Exploring bivariate numerical data

This course is a part of Statistics and probability, a 16-course Topic series from Khan Academy.

We use scatter plots to explore the relationship between two quantitative variables, and we use regression to model the relationship and make predictions. This unit explores linear regression and how to assess the strength of linear models.

This course contains 6 segments:

Introduction to scatterplots

Let's go over the basics of how to construct a scatter plot and describe the relationship we see with its form, direction, and strength.

Correlation coefficients

We'll learn to calculate and use the correlation coefficient r to describe the strength and direction of a linear relationship.

Introduction to trend lines

When we see a pattern in a scatterplot, it's often useful to model that pattern with a trend line. This lesson covers how to informally fit a line to data, make predictions, and think about the slope and y-intercept of the trend line.

Least-squares regression equations

Once we see a linear pattern in a scatterplot, we fit a line to model the relationship and make predictions. But how do we know what line fits the data "best?" To solve this problem, we'll learn about residuals and least-squares regression .

Assessing the fit in least-squares regression

Let's learn how to measure how well a linear model actually fits the relationship in a scatterplot using residual plots and r-squared. We'll also look at how outliers can be influential points in least-squares regression.

More on regression

This is a more advanced collection of videos that goes into some more advanced topics and proofs in least-squares regression.