directly causes a change in another variable.
• Lurking Variable: An unmeasured variable that affects the relationship between the two observed variables.