The Benefits of Active Learning for Data Science Skills

Since the late 1970s, educators have promoted the adoption of active learning principles in teaching practices. Active learning is a method that involves students directly in the learning process. This contrasts with traditional learning methods, like lectures, where students passively receive information without taking measures to engage with the material and ensure they have sufficiently understood it. Active learning involves getting students to do activities and to think about the purpose behind these activities[1]. At The Data Incubator, we believe that active learning is the best way to approach data science training and education, and we’ve built our curricula based on these concepts. Below we discuss the benefits of active learning and how we’ve employed active learning methods in our data science training programs.

In the seminal text “A Taxonomy of Educational Objectives”[2], Bloom defines the six learning objectives of cognitive domain, the area of mental skill acquisition.[3] They are:

Effective learning in the cognitive domain is achieved by activating all of these objectives. Everyone learns differently; you often hear people state “I’m a visual learner” or “I’m an auditory learner”. The problem with this thinking is that it over simplifies how people learn into only two categories. Additionally, it adheres to the traditional passive learning techniques that rely on audio and visual cues only. With regards to the six objectives listed above, passive learning only addresses the “low order” thinking skills, remembering and understanding,without activating the “high order” thinking skills . During a one or two hour lecture, there’s no opportunity for students to effectively apply and analyze the information – let alone to evaluate or create something from the new information. While students may do so on their own time after lecture, the opportunity is lost to solidify the concepts when information has been most recently seen.

Critics of active learning often decry it as just another fad. However, numerous research studies have refuted this claim. A review of active learning studies found support for various forms of active learning[4]. Given the various studies analyzed in the review, the author suggests that introducing activities during lecture and promoting student engagement will improve learning outcomes.

The success of active learning has led institutions of higher learning to implement active learning principles. For example, MIT has replaced their traditional passive learning introductory physics classes with what they refer to as TEAL, Technology-Enabled Active Learning. These changes were prompted by low lecture attendance and high failure rate in the previous traditional lecture style courses. A study on TEAL performance reveals improvements in conceptual understanding, class attendance, and passing rate[5]. The study shows the failure rate dropped from 13% to 5% and lecture attendance increased from 50% to 80%, compared to a control group.

Active Learning at The Data Incubator

We understand the benefits of active learning, and we built our curricula based on the evidence that active learning supports better outcomes for students than passive learning. Our data science training programs include various features that promote active learning.

Interactive Lectures: Lectures are presented via an interactive learning environment, where students can follow along and interact with the material on their own device. Students are encouraged to experiment with the variables in real time during lectures, to see how they affect results. Additionally, we demonstrate concepts using interactive figures and plots, allowing students to study the effect of changing parameters. One example lecture activity would be to visualize the effect on performance of a machine learning model by adjusting a hyperparameter. Students can engage with the visualization and confirm the effect we’re discussing in the lecture. Students are no longer merely remembering a fact, they’re analyzing and applying the concepts to actively engage in the learning process.

Flexible Format: Additionally, we avoid long lecture formats to encourage active learning- for this reason, a typical day of data science training will involve frequent breaks from lecture. During these breaks, students work on small exercises that reinforce the concepts that was just discussed. Breaks from lecture are important because people have limited attention spans. Additionally, they ensure students have a chance to employ “high order” thinking skills to essential concepts before moving on to more advanced material. If there’s not enough time to apply and analyze the material, students will not be able to effectively learn new material presented.

Real-world Miniprojects: We include a miniproject as part of each teaching module we create. Miniprojects help students to meet all of the learning objectives outlined by Bloom’s taxonomy by having students start applying the information they’ve just learned on a real-world problem, using real-world data. Students are challenged go beyond remembering and understanding material, to exercise “high order” thinking skills by evaluating lecture material against practical examples and creating solutions with hands-on practice. For example, students will evaluate different machine learning models to determine not only which approach would be best for a given application, but also what makes it better than other models in that particular instance.

Group Learning: Students are encouraged to work in groups; not only does this prevent students from falling behind (by externalizing accountability and encouraging collaboration), it enables them to exercise the “high order” thinking skills required to meet those learning objectives. Peer-to-peer engagement helps build confidence in students by creating more opportunities for reinforcing the course material. When reviewing or explaining a piece of information to a fellow student, that student is engaged in applying, analyzing, evaluating, and creating information based on the course material.

Active learning has been extensively explored and advocated by teaching experts because of the vast amount of benefits it realizes over passive learning. It helps to maintain student concentration and deepens learning towards the “high order” thinking skills. It also helps to engage students who might otherwise struggle. Active learning is the guiding principle behind the creation of all of the data science training curricula at The Data Incubator because of these proven benefits. Data science is not a spectator sport – it requires engagement with the material to master data science skills.


1.) Bonwell, Charles C., and James A. Eison. (1991). Active Learning: Creating Excitement in the Classroom. ASHE-ERIC Higher Education Report No. 1. Washington D.C.: The George Washington University, School of Education and Human Development.
2.) Bloom, Benjamin Samuel. (1956). Taxonomy of educational objectives: The classification of educational goals. Handbook I: Cognitive domain. New York: David McKay Company.
3.) Anderson, Lorin W.; Krathwohl, David R., eds. (2001). A taxonomy for learning, teaching, and assessing: A revision of Bloom’s taxonomy of educational objectives. Allyn and Bacon.
4.) Prince, Michael. (2004). Does Active Learning Work? A Review of the Research. Journal of Engineering Education. 93. 223-231.
5.) Dori, Yehudit & Belcher, John. (2005). How Does Technology-Enabled Active Learning Affect Undergraduate Students’ Understanding of Electromagnetism Concepts?. Journal of the Learning Sciences. 14(2), 243-279.

Learn more about our offerings:

Related Blog Posts

Moving From Mechanical Engineering to Data Science

Moving From Mechanical Engineering to Data Science

Mechanical engineering and data science may appear vastly different on the surface. Mechanical engineers create physical machines, while data scientists deal with abstract concepts like algorithms and machine learning. Nonetheless, transitioning from mechanical engineering to data science is a feasible path, as explained in this blog.

Read More »
Data Engineering Project

What Does a Data Engineering Project Look Like?

It’s time to talk about the different data engineering projects you might work on as you enter the exciting world of data. You can add these projects to your portfolio and show the best ones to future employers. Remember, the world’s most successful engineers all started where you are now.

Read More »
open ai

AI Prompt Examples for Data Scientists to Use in 2023

Artificial intelligence (AI) isn’t going to steal your data scientist job! Instead, AI tools like ChatGPT can automate some of the more mundane tasks in your future career, saving you time and energy. To make life easier, here are some data science prompts to get you started.

Read More »