Initial Publication Date: December 10, 2020

Linear Regression

This statistical vignette was developed by Dax Soule, Michelle Weirathmueller, Steve Juliano, and Diana Garcia Silva

Summary

Linear Regression is similar to the correlation coefficient in that both are utilized to understand the relationship between two sets of measurements. Regression expands this concept to create modeled predictions from the extension of the best fit line between a dependent and independent variable. This vignette will help build a student's understanding of linear regression and how to interpret R² coefficients by analyzing lines of best fit.

Learning Objectives

Demonstrate the ability to identify the independent and dependent variables
Investigate the relationship between two datasets and use the extent to which they vary together to predict future values
Recognize that not all data relationships can be described using the Correlation Coefficient

Context for Use

This vignette can be used in either a single lab or lecture session and should take between 15 and 20 minutes for introductory or intermediate level students.

The vignette has been implemented in conjunction with the EDDIE Climate Change Module and utilized in both lab and lecture for the introductory Environmental Science courses at Queens College (CUNY) and at Loyola University of Chicago and is suitable for use as a stand alone activity or as a supplement to where the relationship between two variables is of interest.