Data Science: Linear Regression

  • Free to audit
  • Paid certificate
  • Course certificate
  • Beginner
  • 8 weeks
Provider
Harvard
Cost
Free to audit
Certificate
Paid, $149
Level
Beginner
Duration
8 weeks
Effort
1–2 hrs/week
Format
Self-paced
Language
English
Subjects
Data Science, Mathematics
Source
Harvard Online
Last verified
14 Sep 2026

Free access to the material ends after the course length; graded work and the certificate are paid.

Data Science: Linear Regression covers how to implement linear regression in R and how to use it to adjust for confounding. The motivating case study is the data-driven approach to building baseball teams described in Moneyball: which measured outcomes best predict runs? Along the way the course explains how Galton originally developed regression, what confounding is and how to detect it, and when regression is and is not the appropriate tool, since extraneous variables can create spurious associations between the variables you care about.

Part of HarvardX's Professional Certificate in Data Science, it is self-paced, eight weeks long at 1 to 2 hours per week, and rated introductory. Harvard lists the course as free to audit on edX, with a verified certificate available for $149.

Advertisement

What you’ll learn

  • Explain how Galton developed linear regression
  • Implement linear regression in R
  • Detect confounding and adjust for it
  • Judge when regression is an appropriate technique

Who it’s for

Learners who know some R and probability and want to quantify relationships between variables.

Source: Harvard Online (opens in a new tab) · Verified · Report a change

Advertisement
Free to auditOpens pll.harvard.edu Go to course (opens in a new tab)