What Active Learning Actually Looks Like in an Online Course

There is a version of an online course that looks like this: ten video lectures, each ten minutes long, chunked neatly into modules. A multiple-choice quiz at the end. A certificate when the final quiz is passed. The educator spent months building it. The production value is high. The learner watches, clicks through, and finishes.

And learns very little.

Not because the content is wrong. Not because the educator does not know their subject. But the course is built on a fundamental misunderstanding of how learning actually works. Watching is not learning. Clicking through is not learning. And a quiz that you can pass by scrolling back up is not an assessment. It is performance.

Watching is not learning. Clicking through is not learning. And a quiz you can pass by scrolling back up is not assessment.

Active learning is the alternative. But in online course design, the term gets used so loosely it has almost lost meaning. “Interactive” content, gamification badges, and discussion boards nobody reads. These are not active learning. They are the furniture of active learning placed in rooms where no real learning happens.

This article is about what active learning actually is, what the research says it does to learner outcomes, and what it looks like when it is designed well into an online course. Specifically, into an African online course, built for learners who may be studying asynchronously, solo, on a phone, with patchy connectivity. Because the evidence says it works there too.

The Definition Worth Starting With

In 1991, Charles Bonwell and James Eison published a landmark report for the Association for the Study of Higher Education that gave the field its most durable definition. Active learning, they wrote, involves instructional strategies that engage students in “doing things and thinking about what they are doing”, as opposed to passively receiving information. Approaches that promote active learning focus more on developing students’ skills than on transmitting content.

That definition is thirty-five years old and still routinely ignored in online course design.

The key phrase is doing and thinking. Both. An activity that involves doing without thinking: clicking through a drag-and-drop interaction, for instance, is not active learning. An activity that involves thinking without any output: watching a lecture while taking notes, sits at the lower end of the engagement spectrum. What moves learning forward is when the learner is required to produce something, apply something, or exchange something with another person.

What the Research Actually Says

The most important study in this space is Freeman et al. (2014), published in the Proceedings of the National Academy of Sciences. The research team conducted a meta-analysis of 225 studies comparing undergraduate STEM courses under traditional lecturing versus active learning methods.

The finding was not subtle. Students in traditional lecture classes were 1.5 times more likely to fail than students in active learning classes. Exam scores improved by about 6% on average under active learning conditions. And failure rates under traditional lecturing were 55% higher than under active learning. The researchers were direct: this evidence should prompt serious questions about the continued use of passive lecture as a default teaching method.

THE RESEARCH IN NUMBERS
55% — the increase in failure rates under traditional lecturing compared to active learning (Freeman et al., 2014)
225 studies — the scale of the meta-analysis; results held across STEM disciplines, class sizes, and institution types
1.5x — how much more likely students were to fail in a passive lecture environment
61% vs 40% — one-week retention for students who practised recall versus students who re-read (Roediger & Karpicke, 2006)

There is also a counterintuitive finding that every online educator should sit with. A Harvard study by Deslauriers and colleagues found that students in active learning classrooms actually learned more than their peers in passive lectures. But they felt like they were learning less. Passive learning is comfortable. It feels like mastery because the information is flowing in without friction. Active learning is effortful. Generating an answer, constructing an argument, explaining something to a peer: these feel harder, because they are harder. That productive difficulty is exactly what makes them work.

MIT Open Learning has described this principle as the “doer effect”: learners who actively engage with course content and receive feedback on their responses show higher learning gains than those who only passively read or watch. The act of doing is not a supplement to learning. It is the mechanism.

A Framework for Designing Engagement

If Bonwell and Eison give us the definition, Michelene Chi and Ruth Wylie give us the design tool. Their ICAP framework, published in 2014, categorises learner engagement into four modes based on what students actually do with instructional material. The modes sit on a spectrum, and learning increases as you move up it.

Engagement Mode	What the learner does	Online example
Passive	Watches, listens, reads: no output required	Watching a lecture video without stopping
Active	Writing a summary in their own words, drafting an application	Completing a reflection question below a video
Constructive	Creates something new from the material	Writing a summary in their own words; drafting an application
Interactive	Exchanges ideas with a peer and revises thinking	Async discussion: post, then respond to two classmates

Most online courses operate almost entirely in Passive mode, with occasional dips into Active. The goal of active learning design is not to eliminate every passive moment; some content genuinely needs to be absorbed before a learner can do anything with it. The goal is to move the centre of gravity from Passive toward Constructive and Interactive, and to do it deliberately.

The ICAP framework was recently applied in a peer-reviewed audit of asynchronous online courses and confirmed as a practical tool for evaluating the active learning potential of course design, not just a theoretical model. It works in the online context because it is based on observable learner behaviour, not on physical presence.

What It Looks Like in Practice

Here is a business writing course. It exists in two versions.

Version A (Passive)

• Module 1: Watch a 9-minute video on professional email structure
• Download a PDF template
• Complete 5 multiple-choice questions (answers visible in the PDF)
• Move to Module 2

Version B (Active)

• Module 1: Watch the same 9-minute video, but it stops at 4 minutes with a prompt: “Before we look at the structure breakdown, write down what you think the three main problems are with the email you just read. Do not scroll forward until you have written something.”
• After the video: draft a real email using the principles taught. The assignment specifies a professional scenario, a recipient, and a goal. There is a word limit. The learner cannot just write anything; they have to apply the structure.
• Discussion prompt: Post your draft in the course community. Reply to one peer’s draft with one thing that works well and one specific suggestion. Responses required before you can progress.
• End-of-module reflection: What did you change about your draft after reading the peer response?

Same content. Same video. Completely different learning experience. Version B moves through Active, Constructive, and Interactive modes in a single module.

What About the African Online Context?

