College Board-Aligned Original Notes

AP Statistics Unit 2 Topic 5: Linear regression models

Use Linear regression models across graphical, numerical, algebraic, and verbal representations.

Unit 2: Exploring Two-Variable Data. College Board exam weighting listed for this unit: 5%-7% of Score.

What to Know

  • Check the conditions of a theorem or method before applying it.
  • Show the setup before the calculation.
  • Interpret the result in context, including units when the problem supplies them.
  • Always connect this topic back to the larger unit: Exploring Two-Variable Data.

Detailed Notes

Linear regression models should be studied through multiple representations. A graph may show behavior quickly, an equation may make calculation possible, and a verbal interpretation explains what the result means.

In AP Statistics, AP questions often award credit for setup and reasoning, not just final answers. Write the expression, theorem, condition, or model before doing the computation.

When this topic appears in free response, check whether the question asks for a value, a rate, an interval, a comparison, or a justification. Use units and context to make the final answer precise.

Key Vocabulary

Distribution

The pattern of variability in a dataset.

Normal distribution

A symmetric bell-shaped distribution described by mean and standard deviation.

Correlation

A measure of direction and strength of a linear association.

Regression line

A line used to model or predict a quantitative response from an explanatory variable.

Residual

Observed value minus predicted value.

Quick Practice

How would you explain Linear regression models in one or two AP-style sentences?

Name the concept, apply it to a specific example or source, and explain the reasoning that connects the evidence to your answer.

Related Topics in This Unit

  • Comparing representations of 2 categorical variables
  • Calculating statistics for 2 categorical variables
  • Representing bivariate quantitative data using scatter plots
  • Describing associations in bivariate data and interpreting correlation