• Edited by:Priyanka Negi
• Education
• Jan 22, 2026

Explore Data Science Topics With Harvard's Free Online Classes

Harvard University is offering seven free data science online courses. The course duration is eight to nine weeks, with one to two hours of study time per week. The last date of application is June 17, 2026 and candidates can visit the official website to apply.

The courses names are Visualization, Inference and Modeling, Causal Diagrams: Define Your Hypotheses Before Drawing Conclusions, Capstone, Digital Humanities in Practice: From Research Questions to Results, Probability and Linear Regression.

Below you can check about the courses:

1. Data Science: Inference and Modeling

This course explains how to use inference and modeling to develop statistical methods that are useful in conducting effective opinion polls.

2. Causal Diagrams: Define Your Hypotheses Before Drawing Conclusions

The first part of this course consists of five lessons that explain the principles of causal diagrams and their use in the context of causal inference. The second part, through case studies, demonstrates how causal diagrams are applied to real-world situations in health and social sciences.

3. Data Science: Capstone

This is a two-week specialized course, requiring 15 to 20 hours per week. Through this capstone project, students have the opportunity to apply the R data analysis knowledge and skills learned during the course series.

4. Digital Humanities in Practice: From Research Questions to Results

In this course, students work on components of a search engine based on the requirements of academic research. They are also given an understanding of basic text analysis techniques, which are considered the foundation of digital humanities.

5. Data Science: Probability

This course introduces important statistical concepts such as random variables, independence, Monte Carlo simulations, expected value, standard error, and the central limit theorem.

6. Data Science: Linear Regression

This course teaches how to implement linear regression using R and how to balance confounding factors in real-world situations.