Building a Regression Model

Statistics · Axiom Academy

EXAMPLE Building a Regression Model Step-by-step walkthrough of linear regression analysis from scatterplot to predictions Study Time vs. Test Score Data Excellent work! You've completed a full regression analysis. Here's what we learned: The regression line equation: y = 4.5x + 56 describes the relationship between study hours and test scores. The slope (4.5) tells us that each additional hour of study is associated with a 4.5-point increase in test score. Computing the slope and intercept: We use the formulas b = Σ[(x - x̄)(y - ȳ)] / Σ(x - x̄)² for slope and a = ȳ - b·x̄ for intercept. These minimize the sum of squared errors. Making predictions: Once we have the equation, we can predict scores for any study time. For 8 hours, we predict 92 points. R² measures goodness of fit: Our R² = 0.976 means 97.6% of the variation in test scores is explained by study hours, indicating an excellent model fit. Interpretation matters: Statistical results must be interpreted in context. Correlation doesn't prove causation, and predictions are most reliable within the range of observed data. This regression approach is fundamental to statistics and can be applied to countless real-world scenarios from economics to biology to social sciences!

This is the written version of the interactive lesson above. See the full Statistics course.