Here is the objection I hear most often from African online educators: active learning is designed for well-resourced classrooms with live cohorts, fast internet, and learners who have time to collaborate. My learners are studying alone, at night, on their phones, with expensive mobile data.

The research does not support that objection. It dismantles it.

A 2024 study from Carnegie Mellon University tested the doer effect with 234 Ugandan students learning via phones and community radio broadcasts. These learners were studying in remote areas, on low-bandwidth technology, without the infrastructure of a conventional online course. The findings still showed that active engagement with practice questions predicted better learning outcomes than passive listening. The active learning advantage held.

The evidence says active learning works in low-resource, mobile-first African contexts. The constraint is not the bandwidth. It is the course design.

The constraint, in other words, is not the bandwidth. It is the course design. And that is something educators control.

There are practical adaptations for the African async context. Discussion prompts do not need a live response window; they can run over 72 hours. Reflection tasks can be text submissions that work on any device. Peer review can be structured so it requires only three to five sentences. Application tasks can be designed around the learner’s own professional context, which means no external resources are needed. Low bandwidth does not prevent thinking. It does not prevent writing. It does not prevent applying.

The Retrieval Trap: Why Your Quiz Probably Is Not Working

One more piece of research worth naming directly. Roediger and Karpicke’s 2006 studies on retrieval practice found that students who tested their recall of material retained 61% of it one week later. Students who spent the same time re-reading retained 40%.

Re-reading feels like studying. Re-reading raises familiarity with the material, and familiarity feels like mastery. But familiarity decays fast. Retrieval: the act of pulling information out of memory without being able to look it up — physically strengthens the memory trace. The act of retrieval is itself a learning event.

This has a direct implication for course design. A quiz where learners can scroll back up and find the answer is not retrieval practice. It is recognition practice, and the memory benefit is minimal. True retrieval practice means the learner must generate the answer from memory. Short open-ended questions. Application problems. “Without referring to the module, explain in two sentences why X matters.” These are retrieval tasks. Multiple choice with the module open beside you is not.

RETRIEVAL PRACTICE: WHAT IT IS AND WHAT IT IS NOT
IS retrieval: ‘Without referring back, write down the three steps in this process.’
IS retrieval: ‘A new client sends you this email. Identify two structural problems and explain why they matter.’
IS NOT retrieval: Multiple choice with the module open beside the learner
IS NOT retrieval: Drag-and-drop matching where the options are visible on screen
IS NOT retrieval: ‘According to the video, what is the first step?’ (Answer is directly stated)

Four Things You Can Do in Your Next Course

Active learning design does not require rebuilding everything from scratch. It requires adding the right kind of friction in the right places. Here are four practical moves.

TRY THIS NOW
1.  Add a mid-content prompt. Stop your video or reading at a key concept and ask a question that requires the learner to produce an answer before continuing. Even a text box that says ‘Write your answer before you proceed’ shifts the mode from Passive to Active.
2.  Replace end-of-module recall quizzes with application tasks. Instead of ‘What are the three steps?’ ask ‘Here is a scenario. Walk me through how you would apply the three steps here.’ Closed-book. No scrolling back.
3.  Design one peer exchange per module. It does not have to be complex. Post a 100-word reflection. Respond to one peer with one observation and one question. That is Interactive mode in an async environment.
4.  End every module with a construction task. Ask the learner to make something: a short plan, a revised document, a worked example from their own context. Constructive output is the highest return on learning time.

Courses Built for Learning, Not Just for Watching

The difference between a content library and a course is not the production value. It is whether the design compels the learner to engage, apply, and construct. A well-structured course creates the conditions where active learning is not optional. It is baked into the sequence.

The Fay Institute Learning Framework (FILF), which underpins FayEDU’s Course Studio, was built on exactly these principles. Every course built through FayEDU is structured around the learning engineering insight that passive delivery is the cheapest way to build a course and the most expensive way to produce a learner.

If you are an African educator ready to build online courses that are designed for real learning outcomes, explore what FayEDU’s Course Studio makes possible at fayedu.com.

References

Bonwell, C. C., & Eison, J. A. (1991). Active learning: Creating excitement in the classroom. ASHE-ERIC Higher Education Report No. 1. George Washington University.

Butler, D., Borchers, C., Asher, M. W., Lee, Y., Karnataki, S., Dangi, S., Athreya, S., Stamper, J., Ogan, A., & Carvalho, P. F. (2024). Does the doer effect exist beyond WEIRD populations? Toward analytics in radio and phone-based learning. arXiv. https://arxiv.org/abs/2412.20923

Chi, M. T. H., & Wylie, R. (2014). The ICAP framework: Linking cognitive engagement to active learning outcomes. Educational Psychologist, 49(4), 219–243.

Deslauriers, L., McCarty, L. S., Miller, K., Callaghan, K., & Kestin, G. (2019). Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom. Proceedings of the National Academy of Sciences, 116(39), 19251–19257.

Freeman, S., Eddy, S. L., McDonough, M., Smith, M. K., Okoroafor, N., Jordt, H., & Wenderoth, M. P. (2014). Active learning increases student performance in science, engineering, and mathematics. Proceedings of the National Academy of Sciences, 111(23), 8410–8415.

MIT Open Learning. (2024). 2023–2024 Open Learning Impact Report. Massachusetts Institute of Technology.

Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255.

Vale, R., & Falloon, G. (2024). ICAP framework as an audit tool for asynchronous online learning activities. Online Learning Journal, 29(3).

ABOUT THE AUTHOR

Danielle Thomas is a pedagogy specialist and educator writing for Grounding EdTech Magazine. She focuses on translating learning science into classroom practice for online educators across Africa. Her work covers active learning design, student engagement, and the practical craft of building courses that actually develop learners.

Published in Grounding EdTech Magazine | Pedagogy & Learning Methodologies